Viro v The Queen

Karen drove down Clarke Street and I turned around and hit John on the head with the jack handle a couple of times to stun him so that I could rip him off. Karen stopped the car quickly, the dark bloke was screaming out, "Stop it Fred, leave him alone Fred." He was making out that he wasn't in the rip off and he got out of the car and I think he ran, because I didn't see him again. Anyhow after I hit John the second time with the jack handle, he had a knife out, I think he got it out of his pants, but Seppe later said that he thought that he had it in his hand all the time. It was a sort of a flick knife, one that opens out and he had it in his left hand. I grabbed hold of his arm and dropped the jack handle and tried to stop him from sticking us and Seppe got hold of him around the neck and I said to him "Stop it man." He cut me on the fingers, the three fingers on my right hand. [indicates cut on knuckle index finger, knuckle of middle finger and tip of little finger.] He cut Seppe on the arm. He was a madman and wouldn't stop struggling. I got the knife out of the glovebox and held it at him and said, "Drop the knife, cool it man." All the time Seppe was still struggling with him and John was kicking at me and he kicked me in the left shin. [indicates two scabs left shin.] Seppe said, "Quick man, you'll have to give it to him. Give it to him. Give it to him. Otherwise we are dead." I stuck him once in the stomach and said, "Stop it and drop the knife". But he was crazy and kept struggling and Seppe kept saying to me, "Give it to him". I stuck him three or four times then, I don't know just where, but it was in his body and once in his left arm to make him drop the knife, I purposely tried to get him in the arm to make him drop the knife, but he was just too strong and by then we were all covered in blood, so I opened my door and said, "Let's get him out of here, don't worry about the money." The whole thing sickened me. Some of the inhabitants of Clarke Street, Bass Hill, saw Rellis precipitated from the car shortly before 11 p.m. He died soon afterwards. His body bore about seventeen cuts or wounds - on the face, chest, side, shoulder and back. One wound, on the right side of the back, was 9 cm deep and had penetrated the heart - that was the fatal wound. The steak knife, which is an exhibit in the case, had a serrated blade, about 10½ cm in length. The flick knife that the applicant said had been used by Rellis had a blade 9½ cm in length but it was a heavier blade than that of the steak knife.

At the conclusion of the evidence the applicant made an unsworn statement from the dock. It seems necessary to set it out in full:

Your Honour, members of the jury: I am where I am today because I am addicted to heroin. I have tried to get off the habit and have had treatment, but it is very hard.

At the time we thought up this crazy scheme to get the money off Rellis we were badly affected by heroin. I was well affected by heroin at the time of the stabbing too. You just can't think properly with drugs in you. I certainly wasn't thinking straight. I never formed any intent to harm Rellis.

I have set out what the position was as best I can recall in the record of interview. I never intended to stab Rellis. I was just trying to get the money so I could buy more heroin for myself.

As soon as the struggle in the car started Rellis had that knife of his out. I didn't know he had it until I saw it in his hand. I have got no doubt Rellis was trying to get the heroin off us and not pay for it. He was a big trader, drug trader. I dropped the jack handle and tried to stop him from stabbing us. It is hard to explain, but Rellis seemed like a madman. I don't know if he was on drugs or not. I got cut on the hand. I said I got stabbed in the arm, as you heard in the evidence. As soon as I saw the knife in Rellis' hand all thoughts of taking the money left my mind. The only way out I could think of at the time was to use the small knife I kept in the glovebox of the car. I kept it in the glovebox of the car to cut up lemons with and fruit that we used to take when we were having drugs.

I never intended the knife to come into it at all, and I only took it out because I was in panic for my life. You know I didn't take the money. If I had meant to use the knife I would have taken the money before I left Rellis. I was just glad when we got him out of the car and I only got out of the car to get him out. It was very confused and Rellis was very violent.

Looking back now, I know I would never have tried to get the money off him if I had not been affected by heroin. It obviously could not have worked. Rellis was a big man and he knew how to look after himself.

I ask you not to convict me, because in my own heart I know that the only reason I stabbed Rellis was because I was afraid for my life. I was very afraid of him, and I am sorry that all this happened.

Before dealing with the two main grounds on which the applicant seeks leave to appeal it is convenient to mention a comparatively minor submission made on behalf of the applicant. This was that the learned trial judge failed adequately to direct the jury as to the effect of the written record of interview tendered by the Crown against Greco. When asked about the knife used by the applicant, Greco had said "That's what we were going to hold him up with." That was less favourable to the applicant than the applicant's own version. The statement made to the police by Greco, in the absence of the applicant, was of course inadmissible against the applicant. It was necessary that the learned trial judge should make it quite clear to the jury that they could not rely on anything said by Greco in his record of interview when they were considering the case against the applicant. In fact the learned trial judge did give the necessary direction. During the course of evidence, as each record of interview was tendered, he told the jury that the statements were admissible only against the accused who had made them. Later, in the course of his summing up, he repeated and elaborated that direction. This ground is completely devoid of merit.

The two principal grounds argued on behalf of the applicant were that the learned trial judge failed to direct the jury (1) that in deciding whether the applicant had formed the intention necessary to constitute murder they should take into account the fact that he had used drugs, and (2) that they were entitled to return a verdict of manslaughter if the applicant acted in self-defence but used excessive force in doing so. These submissions raise matters of some importance.

In New South Wales the elements of a charge of murder are set out in s. 18 (1) (a) of the Crimes Act, 1900 which provides as follows:

Murder shall be taken to have been committed where the act of the accused, or thing by him omitted to be done, causing the death charged, was done or omitted with reckless indifference to human life, or with intent to kill or inflict grievous bodily harm upon some person, or done in an attempt to commit, or during or immediatley after the commission, by the accused, or some accomplice with him, of a crime punishable by death or penal servitude for life. It was not suggested in the present case that the acts of the applicant were done with reckless indifference to human life. The Crown case was that the jury should be satisfied that the applicant had an intention to kill Rellis or an intention to inflict grievous bodily harm upon him. Alternatively the Crown relied on what it may be still convenient (if inaccurate) to call felony-murder. In other words, the Crown submitted that the acts done by the applicant which caused the death of Rellis were done in an attempt to commit, or during or immediately after the commission, by the applicant, of a crime punishable with penal servitude for life. The crime in question, according to the Crown case, was that provided by s. 98 of the Crimes Act which is in the following terms: Whosoever, being armed with an offensive weapon, or instrument, or being in company with another person so armed, robs, or assaults with intent to rob, any person, and immediately before, or at the time of, or immediatley after, such robbery, or assault, wounds, or inflicts grievous bodily harm upon, such person, shall be liable to penal servitude for life.

Where the act of the accused which caused the death was done in an attempt to commit, or during or after the commission, by the accused, of a crime punishable with penal servitude for life, it is unnecessary for the Crown to establish, as an element of the murder charged, that the accused had any intention to kill or to inflict grievous bodily harm. So much is clear from the words of s. 18 itself and is established by the authority of Ryan v. The Queen [25] . In that case it was held that to constitute murder it is necessary that the act of the accused should have been a voluntary act - an act resulting from the exercise of his will - but that no intention to kill or wound is necessary. It has never been suggested in the present case that the actions of the applicant were involuntary. In the light of the decision of the House of Lords in Director of Public Prosecutions v. Majewski [26] it could not successfully have been contended that the applicant would have been exonerated from criminal responsibility if the drugs which he had voluntarily taken had deprived him of the ability to exercise self-control or to realize the possible consequences of what he was doing. However, it is not right to say (and it was not said in Ryan v. The Queen) that no intention is necessary on the part of an accused before he can be convicted of felony-murder under s. 18. The lethal act must have been done in an attempt to commit, or during or immediately after the commission, by the accused or by some accomplice with him, of a crime punishable with penal servitude for life. Where the attempt has been made, or the crime has been committed, by the accused, the accused must necessarily have had the intent (if any) involved before he could be held to have made the attempt or to have committed the crime. In Ryan v. The Queen the crime committed was robbery with arms and wounding. In the present case there was no robbery - no property was taken - and the Crown case was that the relevant crime was assault with intent to rob with arms and wounding. The intention which the Crown had to prove on this aspect of the case was an intent to rob. The evidence that the applicant had such an intent was overwhelming - he did not ever suggest that the use of the drugs meant that he had not formed an intent to rob. On the contrary, he in effect reaffirmed, in his unsworn statement, what he had already said in his record of interview, that he did intend to rob Rellis. It would have been a question for the jury whether the stabbing had occurred during or immediately after the commission of the crime, but on this aspect of the case the failure to give a direction that the jury, in considering whether the applicant had formed the necessary intention, should consider the fact that the applicant had used drugs could not have misled the jury.

However, the case of felony-murder, although never abandoned, and strongly supported by the evidence, does not appear to have been put in the forefront of the Crown case; it certainly occupied a subsidiary place in the summing up. The jury may have convicted the applicant because they were satisfied that he intended to kill or inflict grievous bodily harm upon Rellis. It is therefore necessary to consider whether the directions on that question were sufficient.

The learned trial judge did direct the jury on a number of occasions in the course of the summing up that it was necessary that they should find an intent to kill, or an intention to do grievous bodily harm, before they could convict. It was not submitted that this direction was inadequate in itself. However, the learned trial judge did not expressly tell the jury that the fact that the applicant had used heroin might be relevant to the question whether he had formed the necessary intention. He did repeat to the jury in full the unsworn statement which the applicant had made and which the jury had recently heard. After reading the words "I certainly wasn't thinking straight. I never formed any intent to harm Rellis", he interpolated the words "He is saying in his statement he never had the intent to kill or harm him". A little later in the course of his summing up the learned trial judge said the following:

Gentlemen, in relation to each of the accused who said they were heroin addicts or on drugs, the fact that a person is a drug addict is an irrelevant situation so far as the commission of this crime is concerned. It is no excuse or no defence and it only comes into the matter, as I understand, because it is suggested on their behalf that if you are on drugs you are not able to think clearly, not able to remember things and things of that nature. That is the only way it comes into this case as I understand it.

Their records of interview are put before you as documents of truth which you can rely on and the learned Crown Prosecutor suggests that an examination of those will show in each case they were clear in their evidence, able to think and they gave detailed explanations and there was no impairment or infringement of their faculty to remember or express themselves and to understand.

At the conclusion of the summing up the learned trial judge was asked to direct the jury that in deciding whether or not the applicant in fact formed an intent they should take into account the effect of the drugs, but he refused to do so.

The rules of the common law as to the criminal responsibility of a person who has voluntarily taken drugs are the same as those which relate to drunkenness voluntarily caused. There is no reason why intoxication caused by the use of drugs should be regarded any differently from intoxication caused by alcohol, and at common law (as under most Criminal Codes the effect of drink and the effect of drugs on criminal responsibility are governed by the same principles: Reg. v. Lipman [27] ; Director of Public Prosecutions v. Majewski [28] . It is well established that intoxication voluntarily caused is no excuse for the commission of a crime. However, when an intention to cause a particular result ("a special intent") is an element of the crime - as in a charge of murder based upon an intention to kill or to inflict grievous bodily harm - the fact that the accused was intoxicated may have an important bearing on the question whether he had the necessary intent. Unfortunately difficulty in this area of the law has been caused by some dicta of Lord Birkenhead L.C. in Director of Public Prosectuions v. Beard [29] . For the purposes of the present case it is not necessary to consider all the matters which were there discussed, and it is sufficient to refer to the two following propositions stated by the Lord Chancellor [30] :

2. That evidence of drunkenness which renders the accused incapable of forming the specific intent essential to constitute the crime should be taken into consideration with the other facts proved in order to determine whether or not he had this intent.

3. That evidence of drunkenness falling short of a proved incapacity in the accused to form the intent necessary to constitute the crime, and merely establishing that his mind was affected by drink so that he more readily gave way to some violent passion, does not rebut the presumption that a man intends the natural consequences of his acts.

For reasons which I shall give neither of these propositions can be accepted as correctly stating the law for Australia. 1. [1970] 1 Q.B. 152, at p. 156. 2. [1977] A.C. 443. 3. [1920] A.C. 479, at pp. 499-505. 4. [1920] A.C., at pp. 501-502.

The former of the two propositions just quoted was criticized by the Judicial Committee in Broadhurst v. The Queen [31] . Lord Devlin there said [32] :

If an accused is rendered incapable of forming an intent, whatever the other facts in the case may be, he cannot have formed it; and it would not therefore be sensible to take the incapacity into consideration together with the other facts in order to determine whether he had the necessary intent. This criticism is unanswerable. If Lord Birkenhead L.C. had stated that evidence of drunkenness should be taken into consideration with the other facts proved in order to determine whether the accused had the necessary special intent, his statement would in my opinion have been correct. 1. [1964] A.C. 441. 2. [1964] A.C., at p. 461.

The latter of the two propositions of Lord Birkenhead L.C. suggests not only that evidence of drunkenness will not be material unless it shows that the accused was incapable of forming the necessary intent, but also that an onus lies on an accused person to prove that he did not intend the natural consequences of his acts. Director of Public Prosecutions v. Beard [33] was decided before Woolmington v. The Director of Public Prosecutions [34] which effected an important change in the law as to onus of proof. The doctrine that the question whether an accused person acted with a particular intention is to be answered by reliance upon a presumption that he intended the natural consequences of his acts has been rejected in Australia: see particularly Smyth v. The Queen [35] and Parker v. The Queen [36] . It may be that this doctrine influenced Lord Birkenhead L.C. to hold that evidence of drunkenness which did not render the accused incapable of forming the necessary intent was no answer to the charge. It is however right to say that some of the earlier authorities use language that appears to support his dictum, and that later dicta, of high authority, have echoed the suggestion that drunkenness is only an answer to a charge of an offence in which a special intent must be proved if it rendered the accused incapable of forming that intent: Attorney-General (Northern Ireland) v. Gallagher [37] ; Bratty v. Attorney-General (Northern Ireland) [38] . In Broadhurst v. The Queen [39] ; the Judicial Committee did not make it clear what opinion it held on this question, their Lordships did however criticize what Lord Birkenhead L.C. said as to the burden of proof [40] . In New South Wales the courts have not accepted Lord Birkenhead's statement of principle as correct. In Reg. v. Gordon [41] the Court of Criminal Appeal held that on a charge of murder:

(a) "the burden of disproving intent does not lie upon the accused but always the onus is on the Crown to prove that the accused had such intent"; and

(b) "the test is not incapacity to form but the absence of the intent in the accused himself".

The law thus stated is the same as that expounded in decisions on the Criminal Code of Queensland, such as Reg. v. Nicholson [42] and Reg. v. Crump [43] . In England, since s. 8 of the Criminal Justice Act, 1967 U.K. has provided in effect that the question whether an accused person intended a particular result is not governed by a presumption, but is to be decided by reference to all the evidence, the Court of Criminal Appeal has cast doubt on the authority of Director of Public Prosecutions v. Beard [44] , and has indicated that in a case of drunkenness the jury should be instructed to consider not whether the accused had the capacity to form, but whether he in fact had formed, the requisite intent: Reg. v. Sheehan and Moore [45] ; Reg. v. Pordage [46] . In Director of Public Prosecutions v. Majewski [1] , where Director of Public Prosecutions v. Beard was much discussed in the House of Lords, the crime was not one involving a special intent, and the present question did not fall for direct decision. 1. [1920] A.C. 479. 2. [1935] A.C. 462. 3. (1957) 98 C.L.R. 163. 4. (1963) 111 C.L.R. 610, at p. 632. 5. [1963] A.C. 349, at p. 381. 6. [1963] A.C. 386, at p. 410. 7. [1964] A.C., at pp. 461-462. 8. [1964] A.C., at p. 463. 9. [1963] S.R. (N.S.W.) 631, at p. 635. 10. [1956] St. R. Qd. 520. 11. [1966] Qd. R. 340. 12. [1920] A.C. 479. 13. [1975] 1 W.L.R. 739; [1975] 2 All E.R. 960; 60 Cr. App. R. 308. 14. [1975] Crim. L.R. 575. 15. [1977] A.C. 443.

It would be contrary to fundamental principle to hold that evidence of intoxication not amounting to incapacity is irrelevant to criminal responsibility where the commission of the crime requires a special intent. In the case of such a crime the issue is not whether the accused was incapable of forming the requisite intent, but whether he had in fact formed it. The Crown must prove beyond reasonable doubt that the accused actually formed the special intent necessary to constitute the crime. If no more were proved than that the accused was capable of forming such intent, the case for the prosecution would not have been established. Proof of capacity will not be elevated to proof of intent by the operation of a presumption that a man intends the natural consequences of his acts. Of course if the jury were not satisfied that the accused was capable of forming the requisite intent that would be the end of the matter, but if they were satisfied that he was capable of forming that intent they would have to go on to consider whether in fact he did so. The state of intoxication of an accused person is one of the matters to be considered by the jury in deciding whether they are satisfied that he had the requisite intent. In Director of Public Prosecutions v. Majewski, Lord Salmon said [2] :

This does not mean that drunkenness, of itself, is ever a defence. It is merely some evidence which may throw a doubt upon whether the accused had formed the special intent which was an essential element of the crime with which he was charged. Often this evidence is of no avail because obviously a drunken man may well be capable of forming and does form the relevant criminal intent: his drunkenness merely diminishes his powers of resisting the temptation to carry out this intent. That in my respectful opinion is good sense as well as good law, and it applies equally to intoxication caused by drugs. 1. [1977] A.C., at p. 481.

In a case where there is evidence fit to be considered by a jury that the accused was intoxicated as a result of the consumption of drink or drugs, it is not enough to tell the jury that the Crown must prove beyond reasonable doubt that the accused had in fact formed the requisite special intent. They should also be told that the fact that the accused was intoxicated, whether by drink or drugs or by a combination of both, may be regarded for the purpose of ascertaining whether the special intent in fact existed. It should be explained that evidence that the accused was intoxicated will not in itself entitle him to an acquittal, because a person when intoxicated may form the necessary intent, and one who has formed the intent does not escape responsibility because his intoxication has diminished his power to resist the temptation to carry it out. However, the jury should be told that if, because of the evidence as to the effect of the intoxication or otherwise, they are not satisfied that the accused did in fact have the necessary intent, they must acquit of the crime which involves that intent.

The jury in the present case were not told that they were entitled to have regard to the fact that the applicant was high on heroin in deciding whether he had formed the necessary intent. On the contrary, the passage from the summing up that I have quoted may well have led them to think that the evidence that drugs had been taken was irrelevant to that issue. It there was evidence fit to be considered on this question, the jury were not properly directed. It is true that there was no evidence as to the effects which heroin is likely to have on the mental processes of the person who uses it. However, the jury were entitled and bound to consider the meaning of the statements that the applicant was "pretty high", and "well affected by heroin". It is also true that on one view the applicant's record of interview and unsworn statement showed that he knew what he was doing at the time and that he afterwards had a clear recollection of what he had done. But the effect of that material was entirely for the jury to evaluate. In the circumstances the jury should have been told that in deciding whether they were satisfied that the applicant intended to kill or harm Rellis they might have regard to the fact that the applicant had taken heroin. The summing up was defective in this respect.

I turn now to consider the submission that the jury should have been directed that where upon a charge of murder a plea of self-defence fails only because the force used by the accused was excessive, the homicide is reduced from murder to manslaughter. The decision of this Court in Reg. v. Howe [3] would require a direction of that kind to be given. However in Palmer v. The Queen [4] the Judicial Committee of the Privy Council (on appeal from the courts of Jamaica) refused to follow Reg. v. Howe. Their Lordships there stated the law as follows [5] :

A jury will be told that the defence of self-defence, where the evidence makes its raising possible, will only fail if the prosecution show beyond doubt that what the accused did was not by way of self-defence. But their Lordships consider that if the prosecution have shown that what was done was not done in self-defence then that issue is eliminated from the case. If the jury consider that an accused acted in self-defence or if the jury are in doubt as to this then they will acquit. The defence of self-defence either succeeds so as to result in an acquittal or it is disproved in which case as a defence it is rejected. The law as stated in Palmer v. The Queen has since been applied by the Privy Council on appeal from Hong Kong (Edwards v. The Queen [6] , and by the Court of Appeal in England (Reg. v. McInnes [7] ). On behalf of the applicant it was submitted that the law is correctly stated in Reg. v. Howe and that the decision in that case rather than the decisions of the Privy Council should be followed in New South Wales. The learned Solicitor-General for New South Wales did not dispute the correctness of this submission. He agreed that Reg. v. Howe correctly states the law. His submission was that in the present case the question of self-defence should not have been left to the jury at all, because a reasonable jury could only have taken the view that the applicant was the aggressor throughout the whole incident and never withdrew from the conflict which he had commenced. He further submitted that in any case the applicant was not entitled to rely on a plea of self-defence, when the force which he employed was used in the course of a struggle which he had commenced while making a felonious assault on Rellis. I shall later deal with the submissions made on behalf of the Crown in opposition to the present application in so far as they rest on the facts of the case. But I may say immediately that the principle of law for which the Solicitor-General contended is too absolute to be accepted. In my opinion it cannot be said as a matter of law that the fact that the applicant was the aggressor and made a violent criminal assault upon Rellis meant that the subsequent killing of Rellis could not under any circumstances be excused on the ground of self-defence. 1. (1958) 100 C.L.R. 448. 2. [1971] A.C. 814. 3. [1971] A.C., at p. 832. 4. [1973] A.C. 648, at p. 658. 5. [1971] 1 W.L.R. 1600; [1971] 3 All E.R. 295; 55 Cr. App. R. 551.

The rules of the common law on the subject of self-defence were formerly strict and technical. An account of some of the earlier writings will be found in an article by Mr. Harold Snelling Q.C. in the Australian Law Journal, vol. 34 (1960), p. 130. A distinction was drawn between justifiable homicide on the one hand and excusable homicide by self-defence on the other. It was justifiable for one man to kill another in the course of resisting a felonious assault. Blackstone said (4 Comm. 180): "If any person attempts a robbery or murder of another and shall be killed in such attempt, the slayer shall be acquitted and discharged." On the other hand, homicide might be excusable in the case of a killing in the course of a sudden brawl or quarrel, but in that case it was required that the accused "should have retreated as far as he conveniently or safely can, to avoid the violence of the assault, before he turns upon his assailant": ibid., 184-185. Modern writers still attach significance to the distinction between self-defence to a felonious attack and other cases. In Russell on Crime, 12th ed. (1964), vol. 1, p. 436, it is said:

There is, however, still a substantial difference in cases where death has been inflicted in self-defence: for if the attack resisted is a felonious one then the victim of it may stand his ground and kill with impunity, so long as the means of resistance which he employs are held to be reasonable in the circumstances. If, however, the attack is not felonious then the victim must if possible retreat, and can only be excused for a death caused by resistance if it was no longer possible for him to withdraw in safety. In Kenny's Outlines of Criminal Law, 19th ed. (1966), p. 144, a similar view is expressed. It would appear that under the old rules a plea of self-defence would not have been availiable to the applicant. Under those rules Rellis, the victim of a felonious assault, would have been entitled to stand his ground and defend himself; on the other hand, the applicant would have been bound to retreat.

However the law of self-defence is now more simple and rational. A person who, acting in self-defence, uses force which is no greater than is reasonably necessary in all the circumstances, and is not out of all proportion to the injury which it is intended to prevent, is not guilty of any crime if his or her assailant is killed. In Palmer v. The Queen their Lordships said [8] :

In their Lordships' view the defence of self-defence is one which can be and will be readily understood by any jury. It is a straightforward conception. It involves no abstruse legal thought. It requires no set words by way of explanation. No formula need be employed in reference to it. Only common sense is needed for its understanding. It is both good law and good sense that a man who is attacked may defend himself. It is both good law and good sense that he may do, but may only do, what is reasonably necessary. But everything will depend upon the particular facts and circumstances. Of these a jury can decide. It may in some cases be only sensible and clearly possible to take some simple avoiding action. Some attacks may be serious and dangerous. Others may not be. If there is some relatively minor attack it would not be common sense to permit some action of retaliation which was wholly out of proportion to the necessities of the situation. If an attack is serious so that it puts someone in immediate peril then immediate defensive action may be necessary. If the moment is one of crisis for someone in imminent danger he may have to avert the danger by some instant reaction. If the attack is all over and no sort of peril remains then the employment of force may be by way of revenge or punishment or by way of paying off an old score or may be pure aggression. There may no longer be any link with a necessity of defence. Of all these matters the good sense of a jury will be the arbiter. There is nothing in that passage inconsistent with the decision in Reg. v. Howe [9] . In accordance with this view of the law it is now accepted, both in England and Australia, that it is no longer essential, even in circumstances where the homicide would have been regarded as excusable rather than justifiable, that, before a plea of self-defence can succeed, it should appear that the accused had retreated as far as possible having regard to the nature of the attack. "The failure to retreat is a circumstance to be considered with all the others in order to determine whether the defendant went farther than he was justified in doing": per Holmes J. in Brown v. United States [10] , cited in Reg. v. Howe [11] . In other words, a failure to retreat is "simply a factor to be taken into account in deciding whether it was necessary to use force, and whether the force used was reasonable": Reg. v. McInnes [12] , citing Smith and Hogan, Criminal Law, 2nd ed. (1969), p. 231. 1. [1971] A.C., at pp. 831-832. 2. (1958) 100 C.L.R. 448. 3. (1920) 256 U.S. 335, at p. 343 [65 Law. Ed. 961, at p. 963]. 4. (1958) 100 C.L.R., at p. 463. 5. [1971] 1 W.L.R. 1600, at p. 1607; [1971] 3 All E.R., at p. 300; 55 Cr. App. R. 551, at p. 560.

In the United States of America it appears to be established that an aggressor who provokes an attack upon himself, or produces the occasion which makes it necessary for him to take life, cannot assert that he acted in self-defence unless when attacked he has retreated as far as possible and announced his desire for peace. The aggressor, to be entitled to defend himself, "must have withdrawn actually and really and in good faith. He must retreat as far as the fierceness of the assault by the other will permit without danger of death or great personal injury, before he will be justified in killing. An insufficient retreat is of no effect. Nor is this all; the aggressor must inform his antagonist of his purpose to withdraw from the conflict. If the circumstances are such that he cannot do this, it is attributable to his own fault and he must abide by the consequences.": 40 Am. Jur. 2d, Homicide, ss. 140, 145, 150; see also 55 A.L.R. 3d, pp. 1003-1004, and Perkins, Criminal Law (1957), pp. 896-897. In my opinion, in Australia the fact that the person raising self-defence was the aggressor is an important consideration of fact, but not a legal barrier to the success of the plea. The matter may be regarded in a similar light to a failure to retreat. It is obvious enough that a person cannot rely upon the plea of self-defence unless the violence against which he sought to defend himself was unlawful. If, therefore, one man makes a violent attack upon another with intent to rob him, and the man attacked defends himself, using no more force than is reasonably necessary, the original assailant cannot be said to be acting in self-defence in trying to overcome the other's resistance, since that resistance was lawful. However, if the original assailant has desisted from his attack, and his intended victim no longer needs to defend himself, and can not reasonably believe that he is still in danger, but nevertheless takes the offensive and out of anger or revenge himself becomes the attacker, the original assailant is not obliged to let himself be killed or injured without any attempt at resistance. Nevertheless, in such a case it is difficult to see how, as a matter of fact, the conduct of the aggressor, which commences as a criminal assault with an intent to commit a serious crime, can become transmuted in split seconds into lawful self-defence, unless the aggressor has clearly broken off his attack. In such circumstances the fact that he did not retreat when he had the opportunity to do so assumes a special significance.

In the present case, the applicant, while in company with a number of accomplices, attacked Rellis with a jack handle in an endeavour to stun and rob him. Rellis produced his knife only after the attack had commenced. Thereafter the applicant picked up a knife. The applicant did not suggest that he made any attempt to disengage himself from the ensuing struggle until after Rellis had been mortally wounded. He was in the front of the car and Rellis was in the back. The car was stopped when the attack on Rellis commenced. It might have been supposed that the applicant, if attacked, could have got out of the vehicle, as Fernando in fact did. In the end, when the applicant did open the door of the car, he used words which did not suggest that he was endeavouring to escape: "Let's get him out of here, don't worry about the money." In all these circumstances the assertion that the applicant was acting in self-defence when he stabbed Rellis seems far fetched and the jury were entitled to view it with scepticism. It would not have been unreasonable for the jury to conclude that the applicant killed Rellis in the course of a struggle which began with his violent assault upon Rellis and continued without interruption as Rellis endeavoured to defend himself - in other words that the applicant was never acting in self-defence. However it must be remembered that it was entirely for the jury to interpret the evidence contained in the applicant's statement: Reg. v. Howe [13] . And it is well settled that the onus is never upon an accused person to establish that he acted in self-defence: Chan Kau v. The Queen [14] .

"Once a ground is disclosed by the evidence upon which a plea of self-defence may arise, it is essential to a conviction of murder that the jury shall be satisfied beyond reasonable doubt that one or other or all of the ultimate facts which establish that plea were not present.": Reg. v. Howe. The jury in the present case might have accepted that the applicant feared for his life when Rellis produced the knife. They had to consider the effect of the statements that the applicant dropped the jack handle and that Rellis seemed like a madman and would not stop struggling. They could consider the suggestion that Rellis was a drug trader and was trying to get the heroin from the applicant without paying for it. The judge was on any view justified in leaving the issue of self-defence to the jury, because a judge, if in any doubt as to whether there is sufficient material to raise such an issue, should leave the issue to the jury. Although an appellate court is in a different position, it would not seem to me to be satisfactory to determine this case by holding that no reasonable jury could have failed to be satisfied that the applicant was not acting in self-defence, even if that were the view that ultimately commended itself to us. Without deciding whether there was material sufficient to raise the issue of self-defence, I proceed to consider the objection made to the summing up in relation to that question. 1. (1958) 100 C.L.R., at p. 459. 2. [1955] A.C. 206.

If the decision of the Privy Council in Palmer v. The Queen [15] is to be followed, the summing up on the issue of self-defence was correct. So the question that immediately arises is whether this Court is bound to defer to that decision of the Privy Council notwithstanding that it is in conflict with an earlier decision of this Court. When the Privy Council was the ultimate court of appeal from the courts of Australia, its decisions were of course binding on all Australian courts. There was an exception to this rule: a decision of the Privy Council upon a question as to the limits inter se of the constitutional powers of the Commonwealth and those of any State or States, or as to the limits inter se of the constitutional powers of any two or more States, pronounced in a case in which this Court had not given its certificate under s. 74 of the Constitution, was not binding: Baxter v. Commissioner of Taxation (N.S.W.) [16] . Another exception has been suggested - that a decision of the Privy Council given on appeal from the courts of some other part of the British Commonwealth was not binding on Australian courts. In conformity with this view, a court in Canada, faced with the much criticized decision of the Board in Victorian Railways Commissioners v. Coultas [17] held that that case was authority only as to the law of Australia and was not binding on the courts of Canada: Negro v. Pietro's Bread Co. Ltd. [18] . At that time the notion that a decision of the Privy Council on appeal from Australia was not binding in Canada was heretical. Since then it has become recognized by the Privy Council itself that the common law may develop differently in different Commonwealth countries: Australian Consolidated Press Ltd. v. Uren [19] ; Abbott v. The Queen [20] . It has become possible to say that the common law for Australia is not necessarily the same as the common law of England or of some other part of the British Commonwealth: see Mutual Life and Citizens' Assurance Co. Ltd. v. Evatt [21] and Cooper v. Southern Portland Cement Ltd. [22] . Nevertheless the cases in which the common law of Australia differs from that of England are likely to be exceptional. Diversity is certainly not to be encouraged for its own sake. And in any case, once the Privy Council decided a question of general law, without indicating that it was laying down a principle peculiar to the dominion, colony or other state from which the appeal was brought, its decisions were binding upon all courts from which an appeal lay to the Privy Council. That seems to have been held by the Privy Council itself in Fatuma Binti Mohamed Bin Salim Bakhshuwen v. Mohamed Bin Salim Bakhshuwen [23] and by this court in Morris v. The English, Scottish and Australian Bank Ltd. [24] . Those authorities were discussed in Mayer v. Coe [25] and Ratcliffe v. Watters [26] and it was held that the Supreme Court of New South Wales should follow a decision of the Privy Council given on appeal from New Zealand, notwithstanding that it was in conflict with an earlier decision of this Court. At the time when Palmer v. The Queen was decided, it might have been expected that this Court would have treated itself as bound to follow that decision, notwithstanding that it was given on appeal from Jamaica.

1. [1971] A.C. 814. 2. (1907) 4 C.L.R. 1087. 3. (1888) 13 App. Cas. 222. 4. [1933] 1 D.L.R. 490, at pp. 494-496. 5. (1967) 117 C.L.R. 221, at pp. 238-241; [1969] 1 A.C. 590, at pp. 640-641. 6. [1977] A.C. 755, at p. 768. 7. (1968) 122 C.L.R. 556, at p. 563. 8. (1972) 128 C.L.R. 427, at p. 438. 9. [1952] A.C. 1, at p. 14. 10. (1957) 97 C.L.R. 624. 11. (1968) 88 W.N. (Pt 1) (N.S.W.) 549, at p. 555. 12. (1969) 89 W.N. (Pt 1) (N.S.W.) 497, at pp. 503-504.

However the position has changed. There is now no appeal to the Privy Council, by special leave or otherwise, from any decision of this Court, except a decision given in a proceeding commenced before the 8th July 1975, and except for a decision on an inter se question if this Court has issued a certificate under s. 74: see Privy Council (Appeals from the High Court) Act 1975. There is no appeal to the Privy Council, by special leave or otherwise, from any State court exercising federal jurisdiction: s. 39 (2) of the Judiciary Act, as amended in 1968. However an appeal still lies direct ot he Privy Council from the Supreme Court of a State in any matter not given in the exercise of federal jurisdiction.

The effect of these legislative changes is extraordinary and perhaps unprecedented. It is that from the Supreme Courts of the Australian States, when not exercising federal jurisdiction, there are two final courts of appeal, neither of which is subordinate to the other. There is no doubt that the decisions of the Privy Council remain binding on the courts of the States. The question is whether they remain binding on this Court. A further question that arises is what course should be taken by the Supreme Court of a State in the event that a decision of this Court is in conflict with a decision of the Privy Council.

The modern English rule is usually said to be that "every court is bound to follow any case decided by a court above it in the hierarchy": Cross, Precedent in English Law, 2nd ed. 1968, p. 6; Allen, Law in the Making, 7th ed. (1964), p. 363. If this is the rule to be applied the result will be that this Court is no longer bound by decisions of the Privy Council, which now does not occupy a position above this Court in the judicial hierarchy, except in the case of inter se questions - a case that may be regarded as theoretical, since this Court, having regard to its settled practice, is unlikely ever to grant a certificate under s. 74 of the Constitution. Although the rules of precedent are not immutable, it seems to me that when the Parliament made this Court an ultimate court of appeal, with the responsibility of deciding finally and conclusively every question that it is called upon to consider, it must have intended that we should discharge that responsibility for ourselves, and that we should have the power and the duty to determine whether the decision of any other court, however eminent, should be followed in Australia. Part of the strength of the common law is its capacity to evolve gradually so as to meet the changing needs of society. It is for this Court to assess the needs of Australian society and to expound and develop the law for Australia in the light of that assessment. It would be an impediment to the proper performance of that duty, and inconsistent with the Court's new function, if we were bound to defer, without question, to every judgment of the Privy Council, no matter where the litigation in which that judgment was pronounced had originated, and even if we considered that the decision was inappropriate to Australian conditions or out of harmony with the law as it had been developed, and was being satisfactorily applied, in Australia.

It might be thought in theory that in this situation grave difficulties might arise, in that the Supreme Courts of the States may be faced with two lines of authority, each binding upon them, but in conflict. I am sure that fears of that kind will prove to be unfounded. On the one hand this Court will not differ from a decision of the Privy Council any more readily than we will depart from one of our decisions. We no longer treat ourselves as bound by the decisions of the House of Lords, but we nevertheless continue to recognize "their peculiarly high persuasive value": Skelton v. Collins [27] . We ought now to regard a decision of the Privy Council as even more highly persuasive, if that is possible, by reason of the very fact that its decisions remain binding on the States. On the other hand there is no room to doubt that in deciding appeals from Australia the Privy Council will give the same careful consideration to decisions of this Court as we shall give to decisions of the Board. Their Lordships have already shown recognition of the part which this Court has to play in deciding what should be the law for Australia - see e.g. Geelong Harbor Trust Commissioners v. Gibbs Bright & Co. [28] . The possibility that conflicting decisions will be given will be greatly reduced if this Court and the Judicial Committee continue to display to each other the mutual respect that already exists.

1. (1966) 115 C.L.R. 94, at p. 104. 2. [1974] A.C. 810, at pp. 820-821; (1974) 129 C.L.R. 576, at pp. 584-585.

It is not necessary, for the purposes of the present case, to attempt to lay down a set of rules to be applied by the Supreme Courts of the States in the event that a decision of this Court happens to be in conflict with a decision of the Privy Council. If this Court has considered a decision of the Privy Council, and has deliberately decided not to follow it, the State courts will be bound to act in accordance with the law as declared by this Court unless they are directed by a later decision of the Privy Council to take a different course. However it is possible to envisage other situations in which it might be proper for the State courts to follow the Privy Council rather than this Court - for example if the decision of this Court was an old one and obviously out of line with principles more recently established. Such cases may be left to be dealt with as they arise. Nothing that I have said affects the position of the State courts in relation to the decisions of the House of Lords and the Court of Appeal which, though not technically binding, should generally speaking be followed if they are applicable and are not themselves in conflict with a decision of this Court or of the Privy Council.

For the reasons I have given I consider that we are not bound by the decision in Palmer v. The Queen [29] but must decide whether that case or Reg. v. Howe [30] is in future to be followed in those Australian States where the matter is not governed by the provisions of the Codes.

This question cannot be decided simply by reference to authority. It cannot be said that at the time when Reg. v. Howe was decided, the law as to the effect of the use of excessive force in self-defence had been settled in England or in Australia. There were a number of cases in which it had been held that if excessive force is used in self-defence and death results the person responsible is guilty of manslaughter and not murder. These early cases have been fully discussed by the Supreme Court of South Australia in Reg. v. Howe [31] , by this Court in Reg. v. Howe [32] , and by Professors Norval Morris and Colin Howard in their Studies in Criminal Law (1964), pp. 127-131, and it would serve no useful purpose for me to add to the discussion. There was no case in which any appellate court in England had expressly recognized the principle that an accused who acts in self-defence but uses excessive force is guilty of manslaughter, although in R. v. Biggin [33] the Court of Criminal Appeal had mentioned, without comment, the fact that the trial judge had directed the jury that if the appellant had used more violence than was really necessary in the circumstances that would justify a verdict of manslaughter. The principle had been accepted by the Court of Appeal of British Columbia, although with very little discussion, in R. v. Barilla [34] , and that case had been followed in R. v. Ouellette [35] . There was support for the principle in cases decided in some jurisdictions in the United States. On the other hand, the facts in Mancini v. Director of Public Prosecutions [36] might seem to have called for the invocation of the principle, had it existed, but the question was never raised in argument or in the judgments in that case.

1. [1958] S.A.S.R. 95, at pp. 114-117. 2. (1958) 100 C.L.R., at pp. 461-462, 472-475. 3. [1920] 1 K.B. 213, at p. 219. 4. [1944] 4 D.L.R. 344. 5. (1950) 98 C.C.C. 153. 6. [1942] A.C. 1.

In Australia there was early, although somewhat obscure, recognition of the principle by the Supreme Court in New South Wales in Reg. v. Griffin [37] but it first achieved clear acceptance in Victoria in Reg. v. McKay [38] in the summing up of Barry J., and in the judgment of Lowe J. in the Court of Criminal Appeal. Lowe J. there stated the rule as follows [39] :

If the occasion warrants action in self-defence or for the prevention of felony or the apprehension of the felon, but the person taking action acts beyond the necessity of the occasion and kills the offender, the crime is manslaughter - not murder. After a careful review of the authorities the Supreme Court of South Australia in Reg. v. Howe took a similar (but not identical) view, which they expressed as follows [40] : We have come to the conclusion that it is the law that a person who is subjected to a violent and felonious attack and who, in endeavouring, by way of self-defence, to prevent the consummation of that attack by force exercises more force than a reasonable man would consider necessary in the circumstances, but no more force than he honestly believes to be necessary in the circumstances, is guilty of manslaughter and not of murder. 1. (1871) 10 S.C.R. (N.S.W.) (L.) 91, at pp. 100, 107. 2. [1957] V.R. 560. 3. [1957] V.R., at p. 563. 4. [1958] S.A.S.R., at pp. 121-122.

It will have been seen from this brief survey that the decision of this Court in Reg. v. Howe was not governed by authority. Dixon C.J. recognized this, when he said [41] that "there is no clear and definite judicial decision providing an answer to this question " Now that we have to reconsider the matter, in the light of the decision of the Privy Council, it is apparent that the question is not whether the law stated in Reg. v. Howe is supported by earlier authority, but whether it is sound in principle and likely to be beneficial in its operation.

It must immediately be said, with all respect, that it is difficult to define, with precision, the limits of the doctrine accepted by this Court in Reg. v. Howe. Dixon C.J., with whom McTiernan and Fullagar JJ. agreed, said [42] that "in substance the Supreme Court took a correct view of the consequences of the failure of a plea of self-defence to a charge of murder when it fails only because the deceased's death was occasioned by an excessive use of force, that is to say by force going beyond what was necessary in the circumstances or might reasonably be regarded in the circumstances as necessary". As I have already indicated, the Supreme Court of South Australia had stated the principle in such a way that it applied only where the person who was acting in self-defence had been subjected to a violent and felonious attack. However the manner in which the principle is stated in the judgment of Dixon C.J. suggests that it has a wider operation. Dixon C.J. said [43] :

The assumption made for the purpose of this question is that a man actually defending himself from the real or apprehended violence of the deceased has used more force than was justified by the occasion and that death has ensued from this use of excessive force. In all other respects, so it is assumed, the elements of a plea of self-defence existed. That is to say it is assumed that an attack of a violent and felonious nature, or at least of an unlawful nature, was made or threatened so that the person under attack or threat of attack reasonably feared for his life or the safety of his person from injury, violation or indecent or insulting usage. This would mean that an occasion had arisen entitling the person charged with murder to resort to force to repel force or apprehended force. Had he used no more force than was proportionate to the danger in which he stood, or reasonably supposed he stood, although he thereby caused the death of his assailant he would not have been guilty either of murder of manslaughter. But assuming that he was not entitled to a complete defence to a charge of murder, for the reason only that the force or violence which he used against his assailant or apprehended assailant went beyond what was needed for his protection or what the circumstances could cause him reasonably to believe to be necessary for his protection, of what crime does he stand guilty? Is the consequence of the failure of his plea of self-defence on that ground that he is guilty of murder or does it operate to reduce the homicide to manslaughter? He answered this question by saying that "it seems reasonable in principle to regard such a homicide as reduced to manslaughter" [44] . In the passage cited, Dixon C.J. appears to be describing, in decreasing order of gravity, the situations in which excessive force used in self-defence may reduce the homicide to manslaughter. His judgment suggests that it is enough that there was an attack which was unlawful, even though it was not violent or felonious, and that the person under attack feared for the safety of his person from insulting usage - not necessarily from injury. Taylor J. also appears to have given the principle wide scope - it was in his view "sufficient if it appears that what the accused did was done primarily for the purpose of defending himself against an aggressor" [45] . Menzies J. on the other hand expressly confined what he had to say to a case of self-defence against serious violence, though not necessarily felonious violence [46] . 1. (1958) 100 C.L.R., at p. 462. 2. (1958) 100 C.L.R., at pp. 460-461. 3. (1958) 100 C.L.R., at p. 461. 4. (1958) 100 C.L.R., at p. 468. 5. (1958) 100 C.L.R., at pp. 471, 474.

The Supreme Court of Victoria has adopted the more limited view of Menzies J., and has held that homicide will not be reduced to manslaughter if the excessive force was used in self-defence against an assault which merely threatened the person assaulted with some minor form of violence or injury which no reasonable man could consider warranted action to kill or inflict grievous bodily harm: Reg. v. Enright [47] ; Reg. v. Tikos [No. 2] [48] . In the latter case, the Court said [49] :

We would think, on principle, that it must be correct to say that the crime of murder which, in general, involves as a necessary element an intention to kill or to inflict grievous bodily harm, cannot be justified or reduced to manslaughter under a plea of self-defence unless the occasion be one which warranted the accused acting with intent to do some kind of grievous bodily harm at the least. This conclusion is consistent with the judgment of the Supreme Court of South Australia in Reg. v. Howe [50] , but it is narrower than the statement of Lowe J. in Reg. v. McKay [51] , and, it would appear, than the opinions expressed by the majority of this Court in Reg. v. Howe [52] . 1. [1961] V.R. 663, at pp. 668-669. 2. [1963] V.R. 306. 3. [1963] V.R., at p. 313. 4. [1958] S.A.S.R. 95. 5. [1957] V.R. 560. 6. (1958) 100 C.L.R. 448.

The fact that this Court in Reg. v. Howe allowed to remain obscure the question whether the principle which was there approved applies to every case of self-defence, or is limited to cases of self-defence against serious violence, casts doubt upon the validity of the principle, which would operate very differently depending on whether or not it is so limited. Let it first be supposed that it is correct to say that homicide caused by excessive force used in self-defence is not reduced to manslaughter unless the accused was defending himself against an attack involving serious violence. A person threatened with serious violence (which, it is of course assumed, is unlawful) could reasonably think it necessary to defend himself by using force which was intended to cause grievous bodily harm to his assailant. On the supposition now made (which accords with the view accepted in Victoria) an accused person, although placed in such a position of danger that he is lawfully entitled intentionally to inflict some grievous bodily harm on his assailant, may nevertheless be convicted of manslaughter if the amount of harm which he inflicts is greater than is considered reasonably necessary. If the principle is limited in this way, there can be very few cases to which it can properly be applied, because it will not be often that it is possible to say that although it was reasonable to inflict some grievous bodily harm, it was not reasonable to cause harm to the extent that actually occurred, or by the method actually used. However, a principle of this kind would seem completely unreal from a practical point of view. A person faced with a threat of serious violence cannot be expected to weigh calmly and meticulously the exact degree of force which he should use in self-defence. To allow a jury to return a verdict of manslaughter when it was reasonable for the accused to cause some grievous bodily harm in his own defence, but it was unreasonable for him to use as much force, or the kind of force, that he actually used, would detract from the practical value of the defence, and invite the possibility of a compromise verdict of manslaughter.

Let it now be supposed that the other possible view is correct, and that the principle applies where force is used in self-defence against an actual or apprehended attack which is not serious or violent and which causes the person attacked to fear no more than insulting usage. If the principle applied in those circumstances, it would reduce to manslaughter a deliberate killing done by a person for the purpose of defending himself against the most trivial of assaults. Some striking, if rather extreme, instances of the way in which the principle might operate if it is not limited to self-defence against serious violence are given by Sholl J. in Reg. v. Tikos [No. 1] [53] . For example, should it be manslaughter if one man shoots another, intending to kill him, because he honestly believes that that is the only way in which he can prevent the other from throwing a bottle of ink over him? There are some who see no need to retain the distinction between murder and manslaughter, but while it remains there seems no reason in principle or policy to reduce homicides of this kind, which might be utterly unreasonable and inexcusable by ordinary standards, to manslaughter. If this second possible view of the scope of the principle is correct, the principle will have a wide and irrational operation.

Another matter of importance that is not clearly resolved by the judgments in Reg. v. Howe is whether the suggested principle only applies when the accused used no more force than he honestly believed to be necessary. The Supreme Court of South Australia in Reg. v. Howe [1] regarded that as essential for the application of the principle, and in this Court Menzies J. appears to have been of the same opinion [2] . However, the judgment of Dixon C.J. does not make it entirely clear whether he considered that a homicide is only reduced to manslaughter if the force used in self-defence was honestly believed by the accused to be necessary, and Taylor J. appears to have rejected the view that this is the correct test [3] . Taylor J. expressed his final conclusion, somewhat elliptically, as follows [4] :

It is in my view, sufficient if it appears that what the accused did was done primarily for the purpose of defending himself against an aggressor and the jury should be instructed that unless satisfied beyond reasonable doubt that this was not so a verdict of manslaughter should be returned. In Johnson v. The Queen [5] Wooding C.J., delivering the judgment of the Court of Criminal Appeal of Trinidad and Tobago, accepted the test suggested by Taylor J. but further refined it. He said [6] that the question for the jury is "What is the true or primary intention?" and that they should not convict unless satisfied that the accused's true or primary intention was to kill or inflict grievous bodily harm. It seems to me, with the greatest respect, that this judgment brings out the difficulty latent in the view expressed by Taylor J. By hypothesis, the accused in a case such as that under consideration had the intention to defend himself and also had the intention to kill or do grievous bodily harm. Unless each could be said to be a "true intention" the doctrine cannot be applied. Unless the reference to "primary intention" is intended to exclude cases in which the accused used the occasion as a mere excuse to kill or injure the aggressor - cases where he acted "under pretence of necessity" (to use words often cited from Hawkins, Pleas of the Crown (1716), Book 1, ch. 28, s. 2) - it seems impossible to say that one intention is primary and the other is not. In any case it does not seem sensible for a jury on a trial for murder to be asked to decide which of the intentions co-existing in the mind of the accused was the dominant one. 1. [1958] S.A.S.R. 95. 2. (1958) 100 C.L.R., at pp. 471, 477. 3. (1958) 100 C.L.R., at pp. 466-467. 4. (1958) 100 C.L.R., at p. 468. 5. (1966) 10 W.I.R. 402. 6. (1966) 10 W.I.R., at p. 416.

For these reasons I am of the opinion that the principle enunciated in Reg. v. Howe is not one which ought to be accepted. It is likely to lead a jury to confusion and error. Nor do I regard it as sound in legal theory. Taylor J. in Reg. v. Howe [7] said that cases of excessive self-defence "may, as in clear cases of sufficient provocation, be taken as sufficient to prevent the implication that the killing was malicious in the sense in which that term has come to be understood in relation to the crime of murder". But the view formerly expressed that provocation negatives malice (per Viscount Simon in Holmes v. Director of Public Prosecutions [8] ) does not appear consistent with the more recent judgments in Attorney-General (Ceylon) v. Perera [9] and Lee Chun-Chuen v. The Queen [10] . And it is hardly necessary to add that where excessive force has been used in self-defence the doctrine of provocation may reduce the homicide to manslaughter - if the principle of Reg. v. Howe were correct, it would cover part of the same ground as the doctrine of provocation but would extend the defence to cases where the accused had not lost his self-control.

1. (1958) 100 C.L.R., at p. 467. 2. [1946] A.C. 588, at p. 598. 3. [1953] A.C. 200, at p. 206. 4. [1963] A.C. 220, at pp. 227-228.

For these reasons, in my opinion, the decision of the Privy Council in Palmer v. The Queen [11] should be preferred to that of this Court in Reg. v. Howe. But since writing the foregoing I have had an opportunity to read the reasons prepared by the other members of the Court. It is apparent that we hold a diversity of opinions. It seems to me that we would be failing in our function if we did not make it clear what principle commands the support of the majority of the Court. The task of judges presiding at criminal trials becomes almost impossible if they are left in doubt what this Court has decided on a question of criminal law. In the present case the view which appears to have more support than any other is that we should accept as correct the statement of Dixon C.J. in Reg. v. Howe. Contrary to my personal opinion, but in a desire to achieve a measure of certainty, I am prepared to agree. Mason J., in his judgment, has stated the task of the jury where threat of death or grievous bodily harm to the accused is in question and the issue of self-defence arises. In future that statement should be accepted as correct.

On this view, if there had been material sufficient to raise an issue of self-defence in the present case the direction to the jury on the issue on self-defence would have been insufficient. It is unnecessary to consider that matter further because, as I have already held, the learned trial judge erred in omitting to tell the jury that they might have regard to the fact that the applicant had taken heroin when they were considering whether they were satisfied that he had intended to kill Rellis or do him grievous bodily harm. Opinions may well differ as to whether this omission was seriously prejudicial in all the circumstances of the case. However after full consideration I find it impossible to be satisfied that the jury would inevitably have reached the same result if they had been fully directed on the point. I therefore consider that the conviction cannot be allowed to stand.

I would grant special leave to appeal, allow the appeal, set aside the conviction and order a new trial.

On this application for special leave to appeal the applicant relies upon two principal grounds. The first concerns the failure of the learned trial judge to instruct the jury that the applicant's use of heroin might be relevant to the question of whether he had formed the necessary intention to kill or do grievous bodily harm. Gibbs J. has dealt at some length with this ground and has concluded that the charge was defective in this respect. I agree with this conclusion and with his Honour's view that it is in itself enough to require a grant of special leave and an order for a new trial.

The other main ground of appeal also relates to the trial judge's charge. The applicant contends that the jury should have been instructed that they might return a verdict of manslaughter if the applicant had killed the deceased in self-defence but had, in doing so, used excessive force. For this contention he relies upon the decision of this Court in Reg. v. Howe [12] with which is to be contrasted the subsequent, and inconsistent, decision of their Lordships in Palmer v. The Queen [13] .

Two distinct questions are involved, first whether this Court is bound to follow Palmer or is instead free, if it sees fit, to adhere to the law as laid down in Howe; secondly if it be free to choose between these two conflicting decisions, which of them it should prefer.

The answer to the first question cannot, I think, be in doubt. The abolition of all appeals from decisions of this Court, resulting from the Privy Council (Appeals from the High Court) Act 1975 Cth, leaves this Court as the ultimate court of appeal in all cases which come before it and this in turn requires it to decide for itself what is the true state of the law.

The first duty of a court is to administer justice according to law. However in the case of an inferior court operating within a system where the doctrine of precedent applies, the existence of authority binding upon it determines for it what it must understand to be the law. It must accept the law to be as that precedent authority has declared it to be, whatever may be its own inclinations in the matter. The sanction implicit in the doctrine of precedent is simple and effective: if an inferior court fails to observe the doctrine the superior court will correct its decision on appeal. Thus the existence of an appeal is inherent in and essential to the doctrine.

The position of a final court of appeal, its decisions subject to no appeal to a court superior to it, is otherwise. It too must administer justice according to law; as Isaacs J. said in Australian Agricultural Co. v. Federated Engine-Drivers' and Firemen's Association of Australasia [14] "our sworn loyalty is to the law itself". But it may neither surrender, nor be relieved of, its responsibility to find what is the law by any involuntary adoption of the decisions of any other court. It may impose upon itself a rule that it will accept as absolute the binding force of its own past decisions, a course which, for the several reasons referred to by Dixon J. in Attorney-General (N.S.W.) v. Perpetual Trustee Co. Ltd. [15] the High Court has not done. Subject only to that possibility, it must otherwise wholly accept the responsibility of itself declaring what it regards to be the law, even if the views of other tribunals, however respected, are to a contrary effect. Their views will, of course, be of great force in the formation of its own conclusion but at the end of the day what must prevail must be its own conclusion.

1. (1913) 17 C.L.R. 261, at p. 278. 2. (1952) 85 C.L.R. 237, at p. 244.

Accordingly this Court will accord great weight both in their Lordships' views and to those expressed in Howe's Case and will always be very conscious of the need for certainty and consistency in the law. However, it will be for it to arrive at its own conclusion as to what is the law for Australia, for that purpose making its choice, in the present circumstances, between its own earlier decision and that of their Lordships.

Before passing to the second question, involving the outcome of that process of choosing between these two conflicting decisions, it is necessary to say something more concerning the doctrine of precedent and the effect upon it, as it applies within the Australian judicature, of the Privy Council (Appeals from the High Court) Act 1975. The effect of the Act has, in my view, been to distort within Australia, for the time being at least, that symmetrical hierarchy upon which the doctrine of precedent depends for its operation. The doctrine assumes a hierarchy of courts with, at its head, some ultimate court of appeal. It can accommodate itself to the existence of two or more ultimate appellate courts so long as their respective jurisdictions are exclusive, the one of the other. But when their jurisdictions are in any respect coterminous, the doctrine's operation is to that extent impaired. This is now the situation in an extensive area of Australian law. The abolition of appeals from the High Court; the High Court's consequent assumption of the role of a final court of appeal and with it of the responsibility which that involves ultimately to determine for itself the true state of the law; the continued existence of a right of appeal to the Privy Council from certain cases decided in State Supreme Courts; all these together form the ingredients of the present situation.

It is for inferior courts in the hierarchy, when faced with inconsistent precedents each of binding authority, that the coexistence of two such courts of ultimate recourse, neither necessarily deferring to the other, presents a problem in the resolution of which the doctrine of precedent provides no guidance. They cannot tell which precedent should be followed and which cast aside. Nor can any unilateral declaration by either of these two coordinate appellate courts, unless it be of a self-denying character, effectively resolve their dilemma. The vice lies in the decisive power which the present situation confers upon appellants to select the appellate forum of their choice. True it is that the problem will only be encountered within those relatively confined areas where conflict of authority exists, but this, while reducing its incidence, does nothing to render the problem itself any more tractable.

For this Court to declare that State Supreme Courts should, in cases of conflict, prefer its decisions to those of the Privy Council will be ineffective. If observed, unsuccessful Supreme Court litigants will tend always to appeal to the Privy Council, whose earlier decision the Supreme Court will, in observance of this Court's declaration, have rejected. The outcome may well be reversal of the decision of the Supreme Court. Is the Supreme Court thereafter to court further reversals, conduct in itself a denial of the doctrine of precedent? If instead it subsequently follows the Privy Council precedent, the next appeal may come to this Court which may well adhere to its own earlier decision. Supreme Courts will, in areas of conflicting precedent, be at the mercy of appellants; uncertainty will prevail and the outcome in each appeal will in some measure lie in the hands of the loser before the Supreme Court, who will thereby obtain the right to select the appellate forum of his choice.

For this Court to declare that the most recent of two conflicting decisions should be followed offers, I think, no better solution; again the appellant will tend to appeal to the forum whose precedent decision, earlier in point of time, has been rejected by the Supreme Court. It may be said that at least this course will not, as will the other, tend to drive all appellants to resort to the Privy Council; this Court will as often as not be the forum resorted to by the appellant, since the latest decision, to be followed by a Supreme Court, is as likely to be that of their Lordships as of this Court. However, having thus secured for itself a due share of such appeals by directing Supreme Courts to follow the latest of two conflicting decisions, this Court will not itself be constrained by that same rule. Indeed the virtue of this course, curiously enough, lies solely in its effect of requiring Supreme Courts on occasions to reject precedents of this Court in favour of those of their Lordships, with the anticipated consequence that appeals will then come to this Court. If this be its virtue it would be more perfectly attained by requiring Supreme Courts in all cases to prefer Privy Council precedents, so that all appeals might then flow to this Court. Such a sacrifice of principle for expediency cannot, of course, be countenanced and serves only to illustrate the undesirable aspects of the second course suggested.

In my view the present situation is one incapable of remedy by any form of unilateral declaration of one or other of the two final courts of appeal. Until the matter is otherwise resolved the doctrine of binding precedent must, I think, be regarded as a casualty of events whenever a conflict arises between decisions of this Court and of their Lordships. Moreover, despite its purportedly authoritative character, any unilateral declaration will lack effective sanction should the actual course of appeal in a particular case not lead to the Court which issues that declaration. It is a necessary feature of the present situation that there is absent that sanction upon which the doctrine of binding precedent depends, the power of the ultimate court of appeal to reverse decisions which do not conform to the doctrine. The irony of the situation is that only by directing Supreme Courts to prefer the precedents of the Privy Council to its own can this Court ensure that appeals come to it and become subject to that sanction.

So long as the present situation continues I would, for the foregoing reasons, be reluctant to join in any unilateral declaration by this Court directed to the course which Supreme Courts should follow. Unsatisfactory as it may appear, the lesser evil is, I think, to acknowledge the breakdown in the doctrine of binding precedent which now occurs when conflicting decisions exist and to leave it to the Supreme Courts to make their choice as between such decisions.

In what I have already said it is implicit that I regard this Court as not merely free to, but under an obligation to, determine for itself as between Howe [16] and Palmer [17] which it should follow. I favour that view of the law applicable to self-defence as appearing in Howe and conclude that for Australia this should be the law.

I do so primarily in response to what I regard as the need of the criminal law to distinguish, in terms of degree of culpability, between, on the one hand, the crime of murder and, on the other, a killing in circumstances which would have justified a recourse to violence in self-defence but which did not, contrary to the accused's belief, justify that degree of violent response which in fact ensued. The law should not, I think, exclude from account the existence on the part of the accused of an honest although mistaken belief that the degree or kind of violence with which he has answered the aggression against him was reasonable in the circumstances. That belief is an important element in culpability. Due weight will only be given to it if the jury is instructed that, although it concludes that the accused's responsive violence was unreasonably excessive, the killing nevertheless amounts to no more than manslaughter unless the jury is satisfied that the accused did not possess the belief that his response was reasonably proportionate to the danger which he faced.

In the absence of such an instruction it cannot, I think, be assumed that a jury will necessarily treat an accused's honest belief as to reasonable measures of self-defence as answering in the affirmative the question whether "only reasonable defensive action had been taken" (see Palmer [18] ). Assume a jury of good sense, directed that one who kills while defending himself "may do, but may only do, what is reasonably necessary" (Palmer [19] ): such a jury, while perfectly capable of distinguishing between what was reasonably necessary and what the accused in fact believed to be reasonably necessary, may nevertheless feel obliged, in obedience to such a direction, to convict of murder. Yet an accused who honestly believed that his response to aggression was a reasonable one will lack that degree of culpability which a murderer possesses. Because of the excessive force to which he has resorted, the killing, otherwise in lawful self-defence, ill be unlawful; it should not, I think, amount to murder but, rather, to manslaughter.

1. [1971] A.C., at p. 832. 2. [1971] A.C., at p. 831.

In Howe [20] the South Australian Full Court spoke of such a case as one of "unlawful killing without malice aforethought, for although the killer may clearly intend to inflict grievous bodily harm on his assailant, and if necessary, to kill, his state of mind is not fully that required to constitute murder". Accordingly their Honours concluded that when, in defence against a violent and felonious attack, death results from the exercise of more force than a reasonable man would consider necessary but not more than the person attacked honestly believed to be necessary in the circumstances it is manslaughter, not murder [21] .

1. [1958] S.A.S.R. 95, at p. 122. 2. [1958] S.A.S.R., at pp. 121-122.

When the case came on appeal to this Court Dixon C.J. set out [22] the passage in which the Full Court expressed this conclusion and, after reviewing the authorities, he himself concluded that [23] :

in substance the Supreme Court took a correct view of the consequences of the failure of a plea of self defence to a charge of murder when it fails only because the deceased's death was occasioned by an excessive use of force, that is to say by force going beyond what was necessary in the circumstances or might reasonably be regarded in the circumstances as necessary. There are passages in his Honour's judgment which might be thought to suggest that the accused's belief as to the proportionality of his response to the attack upon him is irrelevant. However, read as a whole, I do not understand his Honour as doing otherwise than approve of what had been said in the South Australian Full Court. McTiernan J. and Fullagar J. agreed with the judgment of the Chief Justice. Menzies J., in a separate judgment [24] , concluded that it was manslaughter, not murder, if an accused would have been entitled to acquittal on the ground of self defence but for the fact "that in honestly defending himself" he used excessive force and killed his assailant. Thus his Honour clearly enough regarded the accused's state of mind as to his acts in defence of himself as relevant in arriving at a verdict of manslaughter rather than murder when excessive force had been used. 1. (1958) 100 C.L.R., at p. 456. 2. (1958) 100 C.L.R., at p. 462 3. (1958) 100 C.L.R., at p. 477.

A reading of the extensive academic literature which has appeared since Howe's Case was decided, of decisions in other common law jurisdictions in which Howe's Case has been considered, and of what their Lordships have said in Palmer leaves me preferring, with respect, the view which I regard as adopted by a majority of the members of this Court in Howe's Case. That view was at the time no novelty in the law and has since been followed in a number of reported cases both in Australian jurisdictions and overseas: reference to these is made in the reasons for judgment of Mason J. With all that his Honour there says about the still unresolved limits of the doctrine of self-defence and, more generally, about the developing concept of malice aforethought and the possibility of it being absent despite an intention to kill or inflict grievous bodily harm I agree. At the conclusion of his Honour's judgment appears a formulation of the issues which I regard as properly falling for determination when a jury is called on to consider self-defence. It is by such a formulation that what I regard as the principles of Howe's Case may be expressed.

I would grant special leave and would allow this appeal.

Two questions were argued before the Full Court of seven Justices -

(1) Whether this Court is bound to follow the decision of the Privy Council in Palmer v. The Queen [25] notwithstanding this Court's earlier decision in Reg. v. Howe [26] ; and

(2) If not, what is the proper direction that should be given to the jury when self-defence is raised, the answer to this question requiring a consideration of the conflicting views expressed in Palmer v. The Queen and Reg. v. Howe.

I would answer the first of these questions in the negative and I would do so for the reasons given by Gibbs J. to support this conclusion. To those reasons I would merely add the comment that as this Court has never regarded itself as bound by its previous decisions it would be quite incongruous that it should be bound by decisions of the Privy Council now that this Court's decisions are final and are not subject to appeal to the Privy Council.

I would express the attitude which this Court should take to Privy Council decisions in consequence of the commencement of the Privy Council (Appeals from the High Court) Act 1975 by saying that we should accord them the highest respect. But in so saying I would emphasize that it is well recognized that the common law may develop differently in Australia from the common law as it develops in England and other countries from which an appeal lay or now lies to the Privy Council. It is the responsibility of this Court to determine ultimately what is the law for Australia. In discharging this responsibility this Court can apply its knowledge of Australian conditions and circumstances; it is conversant with the development of Australian law in both its statutory and non-statutory forms; it therefore has a unique capacity to decide whether a particular rule or principle of law is appropriate or adapted to Australian conditions and circumstances, to interpret and apply Australian statutes in the light of Australian circumstances and to apply Australian law as it has developed.

Lord Diplock, speaking for the Judicial Committee in Geelong Harbour Trust Commissioners v. Gibbs Bright & Co. [27] drew attention to the fact that the way in which the law, as enunciated by a particular judicial decision, works in practice is an important factor to be weighed against the more theoretical interests of legal science, and said [28] :

The High Court, sitting regularly in the capitals of the various States which are also the main ports and commercial centres of Australia, is much better qualified than their Lordships are to assess the importance of this factor. 1. [1974] A.C. 810; (1974) 129 C.L.R. 576. 2. [1974] A.C., at p. 819; (1974) 129 C.L.R., at p. 582.

Later, his Lordship, after observing that the extent to which the High Court exercises its power not to adhere to a previous decision of its own must be consonant with the consensus of opinion of the public, of the legislature and of the judiciary as to the proper balance between the respective roles of the legislature and of the judiciary, went on to say [29] :

Even among those nations whose legal system derives from the common law of England, this consensus may vary from country to country and from time to time. It may be influenced by the federal or unitary nature of the constitution and whether it is written or unwritten, by the legislative procedure in Parliament, by the ease with which parliamentary time can be found to effect amendments in the law which concern only a small minority of citizens, by the extent to which Parliament has been in the habit of intervening to reverse judicial decisions by legislation; but most of all by the underlying political philosophy of the particular nation as to the appropriate limits of the lawmaking function of a non-elected judiciary.

The High Court of Australia can best assess the national attitude on matters such as these.

See also Australian Consolidated Press Ltd. v. Uren [30] ; Abbott v. The Queen [31] . 1. [1974] A.C., at pp. 820-821; (1974) 129 C.L.R., at pp. 584-585. 2. (1967) 117 C.L.R. 221, at p. 241; [1969] 1 A.C. 590, at p. 644. 3. [1977] A.C. 755, at p. 768.

By these observations the Privy Council has acknowledged that this Court is pre-eminently equipped to decide what is the law for Australia.

State courts are bound both by decisions of the Privy Council and of this Court. Formerly, when appeals could be taken from this Court to the Privy Council, State courts were bound to follow the Privy Council where a Privy Council decision conflicted with a decision of this Court. Under the new dispensation it is no longer appropriate that decisions of the Privy Council should be accorded higher authority in the Australian judicial system than decisions of this Court. As this Court is the ultimate appellate court in Australia and as it has the advantages mentioned earlier in deciding what is the appropriate law for Australia, State courts should, as a general rule, follow decisions of this Court in case of conflict. Of course every general rule has its exceptions or qualifications. Here an exception must be allowed for the case where the Privy Council, after taking into consideration a decision of this Court, has decided not to follow it. In such a case a State court should follow the Privy Council unless its decision appears to be based on considerations that are not relevant to Australian circumstances or conditions. If the State court follows the Privy Council in such a case it is more likely than not that an appeal will be brought to this Court, thereby enabling this Court to decide whether it will adhere to its earlier decision or whether it will adopt the later decision of the Privy Council. Despite the adoption of the rule in Bakhshuwen v. Bakhshuwen [32] and Morris v. English, Scottish and Australian Bank Ltd. [33] that a Privy Council decision on a question of law is binding on all courts from which an appeal lies directly or indirectly to the Privy Council, the recognition of the potential for different development of the common law in various countries entails that there will be some cases in which Australian conditions and circumstances are such as to require a Supreme Court to decline to follow the Privy Council decision. In some cases it will appear from the terms of that decision that it is based on considerations that are inapposite to Australia, but in other cases this conclusion may be drawn from other materials.

1. [1952] A.C. 1, at p. 14. 2. (1957) 97 C.L.R. 624.

In all that I have said so far I have assumed, without definitely deciding, that the Privy Council (Appeals from the High Court) Act 1975 is a valid enactment of the Commonwealth Parliament as a law "limiting the matters in which such leave may be asked" within the meaning of s. 74 of the Constitution - see Kitano v. The Commonwealth [34] . I express this qualification, not out of any desire to raise doubts as to the validity of the law, but merely because the meaning of the concept of limitation in s. 74 was left open in Kitano's Case and because no argument was addressed to us upon the question.

As the Court is not bound to follow either Reg. v. Howe [35] or Palmer v. The Queen [36] the conflict between the two decisions calls for a re-examination of the question. Speaking broadly, the available choice lies in acceptance or rejection of the proposition that in cases of self-defence the jury should be directed that, if the defence fails only because the jury considers the force used to be excessive, a verdict of manslaughter should be returned. The proposition is affirmed by the unanimous decision in Reg. v. Howe, by other Australian cases (see Reg. v. McKay [37] ; Reg. v. Bufalo [38] ; Reg. v. Haley [39] ; Reg. v. Enright [40] ; Reg. v. Turner [41] ; Reg. v. Tikos [No. 1] [42] ; Reg. v. Tikos [No. 2] [43] ) and by decisions of courts in British Columbia (R. v. Barilla [44] ), Ireland (The People (Attorney-General) v. Dwyer [45] ), Trinidad and Tobago (Johnson v. The Queen [46] ) and in Jamaica (Reg. v. Hamilton [1] ), although in the West Indian cases it was the proposition as formulated by Taylor J. in Reg. v. Howe that was affirmed. This formulation is a variant of the proposition stated by other members of the Court, in particular that expressed by Dixon C.J. The proposition stated by Dixon C.J. is consistent with the law as it has developed in some States of the United States. But its correctness has been denied by the Privy Council in Palmer v. The Queen and subsequently by the Court of Criminal Appeal in Reg. v. McInnes [2] . Earlier the proposition had been rejected in the West Indies in De Freitas v. The Queen [3] but this decision was not followed in Johnson v. The Queen and Reg. v. Hamilton. Since Reg. v. McKay was decided in 1957 the topic has been canvassed extensively and comprehensively in academic writings (see, for example, Norval Morris, The Slain Chicken Thief Sydney University Law Review vol. 2 (1958), p. 414; Morris and Howard, Studies in Criminal Law , Ch. 4 ; Howard, Australian Criminal Law (1965), pp. 80-83; "Two Problems in Excessive Defence" Law Quarterly Review, vol. 84 (1968), p. 343; Smith and Hogan, Criminal Law, 3rd ed. (1973), pp. 263 et seq).

1. (1958) 100 C.L.R. 448. 2. [1971] A.C. 814. 3. [1957] V.R. 560. 4. [1958] V.R. 363. 5. (1959) 76 W.N. (N.S.W.) 550. 6. [1961] V.R. 663. 7. [1962] V.R. 30. 8. [1963] V.R. 285. 9. [1963] V.R. 306. 10. (1944) 4 D.L.R. 344. 11. [1972] I.R. 416. 12. (1966) 10 W.I.R. 402. 13. (1967) 11 W.I.R. 309. 14. [1971] 1 W.L.R. 1600; [1971] 3 All E.R. 295; 55 Cr. App. R. 551. 15. (1960) 2 W.I.R. 523.

I have examined these and other materials in some detail because Mr. Cripps, for the appellant, and the Solicitor-General, who appeared for the Crown, each submitted that the doctrine enunciated in Reg. v. Howe is to be preferred to Palmer v. The Queen. In the result I have come to the conclusion that Reg. v. Howe was correct and that this Court should continue to follow that decision, despite the great weight to be accorded to Palmer v. The Queen. I shall endeavour to express my reasons for this conclusion without discussing all the decided cases in detail, as they have been discussed at length elsewhere.

Notwithstanding some suggestions to the contrary, notably in Reg. v. McInnes [4] and in Reg. v. Hassin [5] , the doctrine enunciated in Reg. v. Howe [6] is not a novel development in the criminal law without any previous foundation in judicial decisions. There are to be found in the cases a number of indications that excessive force used by an accused person in defending himself against an aggressor should result in a conviction for manslaughter, not murder. These cases are referred to in the judgments of the Court in Reg. v. Howe and by Morris and Howard in their Studies in Criminal Law, pp. 127-136. Although they do not themselves furnish a decisive or authoritative answer to the question, these cases are sufficient to dispose of the suggestion that the doctrine emerged of a sudden, so to speak, without benefit of reputable antecedents in the law. In Reg. v. Howe the view was taken, and in my opinion correctly taken, that the weight of authority pointed to the existence of the doctrine. Not that the weight of authority was so strong as to be irresistible. When Reg. v. Howe came to be decided, the issue was very much an open question the answer to which depended more upon an evaluation of the considerations inherent in murder and manslaughter than upon the force of previous authority.

1. [1971] 1 W.L.R., at p. 1608; [1971] 3 All E.R., at p. 301; 55 Cr. App. R. 551. 2. [1963] Crim. L. Rev. 852. 3. (1958) 100 C.L.R. 448.

The underlying rationale of Reg. v. Howe is to be found in a conviction that the moral culpability of a person who kills another in defending himself but who fails in a plea of self-defence only because the force which he believed to be necessary exceeded that which was reasonably necessary falls short of the moral culpability ordinarily associated with murder. The notion that a person commits murder in these circumstances should be rejected on the ground that the result is unjust. It is more consistent with the distinction which the criminal law makes between murder and manslaughter that an error of judgment on the part of the accused which alone deprives him of the absolute shield of self-defence results in the offence of manslaughter.

In Palmer v. The Queen [7] these considerations were dismissed as illusory. The possibility that a jury would distinguish between the accused's belief as to the degree of force which was reasonably necessary and that which a reasonable man placed in the accused's situation would believe to be necessary was held to be academic or hypothetical, a possibility which is never likely to arise, because, so it was said, a jury which finds that the accused honestly believed that the force which he used was necessary will always find that the degree of force which he used was reasonably necessary. I am not persuaded that juries will always come to this conclusion or that there is any sound basis on which a prediction to this effect can confidently be made.

The principle that the jury may take into account and give weight to the accused's belief as to what was necessary in deciding what a reasonable man in his situation would believe to be necessary itself acknowledges that there is a distinction between the accused's subjective belief and the objective standard which may in a given case prove decisive. If it be correct to say that Reg. v. Howe is erroneous, and this because in all cases the jury will invariably find an exact correspondence between the accused's subjective belief and the objective standard, then one would expect the law to acknowledge an identity between the two by formulating the principle in terms of the accused's belief or by requiring that the jury be instructed accordingly. Such an approach is denied by Palmer v. The Queen for the acknowledgement that the accused's belief is "most potent evidence" of the objective standard does not completely bridge the gap. Indeed, the more one reflects upon the approach taken in Palmer v. The Queen the less reason there would appear to be for retaining the objective standard as an element in the defence. Apart from the sense of continuity which it provides by way of a link with earlier authority, it serves little purpose if it is no more than a reflection of the accused's honest belief.

As other criticisms have been made of what was said in Reg. v. Howe I shall say something of that case and of what it decided. It will be recalled that the Full Court of the Supreme Court of South Australia, in allowing the appeal from the conviction, had held that [8] :

a person who is subjected to a violent and felonious attack and who, in endeavouring, by way of self-defence, to prevent the consummation of that attack by force exercises more force than a reasonable man would consider necessary in the circumstances, but no more force than he honestly believes to be necessary in the circumstances, is guilty of manslaughter and not of murder. The majority of this Court in substance agreed with this statement of the law and rescinded the grant of special leave to appeal. This course was taken because the Crown wished to argue other matters on the appeal, matters which would not of themselves have attracted special leave. 1. (1958) 100 C.L.R., at p. 456.

Dixon C.J. (with whom McTiernan and Fullagar JJ. agreed), by way of explanation of the issue, said [9] :

it is assumed that an attack of a violent and felonious nature, or at least of an unlawful nature, was made or threatened so that the person under attack or threat of attack reasonably feared for his life or the safety of his person from injury, violation or indecent or insulting usage. This would mean that an occasion had arisen entitling the person charged with murder to resort to force to repel force or apprehended force. Had he used no more force than was proportionate to the danger in which he stood, or reasonably supposed he stood, although he thereby caused the death of his assailant he would not have been guilty either of murder or manslaughter. But assuming that he was not entitled to a complete defence to a charge of murder, for the reason only that the force or violence which he used against his assailant or apprehended assailant went beyond what was needed for his protection or what the circumstances could cause him reasonably to believe to be necessary for his protection, of what crime does he stand guilty? Later, the Chief Justice stated his conclusion in these terms [10] : in substance the Supreme Court took a correct view of the consequences of the failure of a plea of self-defence to a charge of murder when it fails only because the deceased's death was occasioned by an excessive use of force, that is to say by force going beyond what was necessary in the circumstances or might reasonably be regarded in the circumstances as necessary. 1. (1958) 100 C.L.R., at pp. 460-461. 2. (1958) 100 C.L.R., at p. 462.

Taylor J. thought that a direction founded on Dixon C.J.'s view of the law would tender an artificial view to the jury. He pointed out that in Reg. v. McKay [11] Lowe J., in saying "If the occasion warrants action in self-defence or for the prevention of felony or the apprehension of the felon, but the person taking action acts beyond the necessity of the occasion and kills the offender, the crime is manslaughter - not murder", did not limit the statement to "cases where it appears that the accused entertained an honest belief that the force used, though excessive on any reasonable view, was necessary" [12] . His Honour concluded that the principle rested on a broader basis than the accused's honest, though unreasonable, belief and said that in his view it was "sufficient if it appears that what the accused did was done primarily for the purpose of defending himself against an aggressor and the jury should be instructed that unless satisfied beyond reasonable doubt that this was not so a verdict of manslaughter should be returned" [13] .

1. [1957] V.R., at p. 563. 2. (1958) 100 C.L.R., at pp. 466-467. 3. (1958) 100 C.L.R., at p. 468.

Menzies J. expressed the view that in cases of self-defence against serious, though not necessarily felonious, violence where the accused "in honestly defending himself used greater force than was reasonably necessary for his self-protection and in doing so killed his assailant", the offence was manslaughter, not murder [14] .

The judgment of Taylor J. differs significantly from the other two judgments. For him the issue was whether what the accused did was done primarily for the purpose of defending himself against an agressor. Although the Court of Criminal Appeal of Trinidad and Tobago in Johnson v. The Queen [15] applied this test, they understood Taylor J. to be drawing a distinction between a primary intention of the accused to defend himself and a primary intention to kill the deceased or cause him grievous bodily harm. For my part, I should have thought that his Honour, in emphasizing that what the accused did must have been done primarily for the purpose of resisting the aggressor, was seeking to exclude those cases in which the accused was using the occasion as a pretext for killing or injuring the deceased. However, it is not a matter which merits further consideration because his Honour's view, though it has its attractions, did not commend itself to the other members of the Court and it did not form the basis of the decision in Reg. v. Howe.

Menzies J. was more explicit than Dixon C.J. in confining the qualified defence of manslaughter to cases of serious violence, whether felonious or not, or to cases of apprehended violence. Although Dixon C.J. mentioned fear of "violation or indecent or insulting usage", there is scarcely any justification for thinking that the Chief Justice contemplated that the doctrine would apply to trivial or minor assaults. His reference to the use by the accused of "no more force than was proportionate to the danger in which he stood" makes it reasonably clear, I should have thought, that his Honour was looking to situations of danger or apprehended danger, evidently including within those situations such cases of sexual molestation and insulting usage as might give rise to an occasion justifying resort to force in self-defence, not to minor or trivial assaults. It was scarcely to be expected that his Honour would express himself more precisely upon this point. The question whether the qualified defence of manslaughter is available in cases other than cases of serious violence is but a reflection of the questions, still unresolved, which arise in relation to the limits of the doctrine of self-defence itself.

The Victorian Supreme Court in Reg. v. Enright [16] and in Reg. v. Tikos [No. 2] [17] held that an offence which would otherwise amount to murder cannot be reduced to manslaughter under a plea of self-defence unless the occasion was one "which warranted the accused acting with intent to do some kind of grievous bodily harm at the least". On the assumption that this is a correct statement of the law, it has been suggested that if the doctrine is so confined there will be very few cases in which it is reasonable to believe that it is necessary to cause by way of self-defence grievous bodily harm to an assailant and yet it is unreasonable to believe that it is necessary to kill him. The suggestion proceeds on the view that the offence will be reduced from murder to manslaughter only if it appears that the accused's belief that his response was proportionate to the danger which threatened him was a reasonable belief. To my mind this is an incorrect interpretation of what Dixon C.J. said in Reg. v. Howe. A distinction is to be drawn between the accused's belief as to the danger which beset him and his perception of the proportionality of his response to that danger. For the offence to be reduced from murder to manslaughter it must appear that the accused reasonably believed in all the circumstances in which he found himself that an unlawful attack which threatened him with death or serious bodily injury was being or was about to be made upon him. But when it comes to the accused's belief as to the appropriateness of his response it is sufficient that he honestly believed that the force which he used was reasonably proportionate to the danger which he believed he faced. There is no additional requirement that his belief in this respect should be reasonably held or that it should be based on reasonable grounds. It is enough, as I have said, that the belief is held.

1. [1961] V.R., at pp. 668-669. 2. [1963] V.R., at p. 313.

The criticism therefore is misdirected. But even if it were not so, there are no means by which one can gauge the accuracy of the prediction that there will be very few cases in which it is reasonable to believe that it is necessary to cause by way of self-defence grievous bodily harm to an assailant and yet it is unreasonable to believe that it is necessary to kill him. It calls for a remarkable pre-vision of conceivable fact situations. For my part, I doubt whether the prediction is correct. There may well be situations in which it is reasonable for a person by way of self-defence to believe that it is necessary to inflict grievous bodily harm, to wound the assailant or to cause injury to his arm or leg for example, where it would be unreasonable to believe that it was necessary to kill him. Indeed, in New South Wales alone there have been cases in which the jury, following a Reg. v. Howe [18] direction, has returned a verdict of manslaughter - see "Self Defence, Provocation and Duress", a paper by Mr. Justice Lee delivered at the 19th Australian Legal Convention in Sydney, July 1977. But even assuming the prediction to be accurate, it concedes the existence of some cases in which a distinction will or may be drawn between the reasonableness of intending to inflict grievous bodily harm and the unreasonableness of intending death. The fact, assuming it to be a fact, that the application of the doctrine will result in the reduction of murder to manslaughter in a small number of cases only can be no reason of refusing to accept or apply the doctrine, if it be otherwise soundly based and if it produces an acceptable outcome. I do not agree that the doctrine introduces a test which is so complicated that it will confuse juries.

Some of the uncertainty that is said to surround the application of the principle enunciated in Reg. v. Howe is attributable not to the principle itself but to the generality of concepts which have been traditionally associated with the crime of murder, such as grievous bodily harm and violence, terms which have not proved to be susceptible of precise definition, and to some degree of uncertainty as to the precise limits of self-defence itself. Again, the problems that arise in connexion with the killing of a person in the course of resistance to unlawful arrest and in the course of apprehending a felon are but a reflection of the difficulties that have been experienced in bringing these situations within the common law conception of murder.

The fact that all these questions have not been resolved in Reg. v. Howe seems a slender reason for rejecting what it decides. If this reason is to serve as the standard by which existing decisions are to be evaluated, then the casualty rate will be extremely high. The truth of the matter is that the principle of Reg. v. Howe, if it continues to be accepted, will be refined, elaborated and developed in accordance with the tradition of the common law in future cases.

In this respect it depends on the clarification that has taken place, and will continue to take place, in relation to the mental element which underlies the defence of murder. In its gradual evolution we have seen emphasis given to the existence of an intent to kill or do grievous bodily harm and the confinement by statute of constructive malice to cases in which death has resulted during or after the commission of an act obviously dangerous to human life, or of an offence punishable by death or penal servitude for life. I refer, of course, to s. 18(1) of the Crimes Act, 1900 N.S.W., as amended, which also includes within its definition or description of "murder" that which was done or committed "with reckless indifference to human life". It goes on to provide that every punishable homicide not falling within the statutory definition "shall be taken to be manslaughter".

The provisos in sub-s. (2) are in the following terms:

(a) No act or omission which was not malicious, or for which the accused had lawful cause or excuse, shall be within this section.

(b) No punishment or forfeiture shall be incurred by any person who kills another by misfortune only, or in his own defence.

Proviso (b) was re-enacted in substantially similar terms in s. 10, 9 Geo. IV c. 31 (1828) which ended the distinction between justifiable and excusable homicide se defendendo (see H. A. R. Snelling, Killing in Self-Defence, Australian Law Journal, vol. 34 (1960), p. 135; Sir William Holdsworth, History of English Law, vol. 3, p. 312; vol. 13, pp. 400-401).

But this proviso does not touch the present problem which centres upon the inclusion in the statutory definition of the words "with intent to kill or inflict grievous bodily harm" in s. 18 (1) and proviso (a). In earlier times the element of malice aforethought was supplied by the mere existence of an intent to kill or inflict grievous bodily harm except in those cases in which the formation of the intention was held to be excusable or justifiable, e.g. self-defence. At one time it was thought that the existence of provocation negatived the formation of an intention to kill or inflict such harm: see Holmes v. Director of Public Prosecutions [19] , per Viscount Simon. Now it is accepted that "The defence of provocation may arise where a person does intend to kill or inflict grievous bodily harm but his intention to do so arises from sudden passion involving loss of self-control by reason of provocation" (Attorney-General (Ceylon) v. Perera [20] ; Lee Chun-Chuen v. The Queen [21] ). Consequently, in the case of provocation, the intention to kill or inflict grievous bodily harm which, but for the extenuating circumstances in which it originates would have the quality of malice aforethought, lacks that quality, and the offence sinks to the level of manslaughter. Now that it has been acknowledged that provocation does not deny the existence of such an intention, no insurmountable barrier remains in the way of reaching the conclusion that circumstances giving rise to an occasion of self-defence also deprive an intention to kill or inflict grievous bodily harm formed in consequence thereof of the quality of malice aforethought. Then, if the response is not excessive, the accused commits no offence; if it is excessive, he is guilty of manslaughter. This outcome fits the modern conception of manslaughter, at least the voluntary aspect of manslaughter which embraces a voluntary killing which might have constituted murder but for the existence of a mitigating circumstance. Here that circumstance is to be found in the occasion which entitled the accused to resort to force in his own defence, the offence consisting in causing death by resorting to an excessive degree of force. There is a distinction between (a) the person who kills in self-defence; (b) the person who kills through resorting to excessive force for his own protection; and (c) the person who does not kill at all. The store which we set by the preservation of human life requires that the act of killing through an excessive degree of force, though in self-protection, should be adjudged unlawful and that the offence is of a lesser order than murder.

[1946] A.C. 588, at p. 598. 2. [1953] A.C. 200, at p. 206. 3. [1963] A.C. 220, at p. 228.

What I have already said indicates why I do not accept that the law is correctly stated in Palmer v. The Queen [22] .

In conclusion I state what Stephen J., Aickin J. and I consider to be the issues which arise for determination by the jury according to Reg. v. Howe. In so doing I put to one side cases involving threatened violation of or indecent or insulting usage to the accused's person. Accordingly where threat of death or grievous bodily harm to the accused is in question and the issue of self-defence arises the task of the jury must be stated as follows:

(a) It is for the jury first to consider whether when the accused killed the deceased the accused reasonably believed that an unlawful attack which threatened him with death or serious bodily harm was being or was about to be made upon him.

(b) By the expression "reasonably believed" is meant, not what a reasonable man would have believed, but what the accused himself might reasonably believe in all the circumstances in which he found himself.

If the jury is satisfied beyond reasonable doubt that there was no reasonable belief by the accused of such an attack no question of self-defence arises.

If the jury is not satisfied beyond reasonable doubt that there was no such reasonable belief by the accused, it must then consider whether the force in fact used by the accused was reasonably proportionate to the danger which he believed he faced.

If the jury is not satisfied beyond reasonable doubt that more force was used than was reasonably proportionate it should acquit.

If the jury is satisfied beyond reasonable doubt that more force was used, then its verdict should be either manslaughter or murder, that depending upon the answer to the final question for the jury - did the accused believe that the force which he used was reasonably proportionate to the danger which he believed he faced?

If the jury is satisfied beyond reasonable doubt that the accused did not have such a belief the verdict will be murder. If it is not satisfied beyond reasonable doubt that the accused did not have that belief the verdict will be manslaughter.

In my opinion special leave to appeal should be granted. I would allow the appeal and order a new trial. I agree that the jury were not correctly directed on the significance of the drugged condition of the appellant in its relation to his possession of the intent necessary for the killing to constitute murder. I cannot usefully add anything to the reasons which have been expressed by Gibbs J. for this conclusion.

Having determined that special leave should be granted, it is necessary to deal with the ground of appeal that the jury was not correctly directed on the defence of self-defence. In the circumstances no question arises whether or not the New South Wales Court of Criminal Appeal should have applied or whether this Court should apply the proviso to s. 6 (1) of the Criminal Appeal Act, 1912 upon the ground that no substantial miscarriage of justice occurred as a result of the misdirection on self-defence. I think it most unlikely that the jury would have found in favour of the appellant on this defence, so unlikely that in other circumstances an appeal court could possibly have applied the proviso. But as there must be a new trial on the first ground that is not now to the point.

The appellant raised the matter of self-defence and the trial judge was correct in leaving the question to the jury. There is no nice line which can be drawn between cases where such a defence should or should not be left. However, once the defence is squarely raised, as it was in this case, and where there is evidence of a struggle, I can hardly conceive of circumstances where the facts and the inferences to be drawn from them are so clear that a trial judge would be justified in directing the jury that on no conceivable view of the facts could they not be satisfied beyond reasonable doubt that the Crown had proved that the killing was not done in self-defence. A jury can, indeed must, be trusted to give the matters raised the weight which they deserve and to determine the facts, including questions of degree, and to reach ultimate conclusions on those facts in accordance with the directions of law given to them. A particular conclusion might seem unreasonable to the trial judge and he can express his own view on that conclusion with adequate warning on his and the jury's respective functions; but he cannot direct the jury in law that the conclusion of fact is unreasonable. I accept that it is the law that a felonious aggressor must positively terminate his aggression and thereafter be himself the subject of aggression by his victim before his subsequent act can be regarded as his self-defence against his victim's aggression. But, the law having been stated, it is not possible to be equally definitive about the facts. It is for the jury to determine whether an accused terminated his felonious aggression and whether he was thereafter the subject of aggression by his former victim and so on through the elements of the defence. And it was for the jury so to do in this case after proper direction.

The aspect of the direction which is claimed to be defective is the failure to direct the jury in relation to a possible verdict of manslaughter in accordance with the decision of this Court in Reg. v. Howe [23] . The learned trial judge did not direct the jury that in the event that they found the facts to be such as were the subject of the decision in that case they could find the appellant not guilty of murder but guilty of manslaughter. It would appear that he may have been under the impression that he had done so. The summing-up on self-defence was as follows:

You remember yesterday his counsel put to you that you ought to find him not guilty of anything because he said everything he did by way of inflicting harm on Rellis and causing his death was done in self-defence.

It is the law that any person who is attacked with violence may defend himself. If in the course of defending himself against an attack, the man who is his aggressor is killed, then it will be a question for you to determine whether or not what he did was done primarily for that purpose of defending himself against the aggressor and you would have to be satisfied beyond a reasonable doubt that this was not so before you could return a verdict of murder. It is called self-defence but it is not something that the accused has to raise. There has to be material before you on which you are able to say that the person who did the killing did it in self-defence, that is did it in the course primarily of resisting the attacks of an aggressor.

Of course the learned Crown Prosecutor has put to you that it is impertinent - I do not know whether he used that word - because to put this is to reverse the realities. Who was the aggressor in this case? Beyond doubt the Crown would have you say it was Viro and Greco who planned to be the aggressors. They pleaded guilty before you to agreeing they would rob with violence and they set about doing it and no doubt as I understand the evidence it is not disputed that initially they were.

Once they began attacking this man Rellis, about whom a great deal has been said - but of course who is not here, he is dead - he was entitled to resist. He was entitled to have his life and to fight for it, and to fight for it with all the ferocity that he could command, if it was threatened by two men attacking him with a knife. No question of self-defence can arise in this case unless you take the view that on the facts the man who was killed, was attacking these two men, that he was the aggressor, because the thing explains itself. It is only if you are attacked that you are entitled to defend yourself. If you are the attacker you cannot complain if your victim has the temerity to defend himself and if he bests you in a fight it is just too bad. So you see it never arises unless there is material upon which you can say at the point of time this man was stabbed, Rellis was the aggressor. He was attacking them. They had ceased all attack on him and that is the way it is put.

It is put by the learned counsel for Viro, having hit him on the head with the jack handle, he then ceased to have it in his possession. He had not taken up the knife and it was then that Rellis produced the knife and then Viro and Greco were forced by the attacker to defend themselves and the roles were reversed. The realities were restored and the truth of the matter was that they were the victims of a vicious attack by this man Rellis with a deadly weapon.

Gentlemen, this is your part of the case. How did this vicious attack finish up? It finished up with Viro with cut knuckles and with Greco with the cut in his arm that you have seen. I repeat it is not for the accused to make out they acted in self-defence. It is for the Crown to show that they did not, if you are satisfied that there is material upon which it can be said the accused were attacked.

At the conclusion of the summing up the following interchange occurred between the appellant's counsel and the learned trial judge: Mr. Lloyd-Jones: Would your Honour direct the jury that if the accused Viro used excessive self-defence then manslaughter is available.

His Honour: No, I will not put that. Excessive self-defence?

His Honour: I have put it higher. I have put it that unless they are satisfied they should find him not guilty of murder but guilty of manslaughter.

But a direction on manslaughter in relation to the defence of self-defence had not been given and if it ought to have been given the direction was defective. 1. (1958) 100 C.L.R. 448.

That brings me to the question whether a direction which would leave open a verdict of manslaughter ought to have been given. Before considering whether Howe was correct on this point and whether this Court should for that reason follow its earlier decision, there are the questions whether it is free to follow this earlier decision and whether, even if it is free to do so, it ought to follow it in view of the decision of the Privy Council in Palmer v. The Queen [24] . That was an appeal to the Queen in Council from the Court of Appeal of Jamaica. At the outset I should state that I cannot accept as significant that the appeal was from Jamaica and not from an Australian court. The law discussed and decided was the common law and their Lordships did not intend to distinguish the common law of Jamaica from that of the United Kingdom or of Australia on such a fundamental question as the verdicts open to a jury on the raising of a defence of self-defence to murder.

This Court must consider whether it is now strictly bound by the decision of the Judicial Committee of the Privy Council according to the principles of the common law governing precedent. It needs to be emphasized that it is the strict law on precedent which is now being considered and nothing less. It is obvious to me that this Court would only differ from a decision of the Privy Council with the greatest reluctance even if it is now free to do so. Likewise and with the same reluctance would it decline to adopt strong dicta immediately applicable. Tradition and a desire to ensure uniformity of the common law within the whole British Commonwealth so dictates. But is this Court strictly bound? I do not think that it can be. Since the passing of the Privy Council (Appeals from the High Court) Act 1975 it is clear that there are now two co-ordinate tribunals to which an appellant from the Supreme Court of a State (not exercising federal jurisdiction) can appeal. So far as I am aware, this is a unique position. The law of precedent depends upon the existence of a hierarchy of courts and now there is no longer a hierarchy. Therefore the strict law on precedent cannot be applied. Therefore strictly the decisions of the Privy Council are no longer binding on this Court. It is therefore not bound to follow Palmer.

Before I proceed to consider whether this Court ought to follow Palmer in preference to its earlier decision in Howe, I would say something of the position of State courts. They were always bound to follow a decision of the High Court in preference to dicta or reasoning, however strong, in the reasons given by the Privy Council (Jacob v. Utah Construction and Engineering Pty. Ltd. [25] ). Likewise they were bound to follow a decision of the Privy Council in preference to dicta in this Court. Before the 1975 Act they, as was this Court, were bound to follow a decision of the Privy Council in preference to a conflicting decision of this Court. But since the 1975 Act, they ought in my view to follow a decision of this Court in preference to a conflicting decision of the Privy Council. The High Court is the court of appeal from the Supreme Courts of the States (Constitution, s. 73 (ii.)). There is no right of appeal conferred in the Constitution from the Supreme Court of a State to the Privy Council. The Constitution being silent on the prerogative to receive appeals from the Supreme Courts of the States it has throughout the years been accepted that there remained intact the Royal Prerogative to allow an appeal to the Council from the Supreme Court of a State. The Judicial Committee Act, 1833 Imp. could therefore apply. The Royal Prerogative to grant special leave to appeal to the Privy Council from this Court was preserved with one exception in s. 74. The Privy Council therefore remained in all matters (subject to s. 74) the appellate tribunal from this Court. That is no longer so. Now in all matters the High Court is, without the Privy Council as the appellate tribunal from it, the court of appeal in the hierarchy of precedent. The express constitutional provision governing Australia in particular prevails over the generality of the 1833 Act and of the Prerogative power.

I return to the particular question which now falls to be determined. Although this Court is not bound to follow Palmer [26] it will anxiously examine the question raised and will lean towards doing so even though it thereby would decline to follow its previous decision in Howe [27] . However, it is bound to reach the conclusion which it believes correctly to represent the law. It cannot seek comfort in simply conforming with a binding precedent.

The Privy Council in Palmer discussed Howe at some length and having examined the reasoning of the various members of the High Court concluded that the principles there enunciated were not correct. It is a fair summary of their conclusion that they rejected the conclusion of Dixon C.J. in Howe [28] :

Had he used no more force than was proportionate to the danger in which he stood, or reasonably supposed he stood, although he thereby caused the death of his assailant he would not have been guilty either of murder or manslaughter. But assuming that he was not entitled to a complete defence to a charge of murder, for the reason only that the force or violence which he used against his assailant or apprehended assailant went beyond what was needed for his protection or what the circumstances could cause him reasonably to believe to be necessary for his protection, of what crime does he stand guilty? Is the consequence of the failure of his plea of self-defence on that ground that he is guilty of murder or does it operate to reduce the homicide to manslaughter?

There is no clear and definite judicial decision providing an answer to this question but it seems reasonable in principle to regard such a homicide as reduced to manslaughter, and that view has the support of not a few judicial statements to be found in the reports.

In Palmer, immediately before setting out this passage from the judgment of Dixon C.J., their Lordships stated the effect of the decision in Howe as they saw it [29] :

It will thus be seen that the Full Court of South Australia had posed two questions as being the relevant questions where a plea of self-defence would succeed in toto but for the use of excessive force by the person attacked:

1. Was more force used than a reasonable man would consider necessary?

2. If so, did the accused nevertheless honestly believe that such excessive force was necessary?

and both questions would have to be answered in the affirmative to justify a verdict of manslaughter.

Three members of the High Court of Australia (Dixon C.J., McTiernan J. and Fullagar J.) agreed in substance with the Full Court.

In common with their Lordships in Palmer I do not understand the majority in Howe simply to have held that where more force is used than is reasonably necessary in self-defence the offence of murder is reduced to manslaughter. It was held that where more force is used than is reasonably necessary and where the accused in fact believes that the amount of force which he uses is necessary then the crime of murder will be reduced to manslaughter. It presupposes that, although the accused believes that the amount of force used is necessary, his belief is not a belief which a reasonable man in the circumstances would hold.

The question whether or not Howe is correct appears to me on a final analysis to depend on whether there is any objective constituent in the defence, that is to say, whether there is a constituent of the defence that the amount of force used was no more than was objectively reasonable, no more than a reasonable man would have used or would have believed to be necessary in the circumstances either as they would have appeared to such a reasonable man in the situation of the accused or, on an alternative formulation, as the accused in fact thought them to be. Either of these formulations introduces, the one to a greater and the second to a lesser extent, a truly objective element of reasonableness.

If there be such an objective element in the defence, then an absence of that element and a consequent failure of the defence of self-defence should not result in the accused being guilty of murder. It is the common understanding of modern times, and the jury is repeatedly told, that malice aforethought (or some statutory equivalent) is the requirement of murder. Murder is the most foul killing because of the requisite malice. There is no objective element in the crime. Therefore if there be an objective element of reasonableness in the "defence" of self-defence a failure to find that objective element would prevent the killing being murder. The fact that a man believed that he was killing in self-defence would deprive the act of the requisite malice. Yet the killing would be unlawful and it would be proper that he be found guilty of manslaughter.

On the other hand, if upon analysis it be found that there be no truly objective element in the defence, the conclusion should in my opinion be different. If the correct question be whether the accused believed that the amount of force which he used was necessary a negative answer to that question has the consequence that the force actually used by the accused cannot be regarded as having been used for the purpose of defending himself. The present intent to kill or do grievous bodily harm, the existence of which is presupposed by the subject matter under consideration, stands unaccompanied by a belief that it was necessary to use that degree of force and provides sufficient base for the conclusion that the killing amounted to murder. There is no place for a verdict of manslaughter, if this analysis be correct.

But this is not the usually suggested alternative because it places no limit of any kind on the quality of the accused's belief that the degree of force used by him was necessary. That is why the qualification is commonly introduced that the accused's belief must be based "on reasonable grounds" or words to the same effect. It appears to me that if the effect of the introduction of these words is to require that the accused's belief be that which a reasonable man would have held in the circumstances (whether as they appeared to the accused or as they would have appeared to a reasonable man in the position of the accused) then we are back to the same position as on the first analysis. A jury would be required as reasonable persons to place themselves in the position of the accused and to ask themselves what they as reasonable persons would have believed. If that be so, then a conclusion by them that they would have believed differently from the accused should not, at least in the modern development of the law of murder, result in a conviction for that crime. Yet, I repeat, the killing would be unlawful and the proper verdict would be manslaughter.

But these two alternatives which I have been discussing do not fully exhaust the possibilities. The words "belief on reasonable grounds" may refer not to a belief which a reasonable man would hold but to a belief so wholly unreasonable that no reasonable man could possibly hold it. Should the mere fact of an accused's belief be sufficient in such a case to acquit him of murder? There is a considerable difference between what a reasonable man would believe and what a reasonable man could believe. When we speak of what a reasonable man would believe, we are speaking of that reasonable man most familiar in the law of negligence. He is the type. He would do one thing and omit to do another thing. He would draw one inference, reach one belief. When we move from the conduct of the reasonable man so well known in the law of negligence, civil and criminal, to reasonableness of belief or cause or the like in the case of other torts and crimes, we move to different and often more uncertain ground. I would for example refer to the difficulties which have arisen in regard to the formulation that a comment to be fair must be capable of being reasonably inferred from truly stated facts (and the judge must so find) but the question for the jury is not whether they as reasonable persons would draw the inference but whether the inference was fairly and honestly drawn. See Gatley , 7th ed. (1974), par. 725 et seq. And in crime there is the vexed question whether a "defence" that there was no mens rea based on a mistake of fact requires that the mistake be one based on reasonable grounds in an objective sense. The cases and the learned writing on this subject (and on the associated subject of the defence of mistake in the case of statutory crimes which do not require mens rea) have been collected by McInerney J. in his judgment in Handmer v. Taylor [30] . The great difficulty of introducing an objective test of reasonableness into any part of the concept of mens rea is that crimes other than crimes of negligence can as a consequence be committed by a neglect to observe the standards of conduct of a reasonable man. As Professor Glanville Williams has observed in his Criminal Law - General Part, 2nd ed. (1961), p. 204 when discussing the bigamy cases, the view that unreasonable mistake is no defence to bigamy means that bigamy can be committed negligently. And I would refer without elaboration to the doubts which had arisen on the question whether there is need for reasonableness in an objective sense in an accused's belief as to consent in rape cases. I need do no more than refer to Reg. v. Morgan [31] where a majority in the House of Lords held that reasonableness of belief in a fully objective sense was not a necessary quality of the consent.

1. [1971] V.R. 308, at pp. 311-315. 2. [1976] A.C. 182.

I am of the opinion that in many contexts the concept of reasonableness does not introduce the test of what the reasonable man would think or believe or whatever it may be in the particular case but what a reasonable man, a rational man, could or might think or believe or whatever in such a case. These concepts, although it may be said that they each contain some objective element, are really quite different. The one introduces a wholly objective test. The other simply places the subjective belief of the accused in a sane rational context. Reasonable grounds for a belief are then rational grounds for that belief, grounds consonant with reason. An irrational belief is not allowable as the basis of a defence because it would in effect be a defence of insanity and such a defence has come to be governed by strict rules. The requirements for a defence of insanity cannot be avoided by pleading an irrational belief. But now that the word "reasonable" has very largely acquired objective connotations care must be taken not to import that wholly objective connotation into contexts where such a test is inappropriate.

An alternative formulation is that the question of reasonableness goes only to the question whether the belief was in fact held by the accused. See Glanville Williams, op. cit., pp. 202-203 and the passages from the cases to which he refers. See also Reg. v. Morgan [32] . The difficulty which I feel with this formulation is that it allows the possibility of a senseless irrational belief providing a defence in the absence of a plea of insanity. But speaking generally there is little difference in practical result.

In my opinion it is a constituent of the defence of self-defence that an accused use only that amount of force which he believes to be necessary to repel the attack on him. His belief must be based on reasonable grounds in the sense that the belief must be one which a rational man could hold, but not necessarily a belief which an ordinary reasonable man would hold. Applying the analysis which I have attempted above I therefore conclude that there is no place for a verdict of manslaughter where acting on an irrational belief that such force was necessary an accused has used excessive force to defend himself.

In their Lordships' view the defence of self-defence is one which can be and will be readily understood by any jury. It is a straight-forward conception. It involves no abstruse legal thought. It requires no set words by way of explanation. No formula need be employed in reference to it. Only common sense is needed for its understanding. It is both good law and good sense that a man who is attacked may defend himself. It is both good law and good sense that he may do, but may only do, what is reasonably necessary. But everything will depend upon the particular facts and circumstances. Of these a jury can decide. It may in some cases be only sensible and clearly possible to take some simple avoiding action. Some attacks may be serious and dangerous. Others may not be. If there is some relatively minor attack it would not be common sense to permit some action of retaliation which was wholly out of proportion to the necessities of the situation. If an attack is serious so that it puts someone in immediate peril then immediate defensive action may be necessary. If the moment is one of crisis for someone in imminent danger he may have to avert the danger by some instant reaction. If the attack is all over and no sort of peril remains then the employment of force may be by way of revenge or punishment or by way of paying off an old score or may be pure aggression. There may no longer be any link with a necessity of defence. Of all these matters the good sense of a jury will be the arbiter. There are no prescribed words which must be employed in or adopted in a summing up. All that is needed is a clear exposition, in relation to the particular facts of the case, of the conception of necessary self-defence. If there has been no attack then clearly there will have been no need for defence. If there has been attack so that defence is reasonably necessary it will be recognised that a person defending himself cannot weigh to a nicety the exact measure of his necessary defensive action. If a jury thought that in a moment of unexpected anguish a person attacked had only done what he honestly and instinctively thought was necessary that would be most potent evidence that only reasonable defensive action had been taken. A jury will be told that the defence of self-defence, where the evidence makes its raising possible, will only fail if the prosecution show beyond doubt that what the accused did was not by way of self-defence. But their Lordships consider, in agreement with the approach in the De Freitas Case [34] , that if the prosecution have shown that what was done was not done in self-defence then that issue is eliminated from the case. If the jury consider that an accused acted in self-defence or if the jury are in doubt as to this then they will acquit. The defence of self-defence either succeeds so as to result in an acquittal or it is disproved in which case as a defence it is rejected. In a homicide case the circumstances may be such that it will become an issue as to whether there was provocation so that the verdict might be one of manslaughter. Any other possible issues will remain. If in any case the view is possible that the intent necessary to constitute the crime of murder was lacking then that matter would be left to the jury. 1. [1974] A.C., at pp. 831-832. 2. (1960) 2 W.I.R. 523.

I cannot conclude from this passage that their Lordships were not referring to an objective test of reasonableness. If I am correct in that conclusion then with the greatest respect I would express the opinion that a verdict of manslaughter would be open to a jury if they found that the accused believed that he used no more force than was necessary but that in fact the accused did more than was reasonably necessary. I shall not repeat my reasons for this opinion. On the other hand, if I am wrong in my understanding of the passage and it was not intended to introduce an objective criterion of reasonableness as that criterion is used for instance in the law of negligence, I respectfully agree that the possibility of a verdict of manslaughter need not be considered in relation to the defence of self-defence.

It is of little use attempting definitively to state which interpretation of the passage is correct. In the particular circumstances in which this Court now finds itself, I think it best to state what I understand to be the applicable principle. In my opinion the question for the jury is whether the accused although he had the intention to kill or to do grievous bodily harm acted as he did with the purpose of defending himself and in the belief that the infliction of death or the grievous bodily harm inflicted by him was necessary in order to defend himself. If the facts of the case leave open the view that there were no rational grounds for the belief which the accused had, then, and only then, should the jury be told that the belief of the accused must be a rational one. Then, it seems to me, it would be desirable to make it clear that the belief must be one which a rational man might or could have held, not necessarily the belief which the jury as reasonable persons think that they would have held.

I say nothing upon the question whether a person who intends by his defensive act to inflict death or grievous bodily harm is only justified in using that degree of force if the attack made on him places him or is believed by him to place him in danger of a like injury. That question does not arise in the present case.

I return now to the terms of the summing up in the instant case. In my opinion the learned trial judge was correct in omitting all reference to reasonable grounds for the applicant's belief in the necessity of his action. If the applicant's account of events were accepted, and if the jury had concluded that the applicant was defending himself in a way which he believed to be necessary in order to defend himself against the knife with which the deceased was attacking him, no question of the reasonableness of the applicant's belief in that necessity, in the sense which I have explained, would in the circumstances arise.

I have now had the advantage of reading the reasons for judgment prepared by the other members of the Court. From them it appears that there is a clear majority in favour of the view that an objective element of reasonableness in the degree of force used is a constituent element of the defence of self-defence. In the circumstances I feel that I should accept that view. For the reasons which I have already expressed, I therefore conclude that in a case where the defence fails only because the amount of force used was more than was reasonably necessary or more than the accused should reasonably have believed to be necessary, the proper verdict is manslaughter, not murder.

This appeal was reargued to dispose of the question whether the High Court is bound by the Privy Council's decisions. I will deal with this question before turning to the issues of the appeal itself.

The High Court and the Privy Council. The Judicial Committee of the Privy Council, formerly the Committee for the Business of Trade concerned with the supervision of affairs in the plantations and colonies, evolved to become the judicial arm of British imperialism. It performed useful tasks, particularly in criminal law, such as correcting judicial denials of justice and procedural irregularities where there was no adequate local appeal machinery, but until recently paid little attention (see Geelong Harbor Trust Commissioners v. Gibbs Bright & Co. [35] ) to the natural tendency of the components of the British Empire, British Commonwealth and Commonwealth of Nations to develop systems of law attuned to their people's needs. The political, social and economic advantages to the imperial authorities of preserving harmony between the colonial and United Kingdom legal systems were obvious, and the Privy Council directed colonial and dominion courts to follow decisions of English courts rather than their own (see Trimble v. Hill [36] ; Robins v. National Trust Co. Ltd. [37] ; Bakhshuwen v. Bakhshuwen [38] ). Australian courts readily acquiesced, perhaps even going further than was expected of them (see Piro v. W. Foster & Co. Ltd. [39] ; Waghorn v. Waghorn [40] ). The High Court's acquiescence did not cease until Parker v. The Queen [41] .

1. (1974) 122 C.L.R. 576; [1974] A.C. 810. 2. (1879) 5 App. Cas. 342. 3. [1927] A.C. 515. 4. [1952] A.C. 1. 5. (1943) 68 C.L.R. 313. 6. (1942) 65 C.L.R. 289. 7. (1963) 111 C.L.R. 610.

After the loss of the American colonies, the United Kingdom's response to the Australian colonists' demands to make their own laws was to pass Acts which granted legislative capacity to colonial legislative councils and assemblies. Colonial legislation was, of course, interpreted by courts which were ultimately under the Privy Council's control. Although the judicial interpretations were generally subject to correction by local legislatures, the Privy Council imposed constitutional limits on the legislatures (for example, on the power to make extra-territorial laws). Other restrictions were imposed through the Colonial Laws Validity Act, 1865 Imp.. More importantly, the great body of common or decisional law was under the Privy Council's control. The appeal (as of right or by leave) to the Privy Council thus enabled the United Kingdom to control the interpretation of colonial legislation and the development of decisional law.

Consequently, a significant part of Australian law has been made in the United Kingdom rather than in Australia and, inevitably, much of this law was evolved to meet changing conditions in the United Kingdom and to facilitate its emergence as a world power. During the nineteenth century, the dominant theme was the evolution of rules to serve the needs of British commerce and industry, particularly the factory system. In tort law, it was the century of the defendant, with the exaltation of doctrines under which the injured person would not recover if there were contributory negligence, if the injury was caused by a fellow employee, or if the injury was the death of a person. These doctrines, now regarded as pernicious, were faithfully followed in Australia until they were superseded, often after lengthy periods, by legislation. On the other hand, the judicial modernization of the mercantile law assisted greatly in the expansion of trade throughout the Empire and in making London the commercial capital of the world.

Before Federation, the founding fathers realized the disadvantages of having an ultimate tribunal whose members were not acquainted with the circumstances under which Australian laws were passed and applied, and wanted to replace the Privy Council with a federal Supreme Court to be known as the High Court of Australia. (The history of this is referred to in Baxter v. Commissioners of Taxation (N.S.W.) [42] and La Nauze, The Making of the Australian Constitution (1972), Ch. 16.) Several who later became Justices of this Court (Edmund Barton, Australia's first Prime Minister, Isaac Isaacs who also became Chief Justice and later Governor-General, and Senator O'Connor) made no secret in the Debates on the Bill to establish the High Court (Judiciary Bill, 1903) of their lack of confidence in the Privy Council's handling of Australian cases.

Barton explained that the appeal to the Privy Council in its present form in s. 74 of the Constitution was forced upon the Australian people by the British authorities and was accepted by the Australian negotiators "only as the price that had to be paid to prevent more drastic amendments of the Constitution" which had been agreed to by the people (Commonwealth Parliamentary Debates, vol. 13, p. 803). He then said: "If I had my own way I would have no appeals to the Privy Council" (p. 815). Senator O'Connor described the Privy Council as "altogether an unsuitable body to interpret our Constitution" and as "a most unsatisfactory tribunal" (vol. 15, p. 2699). He stated that it "would be very likely in the future to commit grave errors, as it has already committed them in the past". Isaac Isaacs said, "I cannot feel the same confidence in the Privy Council that I should have in an Australian court, in regard to Australian matters" (vol. 13, p. 731) and "I do not hesitate to say that the Privy Council is not a court in which we can place the fullest reliance in regard to the interpretation of our laws" (p. 732). Referring to the decisions of the proposed High Court, he said, "Those decisions are much more likely to be correct than would be the case if they were given by gentlemen unacquainted with our conditions, and in most cases, therefore, as unable to interpret the meaning of our statutes as if they were living in the planet Mars". In expressing a similar view, Deakin, Australia's first Attorney-General and later Prime Minister, referred to the famous Protest of the New Zealand Bench and Bar against the Privy Council after the decision in Wallis v. Solicitor-General [43] when Stout C.J. described the Privy Council decision as a cardinal blunder, made in ignorance of New Zealand laws and history and listed a series of other blunders they had made (see Appendix to New Zealand Privy Council Cases (1903), p. 730). Deakin then referred to other Privy Council errors, instancing Attorney-General (N.S.W.) v. Macleod [44] as a case where the Privy Council had allowed an appeal from the New South Wales Supreme Court through ignorance of a well-known New South Wales Act (Commonwealth Parliamentary Debates, vol. 13, p. 595).

1. [1903] A.C. 173; [1903] N.Z.P.C.C. 23. 2. [1891] A.C. 455.

The Privy Council's advice in the criminal case of Oteri v. The Queen [45] (in which it gave special leave to appeal from the Supreme Court of Western Australia) is a recent example of what the founding fathers foresaw. The advice contains grave errors on most important questions of Australia's constitutional law, on which it had no right to pronounce. The Privy Council (Lord Diplock, Viscount Dilhorne, Lord Hailsham of St. Marylebone, Lord Edmund-Davies and Lord Fraser of Tullybelton) said [46] : "The legislative power of the Commonwealth of Australia does not extend to criminal law. That lies within the competence of the States."

1. [1976] 1 W.L.R. 1272; (1976) 51 A.L.J.R. 122. 2. [1976] 1 W.L.R., at p. 1275; (1976) 51 A.L.J.R., at p. 124.

This statement is quite wrong. The Commonwealth has full power (which it has freely exercised) to make criminal as well as non-criminal law with respect to the subjects on which it is empowered to legislate. Every year thousands of persons are convicted of offences against the Crimes Act 1914 Cth and the numerous other criminal laws of the Commonwealth. Section 80 of the Constitution (which provides for jury trial of indictments for offences against any law of the Commonwealth) and s. 120 (which provides for imprisonment of offenders against the laws of the Commonwealth) should have deterred the Privy Council from this error. Its statement cannot be excused as being correct in its application to the circumstances of the case (off-shore crimes) though expressed too generally. Even if limited to off-shore crimes, it was wrong. The external affairs power (s. 51 (xxix.) of the Constitution) is sufficient to give the Australian Parliament competence to provide for off-shore crimes (see The Seas and Submerged Lands Case [47] ; Pearce v. Florenca [48] ).

1. (1975) 135 C.L.R. 337. 2. (1976) 135 C.L.R. 507.

However the Privy Council came to this view, it had no right to express it. It was a pronouncement on a question of constitutional relations between the Commonwealth and the States which the Privy Council and at that time State Supreme Courts were not entitled to decide. Such inter se questions were reserved for the High Court and any arising in a State Supreme Court would automatically be removed to this Court under s. 40A of the Judiciary Act 1903-1973 which was inserted (by Act No. 8 of 1907) to prevent inter se questions reaching the Privy Council. This has now been modified to provide for removal on the application of an Attorney-General of the Commonwealth or a State (see s. 9 of the Judiciary Amendment Act 1976).

Section 74 of the Constitution prohibits any appeal to the Privy Council from "a decision of the High Court upon any question, howsoever arising, as to the limits inter se of the constitutional powers of the Commonwealth and the States" unless the High Court certifies that the question is one which should be so determined. Since Federation, the Privy Council has had no right to pronounce on an inter se question whether on appeal from the High Court (unless there is a certificate under s. 74) or from any other court (see Western Australia v. Hamersley Iron Pty. Ltd. [No. 2] [49] ). The Privy Council seemed to have accepted this view in Dennis Hotels Pty. Ltd. v. Victoria where it said [50] :

It has always been recognized that the general purpose of the agreement enacted in s. 74 is to reserve for the final decision of the High Court in Australia, unless the Court itself wishes to refer the matter to the Judicial Committee, "questions which arise in connexion with the federal distribution of power between the Commonwealth on the one hand and the States on the other" (Quck & Garran, The Annotated Constitution of the Australian Commonwealth (1901), p. 757). The clear intention of s. 74, as was said by Isaacs, J. in Pirrie v. McFarlane [51] is that "on the purely Australian question of the distribution of the totality of governmental powers on this continent, the High Court of Australia - the highest judicial organ created by the Australian people - was to be the final arbiter, unless it voluntarily requested the intervention of the Sovereign in Council". 1. (1969) 120 C.L.R. 74. 2. (1961) 104 C.L.R. 621, at pp. 625-626; [1962] A.C. 25, at pp. 50-51. 3. (1925) 36 C.L.R. 170, at p. 196.

Despite this, the decision in Oteri [52] shows that the Privy Council does not hesitate to intrude into Australian constitutional affairs, even though it has no jurisdiction.

Soon after Federation, because of the tension between the Privy Council and the High Court, Parliament acted to limit appeals to the Privy Council. It adopted suggestions made by some members of the High Court (Griffith C.J. and O'Connor and Isaacs JJ.) in Flint v. Webb [53] for legislation to minimize the Privy Council's interference in Australian constitutional affairs. (See the amendments to the Judiciary Act 1903 (made by No. 8 of 1907)).

The Privy Council (Limitation of Appeals) Act 1968, which came into operation on 1st September 1968, allowed appeals from the High Court to the Privy Council only where the High Court decision was on appeal from a decision of a State Supreme Court (not exercising federal jurisdiction) and then only if the High Court decision did not involve the application or interpretation of the Constitution or of a law (or of an instrument made under a law) made by Parliament. It also excluded the possibility of appeals from other federal courts and territorial Supreme Courts. The Act is very restrictive of appeals, as decisions of this Court invariably involve the application of the Constitution, a law made by Parliament (e.g., the Judiciary Act, s. 37), or an instrument made under such a law (e.g., High Court Rules). The application is generally so familiar that it is not referred to in reasons for decision. The limitation did not apply in respect of decisions of the High Court given in a proceeding that was commenced in a court before the commencement of the Act.

The Privy Council (Appeals from the High Court) Act 1975 came into operation on 8th July 1975 and entirely removed the Privy Council's jurisdiction to give special leave to appeal from the High Court after that date (apart from cases commenced in a court before the Act came into operation). Apart from any residual matters, there is now no appeal to the Privy Council from the High Court.

In The Commonwealth v. Queensland [1] , I pointed out the consequences which these two limitation of appeals Acts had on appeals from the State courts. The scheme of the Constitution (embodied in s. 74) is that the High Court is to be the final arbiter on inter se questions (in the absence of a certificate) and in all matters in respect of which Parliament has, by a limitation Act, excluded an appeal by special leave to the Privy Council. If the exclusion by s. 74 of appeals from the High Court to the Privy Council on inter se questions means that no such question is to reach the Privy Council by way of appeal from any other court, then, by the same reasoning, when Parliament excluded appeals from the High Court to the Privy Council in other matters (exercising the power under s. 74 by the limitation Acts of 1968 and 1975), no appeal in those matters lies to the Privy Council (as of right or by leave) from any other court. As Parliament has eliminated all appeals from the High Court to the Privy Council (apart from residual cases commenced in a court before the operation of the Acts), all appeals to the Privy Council from Australian courts are ended (apart from residual cases). I disregard the provision for appeal upon a certificate by the High Court under s. 74. This is now a dead letter, and, in any event, its theoretical operation may have ended with the 1968 Act. Even if s. 74 did not operate in these circumstances to end appeals to the Privy Council, the continuance of such appeals is quite inconsistent with present constitutional relationships between Australia and the United Kingdom.

If the appeal to the Privy Council (by right or by leave) continues, the absurdities and mischief which we face are obvious. The tension between the Privy Council and the High Court which occurred over inter se questions (see Baxter v. Commissioners of Taxation (N.S.W.) [2] ; Webb v. Outtrim [3] ) will be repeated, but the area of conflict will be criminal and civil law generally.

1. (1907) 4 C.L.R. 1178. 2. (1906) 4 C.L.R. 356.

Section 74 refers to "Constitutional powers inter se" which includes judicial powers. If the Privy Council's advice in Ibralebbe's Case [4] means that when hearing an appeal from the Supreme Court of a State the Privy Council is to be regarded as part of the judicial structure of the State, is it to be regarded as exercising State judicial power? If so, the question whether an appeal lies from a State Supreme Court to the High Court only or to the Privy Council as well may involve a question of the constitutional (judicial) powers inter se of the Commonwealth and the States which must not be decided by the Privy Council.

Even if the Privy Council did now have authority to hear appeals from State courts, the exercise of such jurisdiction in criminal cases would be unacceptable. In R. v. Bertrand, the Privy Council stated [5] :

the result is, that any application to be allowed to appeal in a criminal case comes to this Committee labouring under a great preliminary difficulty - a difficulty not always overcome by the mere suggestion of hardship in the circumstances of the case; yet the difficulty is not invincible. It is not necessary, and perhaps it would not be wise, to attempt to point out all the grounds which may be available for the suggestions, if true, raise questions of great and general importance, and likely to occur often, and also where, if true, they show the due and orderly administration of the law interrupted, or diverted into a new course, which might create a precedent for the future; and also where there is no other means of preventing these consequences, then it will be proper for this Committee to entertain an appeal, if referred to it for its decision. Other cases confirm the special and exceptional nature of the criminal appeal. There is an appeal by special leave to the High Court in any case in which such an appeal could have been made to the Privy Council (see s. 73 of the Constitution; s. 35 of the Judiciary Act). Any future granting by the Privy Council of special leave to appeal in a criminal case from a State court can only be regarded as an attempt to undermine the authority and standing of this Court. This conclusion follows irresistibly from the Privy Council's statements in cases such as Bertrand of the circumstances in which it will entertain an appeal. Whatever excuse there was for short cuts when an appeal lay from the High Court to the Privy Council in State criminal matters, there is none now. (This also applies to the Privy Council's granting of special leave in civil cases.) 1. (1867) L. R. 1 P.C. 520, at p. 530.

If appeals were to continue from State courts to the Privy Council what would be the Privy Council's function in relation to Australian cases? It would be mischievous for the Privy Council to state Australian law otherwise than in accordance with this Court's pronouncement. If it did, the consequence would be that the Judicial Committee would make a report to the Privy Council, the formal organ of the United Kingdom Government, which would then issue an order declaring Australian law contrary to that declared by the Federal Supreme Court of this country (see s. 71). This possibility reinforces the view that appeals to the Privy Council from State courts are not consistent with the Australian Constitution (particularly s. 74) since the two limitation Acts of 1968 and 1975.

The High Court is not bound by its previous decisions or by any Privy Council decision. An orderly approach to precedent suggests that previous decisions of the Privy Council, given on appeal from the High Court, should be treated for the present as equivalent to a High Court decision. Any other court in Australia faced with a Privy Council decision and a later conflicting decision of the High Court should follow the High Court; if it is faced with a Privy Council decision on appeal from outside Australia and a conflicting High Court decision (earlier or later), it should follow the High Court (even if, in the case of an earlier High Court decision, the Privy Council had taken account of it).

Australian courts will inevitably pay less regard to past Privy Council decisions on appeals from elsewhere and more regard to decisions of the supreme tribunals which are now the final arbiters of their national legal systems. The Privy Council is an eminent relic of colonialism, representing a dwindling influence (see Enid Campbell, The Decline of the Jurisdiction of the Judicial Committee of the Privy Council Australian Law Journal, vol. 33 (1959), p. 196; H. H. Marshall, The Judicial Committee of the Privy Council: A Waning Jurisdiction International & Comparative Law Quarterly, vol. 13 (1964), p. 697; the jurisdiction has dimished since these articles were published). I use "colonialism" in the generally accepted sense, not limited to what might be termed "colonies" in the strict sense (see Ibralebbe's Case [6] ). Many countries have abolished the appeal to the Privy Council and for Australian purposes it no longer has a useful role in attaining uniformity even where that is desirable.

I should add that no court in Australia is bound by the decisions of the House of Lords or the courts below it in the English system. The expression "not technically bound" is often used, but it should be clear that Australian courts are not bound by such decisions, however persuasive they may be.

In times of rapid social change, the creative role of appellate courts naturally expands to adapt decisional law to the new social environment. The Australian judicial system is not assisted in the definition and development of Australian law by the existence of a tribunal acting as a rival of the High Court. The members of the Privy Council have not been appointed by Australians, are not responsible to anyone in Australia and cannot be removed by any Australian procedure. As the framers of our Constitution understood, the members of the Privy Council generally have little knowledge of our history, our legislation, our society and therefore of how our decisional law should develop. The Privy Council itself seems conscious of this (see Geelong Harbor Trust Commissioners v. Gibbs Bright & Co. [7] ). Decisions in which State Supreme Courts have granted leave to appeal to the Privy Council show that some Australian judges are less conscious. The lesson of cases such as Oteri [8] is that Australian courts should not be encouraged to look to the Privy Council for guidance on Australian law.

1. (1974) 129 C.L.R. 576; [1974] A.C. 810. 2. [1976] 1 W.L.R. 1272; (1976) 51 A.L.J.R. 122.

Consequently, this Court is not bound by the Privy Council decision in Palmer v. The Queen [9] .

The Application and Appeal. The main grounds of the application and appeal are that the trial judge erred in failing to direct the jury: (1) that certain evidence admitted against a co-accused was not to be used against the applicant; (2) that the evidence that the applicant had taken drugs was to be weighed on the issue whether he intended to kill or inflict grievous bodily harm; (3) that if the applicant was acting in self-defence, but used excessive force, he should be found guilty of manslaughter.

There is no merit in the first ground as the trial judge gave the direction. On the second ground, I agree with Gibbs J. that special leave should be granted and the appeal allowed. The jury were entitled in reaching a conclusion whether the prosecution had established the intent beyond reasonable doubt, to consider the effect, if any, of drugs taken by the applicant. I will now deal with the third ground.

The applicant's conviction of murder was under s. 18 of the Crimes Act, 1900 N.S.W., as amended, which states:

(1) (a) Murder shall be taken to have been committed where the act of the accused, or thing by him omitted to be done, causing the death charged, was done or omitted with reckless indifference to human life, or with intent to kill or inflict grievous bodily harm upon some person, or done in an attempt to commit, or during or immediately after the commission, by the accused, or some accomplice with him, of a crime punishable by death or penal servitude for life.

(b) Every other punishable homicide shall be taken to be manslaughter.

(2) (a) No act or omission which was not malicious, or for which the accused had lawful cause or excuse, shall be within this section.

(b) No punishment or forfeiture shall be incurred by any person who kills another by misfortune only, or in his own defence.

Self-defence arises where an accused kills to avoid harm and must be disproved beyond reasonable doubt by the prosecution. The degree of harm required has not been settled, but it clearly includes death or grievous bodily harm and probably actual bodily harm. In Reg. v. Howe [10] Dixon C.J. said that it extended to insult or outrage.

It has been generally accepted that (assuming the other elements of self-defence exist) a person who kills with intent to kill or inflict grievous bodily harm is not guilty of murder (or manslaughter) if he believed that he was defending himself and there were reasonable grounds for believing that what he did was necessary to defend himself. However, if what he did exceeded what was reasonably necessary (that is, there were no reasonable grounds for believing that what he did was necessary to defend himself), the question is whether he is guilty of murder, manslaughter or neither. This question has been closely examined in case law and learned articles. The High Court decided it was manslaughter (Reg. v. Howe), and the Privy Council that it was murder (Palmer v. The Queen). I think it is neither.

The problem arises from the maintenance of the objective test (that there were reasonable grounds for believing what was done was necessary for self-defence) in addition to the subjective test (that he believed he was defending himself). In order to avoid confusion, I use the expressions "subjective test" and "objective test" as ordinarily used in the literature, although from the trial court's point of view, these descriptions would be more accurate if they were transposed.

In my opinion, the objective test should be abandoned. It is quite unrealistic and introduces problems similar to those in provocation (see Moffa v. The Queen [11] ). As Holmes J. said, " detached reflection cannot be demanded in the presence of an upraised knife" (Brown v. United States [12] ). The cases abound with statements like this neutralizing the objective test's application by references to "agony of the moment" considerations which obscure the conclusion that, if the test were dispensed with, the law would be simple and just. It is often doubted that the application of the two tests will yield different answers. As Taylor J. pointed out in Howe, if a jury were satisfied that the killing was not reasonably necessary, they would very likely be satisfied that the accused did not believe he was defending himself. The argument may be turned around: if the jury were not satisfied that he did not believe that he was defending himself, they would very likely not be satisfied that his action was not reasonably necessary.

1. (1977) 138 C.L.R. 601. 2. (1921) 256 U.S. 335, at p. 343 [65 Law. Ed. 961, at p. 963].

The argument that the objective test should be retained in order to preserve the social fabric is not convincing to me. It is a curious jurisprudence which requires acquittal of murder because, as a result of intoxication by drugs or alcohol, the requisite intent (to kill or inflict grievous bodily harm) is absent, but does not require acquittal when the accused, with that intent, killed because he honestly believed that he was defending himself (although he did more than was reasonably necessary).

There is a persistent notion in this branch of the law that murder should be reserved for killings done with intent to kill (not where the intent was only to do less harm) and where there are no mitigating circumstances. This is reflected in the common law principle that the jury although satisfied of murder, may return a verdict of manslaughter (Beavan v. The Queen [13] ). This notion is supported by the argument that a jury will often acquit if the only alternative is a verdict of murder which offends its sense of values as being an excessive assessment of the killing. The decision in Howe [14] seems to me to reflect this notion. I consider it more acceptable that in common law honest self-defence negates guilt.

1. (1954) 92 C.L.R. 660. 2. (1958) 100 C.L.R. 448.

In Howe's Case, the charge was murder at common law. But, in this case, the charge is murder as defined by s. 18 of the Crimes Act, which also deals with self-defence in homicide. The legislative command in s. 18 (2) (b) is that "No punishment or forfeiture shall be incurred by any person who kills another by misfortune only, or in his own defence". This can only sensibly mean that if the person kills in his own defence, he is entitled to be acquitted of murder and manslaughter. Application of the well-known rule which requires strict construction of penal Acts excludes a construction of the words "in his own defence" which would deprive him of the protection of the provision if he used excessive force. In my opinion, s. 18 (2) (b) forbids the application in New South Wales of the decisions in either Palmer [15] or Howe.

If the jury is satisfied beyond reasonable doubt that self-defence is disproved, then the killing is murder unless it is reduced to manslaughter because of provocation (seeJohnson v. The Queen [16] ; Moffa v. The Queen [17] ). In these presentations, the jury of course must be satisfied of the requisite intent (that is, to kill or inflict grievous bodily harm as defined in s. 18).

1. (1976) 136 C.L.R. 619. 2. (1977) 138 C.L.R. 601.

The test of proportionality has been applied as if a proportionate response between the apprehended harm and the action of the accused were essential to the defence. This is not an ingredient. Proportionality between the apprehended harm and force used to repel it merely bears on whether he was defending himself. That an accused took no less action than he was certain would avoid his own death or grievous bodily harm would not, in my view, point against his believing he was defending himself. But that is a factual general observation. Whether an accused retreated, or declared off his own attack are also for the jury on the issue. They are not conditions of the defence. Self-defence is not strictly a defence. Perhaps what is done in self-defence should be regarded simply as an act (or omission) which is not malicious within the meaning of s. 18 (2) (b).

This branch of the law suffers from the general tendency to elevate factual arguments into legal tests which are often not only erroneous but also complicate the criminal law and confuse trials. I favour the instructions on this aspect of self-defence being confined to a direction that the onus is on the prosecution to prove (beyond reasonable doubt) that the accused did not act in his own defence, and that considerations such as excessive force, proportionality and failure to retreat, are not conclusive but may be taken into account when deciding that issue. This applies also to questions of whether an accused believed he was defending himself or that what he was doing was necessary to avoid the apprehended harm, or whether he had any belief at all. In many cases of obvious self-defence, the self-defender acts instinctively. To require "belief" that what he was doing was necessary is unrealistic. If the prosecution fails, the accused should be acquitted of murder and manslaughter.

The trial judge, Taylor J., correctly left self-defence for the consideration of the jury. It is not decisive that Viro was the original aggressor. It was open to the jury if they accepted Viro's statement to infer that he, although the original aggressor, had entirely abandoned any attack on Rellis and that, when stabbing Rellis, he was acting to defend himself from Rellis' attack. Questions of whether he should have submitted (perhaps with a plea to desist) or retreated were for the jury to weigh on the issue of whether he was defending himself. His Honour directed the jury in these terms:

It is the law that any person who is attacked with violence may defend himself. If in the course of defending himself against an attack, the man who is his aggressor is killed, then it will be a question for you to determine whether or not what he did was done primarily for that purpose of defending himself against the aggressor and you would have to be satisfied beyond a reasonable doubt that this was not so before you could return a verdict of murder.

The omission of any reference to the objective test meant that this direction was not in terms of Palmer or Howe. In my opinion, the direction was correct (I leave aside the question, which was not argued, of whether the reference to "primarily" may have been unduly favourable). His Honour did not expressly state that the accused should not in those circumstances be convicted of manslaughter but this did not adversely affect the applicant. His Honour refused a request for redirection, and made some observations on his direction; this occurred in the absence of the jury and did not affect the result. This ground of appeal should fail.

I leave aside the question of whether the conviction could be justified on some basis under s. 18 other than killing with the intent mentioned. Assuming it could be, the jury or some of them may not have acted on it but on the basis to which the grounds of appeal were directed. Also, s. 18 (2) (b) applies to homicide generally.

My opinion on the correct direction to be given differs from that of the other members of this Court. In view of the division of opinion which has existed on this topic, it is important for the daily administration of justice that trial courts be in no doubt of the direction which they should give. Until this Court expresses a different view, a trial judge should give a direction on self-defence in accordance with the majority opinion in Howe's Case.

Special leave should be granted, the appeal upheld, a new trial ordered.

Argument was heard in two stages on this application for special leave to appeal from the Court of Criminal Appeal of New South Wales. The accused was convicted of murder and his appeal from that conviction was dismissed by the Court of Criminal Appeal. In the end two questions were argued, the first before a court of five judges in November 1976. The substantial point then argued was whether the trial had miscarried by reason of the trial judge's failure to instruct the jury that the accused may have been under the influence of drugs, according to his statement from the dock, and that such circumstance may be relevant to the question of whether he had formed the necessary intention to kill or to do grievous bodily harm to the deceased. With respect to that matter I have had the advantage of reading the reasons of Gibbs J. and I am in agreement with the views which he there expresses. That omission from the instructions to the jury makes it inevitable that there must be a new trial.

The case was re-argued before a court of seven judges on a separate point, whether the jury should have been instructed that they might return a verdict of manslaughter if they thought that the applicant had killed the deceased in self-defence, but that in defending himself had used excessive force. A direction in those terms would have been in accordance with the decision of this Court in Reg. v. Howe [18] . The trial judge directed the jury in accordance with the decision of the Privy Council (Lord Morris, Lord Donovan and Lord Avonside) in Palmer v. The Queen [19] which decided the contrary in an application for leave to appeal from the Court of Appeal of Jamaica. Their Lordships considered the decision in Howe and concluded that the view there expressed should not be adopted and that the correct view of the common law was that where the defence of self-defence is raised there are only two possibilities, either conviction or acquittal.

The question therefore arises whether this Court is bound by the decision of the Privy Council and should for that reason overrule Howe, or, if it is not so bound, what course it should adopt.

By virtue of the Privy Council (Limitation of Appeals) Act 1968, as amended, applications to the Privy Council for leave to appeal from the High Court became impossible in all federal matters. By the Privy Council (Appeals from the High Court) Act 1975 such applications for leave to appeal may no longer be made from any decision of the High Court or from any decision of the Supreme Court of a State exercising federal jurisdiction.

This legislation has profoundly altered the position of this Court and has had a substantial effect upon the position of the Supreme Courts of the States. Previously there had been a single judicial hierarchy of which the Privy Council was the apex. Appeals to it from this Court could go only by leave of the Privy Council but from the Supreme Courts of the States appeals lay, and still lie, in many matters as of right, and in other matters upon application for leave or special leave. In criminal matters appeals from the Supreme Courts of the States require leave or special leave, and although such leave is seldom granted there are occasions upon which matters have been fully considered in applications for special leave and other occasions on which special leave has been granted. In that hierarchy there were situations in which a particular appellate court in the heirarchy might be by-passed. For example, it was in many matters possible to by-pass the Court of Appeal or the Full Court of the Supreme Court of a State and appeal direct to this Court, and likewise there were many matters in respect of which an appeal as of right from a State Supreme Court enabled the appellant to by-pass the High Court. Either way the ultimate court of appeal was the Privy Council and its decisions were binding upon all the courts below it in that hierarchy. Those courts were bound to follow its decisions, whether they were given in appeals from the courts of the States or from the High Court, or from other courts in the British Commonwealth from which appeals lay to the Privy Council.

In such a system differences between the views of the High Court and the Privy Council were resolved on appeal to the Privy Council, either from the High Court or from State Supreme Courts, or from other courts so that the Privy Council remained at the apex of such system and there was, after the Privy Council had spoken, no room for difference between the views entertained by this Court and those entertained by the Privy Council. This situation is described in detail in other judgments and it is therefore not necessary for me to elaborate on what I have just said.

The provisions of those Acts have changed that situation and have produced the result that this Court is in respect of all matters which reach it a court of ultimate appeal, i.e. a court from which no appeal, whether by leave or otherwise, lies to any higher tribunal. In parallel with that position however there is the position of the Supreme Courts of the States from which appeals still lie, as of right or by leave, direct to the Privy Council, the unsuccessful litigant having the choice of whether to appeal as of right or by leave (as the case may be) to this Court or to the Privy Council. The State Supreme Courts thus form part of two judicial hierarchies. They run partly in parallel with the hierarchy of courts exercising federal jurisdiction. If the State courts are exercising federal jurisdiction an appeal lies only to the High Court, there being no possibility now of an appeal to the Privy Council. In respect of State jurisdiction, however, the hierarchies may at the level of the Supreme Courts of the States diverge because it is from there that appeals may go either to the Privy Council or to this Court. The doctrine of precedent however is based upon the existence of a judicial hierarchy under which each court is bound to follow decisions of courts above it as a matter of legal policy, so as to give to the system an appropriate degree of certainty by which men may govern their lives. Some intermediate courts of appeal feel free to overrule their own decisions, but in any event the ultimate court of appeal may overrule the decisions of the courts below it and now ultimate courts of appeal are free to reverse their earlier decisions, though they do so only in exceptional circumstances. This has long been the position with the High Court. The High Court however has until now been bound by decisions of the Privy Council, whether given in cases from an Australian court or from other courts because the Privy Council stood at the apex of the hierarchy. The Court however has not in recent years regarded itself as bound by the decisions of the House of Lords.

Now that the Privy Council is no longer the apex of a judicial hierarchy of which the High Court forms part the basis for adherence to the views of the Privy Council has disappeared. This new system must be regarded as leaving this Court free to differ from the Privy Council, if in the proper performance of its duty it feels that it should do so. This is not to say that the decisions of the Privy Council and of the House of Lords will not be treated with the respect properly due to tribunals of that calibre, nor as primary sources for guidance in the development of the common law and as to its content from time to time. In that situation however the High Court cannot regard itself as bound to overrule Howe [20] in deference to the decision of the Privy Council in Palmer [21] . Its task now must be to make up its own mind in the performance of its own duty to declare what is the law on this topic. Whatever the criterion by which this Court should judge whether it should overrule an earlier decision of its own, there can be no doubt that a contrary decision by the Privy Council would provide ample ground, if not compelling ground, for re-examining the earlier decision. Whether the earlier decision should be overruled or not must depend upon this Court's own view of what is the correct position for Australia.

The practical problem arises from the fact that, on an appeal from the Supreme Court of a State, the Privy Council, in the performance of its judicial function, must perform the same task, i.e. it must determine what in its view is the correct law for the relevant State and thus of all other States, as well as other jurisdictions from which an appeal lies to the Privy Council. It is not to be doubted that their Lordships will perform that task with due regard to local considerations but, leaving aside matters of statute, they must determine what in their view is the relevant principle of the common law, including any special developments which have led the common law as understood in Australia to diverge from the position prevailing elsewhere. Moreover decisions of the Privy Council in hearing appeals from other jurisdictions will, in so far as they depend upon the principles of common law, be binding on the Supreme Courts of the States.

This situation is one in which it is obvious that occasions may arise when the Privy Council and this Court will take a different view upon some matter of importance in the development of the law, and that is exactly what has happened in the present case. It is one thing for this Court to be faced with the problem of re-examining one of its own decisions in the light of a subsequent Privy Council decision and arriving at its own conclusion either to adhere to the view previously accepted or to overrule its previous decision in order to conform to the view expressed by the Privy Council, if satisfied that the latter view is to be preferred. It is quite another for the Supreme Courts of the States now to face such a conflict of authority. They are bound by the decisions of this Court, both in federal and non-federal matters, because this is a court from which appeals lie from the Supreme Courts, whatever jurisdiction they may be exercising. Equally however the State courts are bound by decisions of the Privy Council as a tribunal to which appeals from them lie in non-federal matters. If however the decisions of this Court and the Privy Council are in conflict, the Supreme Courts are placed in a position of great difficulty, a difficulty which cannot be solved by any general direction from this Court. Moreover a decision of this Court in an appeal from a Supreme Court to adhere to its previous view will not solve the problem if the same point arises in a later case, for they will still face a conflict of decisions of the Privy Council and the High Court. It is not difficult to visualize the situation where, whichever decision the Supreme Court follows, the unsuccessful litigant will seek to appeal to the tribunal which took the other view. No doubt "judicial restraint" may minimize the number of occasions when this problem will arise, but it remains one beyond the power of this Court and, as it seems to me, beyond the power of the Privy Council to resolve so long as each must, in accordance with its judicial duty, decide cases coming before it according to its best ability in accordance with the law as it sees it to be. Unsatisfactory though one must acknowledge it to be, it does not appear that any pronouncement by this Court or any direction which it may choose to give can solve the problem in the sense of ensuring that it does not occur. The choice of the tribunal to which the appeal may go from the Supreme Court is in the hands of the unsuccessful litigant who will naturally tend to appeal to the tribunal in which he thinks he will fare better and what sounder ground for preference can there be than an existing decision in his favour by that tribunal.

The situation is one without precedent, and, as nothing that we can say can assist the Supreme Courts in resolving for themselves the problems that such possible conflicts will produce, it is, I think, better not to attempt to direct in general terms what course the Supreme Courts should follow. No doubt such cases will be rare, but they will inevitably cause problems when they do arise. It is however otherwise in any particular case where this Court is seised of an appeal from a State Court, and allows that appeal, then upon a new trial being ordered the State Courts are, as I see it, bound upon such new trial to follow the direction of this Court and to decide otherwise is to deny this Court's status as a court of ultimate appeal. No one can say that such a situation is satisfactory but I can see no escape from it.

I turn now to the question of what course this Court should adopt on the present application for leave. Although I have indicated above that there is a ground upon which special leave should be granted and the appeal allowed with an order for a new trial, the question still remains of what direction the trial judge should give at the new trial in relation to the question conveniently called "excessive self-defence". That question arose at the former trial and it was one of the grounds of the application for leave that the trial judge should not have directed the jury in accordance with the decision of the Privy Council in Palmer [22] , but should have directed them according to the decision of the majority of this Court in Howe [23] . When the matter came to be argued it emerged that both the applicant and the Crown urged the Court to decide the point in favour of the adoption of the formulation in Howe rather than in Palmer.

In considering the question of whether the majority in Howe should be preferred to the decision of the Privy Council in Palmer it is not necessary or useful to set out in detail yet another examination of the earlier authorities. They were examined in Howe by Dixon C.J., with whom McTiernan J. and Fullagar J. agreed and with whom Menzies J. substantially agreed. The cases are also the subject of discussion in the Privy Council and are fully examined in Studies in Criminal Law by Professors Morris and Howard (1964), pp. 127-131. The situation as to earlier authority remains as it was prior to Howe. As Dixon C.J. then said [24] :

There is no clear and definite judicial decision providing an answer to this question but it seems reasonable in principle to regard such a homicide as reduced to manslaughter, and that view has the support of not a few judicial statements to be found in the reports. I take the statement referring to the absence of "clear and definite judicial decision" as referring to appellate courts. 1. (1958) 100 C.L.R., at p. 461.

At least since the mid-nineteenth century it has been settled law that it is not murder to kill in self-defence against an attack which threatens the life of the accused or the safety of his person, or which the accused reasonably believes so threatened him. It was expressed by Dixon C.J. in Howe as follows [25] :

it is assumed that an attack of a violent and felonious nature, or at least of an unlawful nature, was made or threatened so that the person under attack or threat of attack reasonably feared for his life or the safety of his person from injury, violation or indecent or insulting usage. This would mean that an occasion had arisen entitling the person charged with murder to resort to force to repel force or apprehended force. Had he used no more force than was proportionate to the danger in which he stood, or reasonably supposed he stood, although he thereby caused the death of his assailant he would not have been guilty either of murder or manslaughter. 1. (1958) 100 C.L.R., at p. 460.

Words to that effect have been used many times by trial judges and by courts of criminal appeal to describe the nature of the defence.

It will be observed that that formulation involves both subjective and objective elements. The objective element lies in the position of a man actually defending himself from real violence on the part of the deceased. The first subjective element is the intention of the accused to defend himself. A further subjective element lies in the alternative situation where the acts are done to defend himself from an attack to which the accused believed on reasonable grounds that he was being or about to be subjected. In such a case the use of force to an extent justified by the real or apprehended danger would complete the defence and entitle the accused to an acquittal, even if he was mistaken in his belief. The formulation in Howe gave the authority of an appellate court to a further element. It decided that in a case where the accused was actually defending himself from the real or apprehended danger or violence of the deceased, but used more force than was justified by the occasion, then if death ensued he was not guilty of murder but guilty of manslaughter. It will be noted that it uses or involves an objective criterion applied to a partly subjective situation, i.e. the amount of force is that which is objectively proportionate to the actual danger or the supposed danger.

It is useful to set out what in this respect appears to me to be the critical passage in the judgment of Dixon C.J. He said [26] :

It will be seen that the second proposition which this summary of the direction formulates raises the question whether, where upon an indictment for murder the accused relies on self-defence as a plea and all the elements of that defence are made out except that which relates to the proportion of the force used to the degree of danger threatened or reasonably apprehended, the verdict against the accused should be, or at all events may be, manslaughter and not murder. The assumption made for the purpose of this question is that a man actually defending himself from the real or apprehended violence of the deceased has used more force than was justified by the occasion and that death has ensured from this use of excessive force. In all other respects, so it is assumed, the elements of a plea of self-defence existed. The judgment then goes on to the passage which I have already quoted. 1. (1958) 100 C.L.R., at p. 460.

The Privy Council after considering Howe and the earlier authorities, as well as some later cases where the same point arose, rejected the decision in Howe. It is not useful to review again the cases to which their Lordships referred. There are however some aspects of the actual decision in Palmer which require examination. The first point to notice is that their Lordships quoted the summing up by the trial judge which contained a statement that there were only "two verdicts which are open to you in this case; guilty of murder or not guilty" [27] . In the course of his directions to the jury he had said, "A man who is attacked in circumstances where he reasonably believes his life to be in danger or that he is in danger of serious bodily harm, may use such force as on reasonable ground he believes is necessary to prevent and resist the attack. And if in using such force he kills his assailant he is not guilty of any crime even if the killing was intentional" [28] . This is in terms which are the equivalent of the basic definition of self-defence to which I referred above. After quoting further passages from the summing up their Lordships concluded, "That there is no room for criticism of the summing up or of the conduct of the trial unless there is a rule that in every case where the issue of self-defence is left to the jury they must be directed that if they consider that excessive force was used in defence then they should return a verdict of guilty of manslaughter" [27] .

1. [1971] A.C., at p. 824. 2. [1971] A.C., at p. 823. 3. [1971] A.C., at p. 824.

Their Lordships' concluding passage is of importance in the ascertainment of the basis of their decision. Their Lordships said [29] :

In their Lordships' view the defence of self-defence is one which can be and will be readily understood by any jury. It is a straightforward conception. It involves no abstruse legal thought. It requires no set words by way of explanation. No formula need be employed in reference to it. Only common sense is needed for its understanding. It is both good law and good sense that a man who is attacked may defend himself. It is both good law and good sense that he may do, but may only do, what is reasonably necessary. But everything will depend upon the particular facts and circumstances. Of these a jury can decide. It may in some cases be only sensible and clearly possible to take some simple avoiding action. Some attacks may be serious and dangerous. Others may not be. If there is some relatively minor attack it would not be common sense to permit some action of retaliation which was wholly out of proportion to the necessities of the situation. If an attack is serious so that it puts someone in immediate peril then immediate defensive action may be necessary. If the moment is one of crisis for someone in imminent danger he may have to avert the danger by some instant reaction. If the attack is all over and no sort of peril remains then the employment of force may be by way of revenge or punishment or by way of paying off an old score or may be pure aggression. There may no longer be any link with a necessity of defence. Of all these matters the good sense of a jury will be the arbiter. There are no prescribed words which must be employed in or adopted in a summing up. All that is needed is a clear exposition, in relation to the particular facts of the case, of the conception of necessary self-defence. If there has been no attack then clearly there will have been no need for defence. If there has been attack so that defence is reasonably necessary it will be recognised that a person defending himself cannot weight to a nicety the exact measure of his necessary defensive action. If a jury thought that in a moment of unexpected anguish a person attacked had only done what he honestly instinctively thought was necessary that would be most potent evidence that only reasonable defensive action had been taken. A jury will be told that the defence of self-defence, where the evidence makes its raising possible, will only fail if the prosectuion show beyond doubt that what the accused did was not by way of self-defence. 1. [1971] A.C., at pp. 831-832.

If that passage were read alone it would suggest that there is no subjective element involved in self-defence at all, for only objective factors are mentioned. With great respect it does not appear to me that it can be left to the jury without explanation. No doubt there is no formula which must be used but the content of the defence must be described and explained. As I have said above their Lordships approved the summing up which included a clear explanation of the subjective factor of a reasonable, though mistaken, belief in the danger of imminent attack, but with great respect the passage just quoted appears to me to state clearly enough that there is no subjective element involved at all.

This uncertainty, which may well trouble trial judges, is a reason for preferring the decision of this Court in Howe [30] . There are however more important reasons than that.

In my opinion there is a real distinction in the degree of culpability of an accused who has killed having formed the requisite intention without any mitigating circumstance, and an accused who, in response to a real or a reasonably apprehended attack, strikes a blow in order to defend himself, but uses force beyond that required by the occasion and thereby kills the attacker. Such a killing is undoubtedly unlawful, but it appears to me to differ significantly from murder.

I have had the advantage of reading the reasons for judgment of Stephen J. and Mason J. and I agree with their reasons for preferring to adhere to this Court's decision in Howe rather than to overrule it in favour of Palmer [31] . In particular I agree with the formulation of the issues which arise in the application of the decision in Howe which is set out in the final paragraph of the reasons for judgment of Mason J.

In my opinion special leave should be granted and the appeal allowed. The order of the Court of Criminal Appeal should be set aside and there should be a new trial.