England and Wales Court of Appeal (Criminal Division) Decisions

You are here: BAILII >> Databases >> England and Wales Court of Appeal (Criminal Division) Decisions >> Roach, R v [2001] EWCA Crim 2698 (4th December, 2001)
URL: http://www.bailii.org/ew/cases/EWCA/Crim/2001/2698.html
Cite as: [2001] EWCA Crim 2698

[New search] [Printable RTF version] [Help]

Roach, R v. [2001] EWCA Crim 2698 (4th December, 2001)

		Neutral Citation Number: [2001] EWCA Crim 2698
		Case No: 2000/02872/X5

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CRIMINAL DIVISION)
ON APPEAL FROM THE CROWN COURT AT
MIDDLESEX GUILDHALL
(HIS HONOUR JUDGE LAWRENCE)

Royal Courts of Justice
Strand, London, WC2A 2LL

4th December 2001

B e f o r e :

LORD JUSTICE POTTER
MRS JUSTICE RAFFERTY
and
HIS HONOUR JUDGE ZUCKER QC
____________________

REGINA
-v-
MICHAEL ROACH

____________________

(Transcript of the Handed Down Judgment of
Smith Bernal Reporting Limited, 190 Fleet Street
London EC4A 2AG
Tel No: 020 7421 4040, Fax No: 020 7831 8838
Official Shorthand Writers to the Court)

____________________

Derek Zeitlin Esq (instructed by the Crown Prosecution Service for the Crown)
Wayne Jordash Esq (instructed by David Phillips, solicitor for the appellant)
____________________

HTML VERSION OF JUDGMENT
AS APPROVED BY THE COURT
____________________

LORD JUSTICE POTTER :

Introduction

On 14 April 2000 in the Crown Court at Middlesex Guildhall before His Honour Judge Lawrence and a jury, the appellant was convicted following a four day trial for Wounding with Intent to do grievous bodily harm, the jury having rejected his defence that he was in a state of automatism at the time of the alleged offence. He was sentenced to four years' imprisonment. He now appeals against conviction by leave of the Single Judge.

The Facts

2. The short facts are that the appellant and the victim, Mr Ahmed Salman, were both employed as co-workers in a catering franchise at the Royal Tournament. They disliked each other. The prosecution alleged that a trivial incident led to the appellant attacking the victim with a knife. As a result Mr Salman suffered a badly cut hand and another lesser wound. Several bystanders witnessed the incident and said that the appellant appeared calm afterwards. It was the defendant's case that he had no knowledge or memory of the incident itself, while being aware of events up to moments before, and then almost immediately after, the stabbing. Two defence psychiatrists described him as having an anti-social personality disorder and said that he had acted in a state of automatism of a psychogenic type as a result of a number of factors working in combination to induce such a state. The consultant forensic psychiatrist for the Crown rejected this diagnosis.

The Evidence

A summary of the evidence for the Crown is as follows.

The employer of the appellant and the victim confirmed that at the time the men were working together on a stand behind the Royal Box at the Tournament. He said it was clear to him the day before that the two were not getting on with each other.

Mr Salman said in evidence that, when he was looking for a mop to clean the floor, he asked the appellant who had come into his area, whether he had found the mop by any chance. The defendant pointed to his face, almost touching his nose and saying, "look, you're not my supervisor". Mr Salman told him to cool down and the defendant walked off. As Salman was packing up, someone from behind him shouted and he turned and saw the appellant with a serrated knife presented at his neck. The appellant said "You see this knife, I'll cut your throat with it". Mr Salman retreated a few paces, the appellant following him, swearing all the time, "Fucking frog bastard". He pushed him back and suddenly the police were there. Salman had no memory of being stabbed, but it was plain afterwards that he had a serious injury to his hand and a comparatively minor wound elsewhere.

There were a number of witnesses. One, a visitor to the Tournament with his son, said that he heard the defendant say "Don't take the fucking piss out of me" and he then saw him stab out at Salman's left neck and shoulder area. He confirmed that Salman moved back away from the knife, but the appellant stabbed out at him two or three times. Salman was driven back and bent back over some boxes, the defendant then changing from an under-arm movement with the knife to an over-arm 'psycho-style' of stabbing action. He stabbed downwards about twice in Salman's left shoulder area.

A Corporal Speight who saw the incident said that Salman appeared to give up as he was arched back and the appellant was doing a gesture with the knife across his body. The appellant then changed his grip into a stabbing action, stabbing over-arm. Immediately after it was all over, the defendant was saying "Everything's cool, don't worry". Speight said the defendant seemed to realise that people had been watching him. He said the defendant seemed to be aware of the severity of what he'd done, i.e. that he had just stabbed someone with people standing around watching.

According to Lance Corporal Tulip who stood next to the appellant after the incident, the appellant said "Everything's ok, calm, sorted" or words to that effect. He said the appellant spoke to him with his palms out saying, "He's taken the piss, he's saying we're no good". He stated that the appellant's face was "pent up and full of anger" during the attack, and afterwards he seemed keen to calm the situation down. He stated that the appellant did not seem bewildered and seemed fully in control of the situation.

At interview following his arrest, the \appellant gave an explanation consistent with his later evidence at trial. He said he could not remember anything about the incident but did not challenge the witnesses' accounts of events. He remembered talking to some cavalrymen before he was arrested. He said a doctor had once informed him that he was schizophrenic but he was not registered as such. He said his own doctor believed he was a manic-depressive.

A summary of the defence evidence is as follows.

The appellant repeated in evidence that he had no recollection of the incident. He said that in childhood he had had an unsatisfactory relationship with his father, an ex-Naval man who was difficult to get on with. He had first seen a psychiatrist in 1998 because his temper had been getting out of control. He had verbally abused his partner and occasionally had the idea of attacking someone, but he never did. He said he had had difficulty in working with Salman who appeared unable to do anything correctly. They had exchanged bad language on the day prior to the event, but had then shaken hands. On the following day, when he was cleaning up at about 9.30p.m, the victim came up and called him "arsehole" and "wanker". He (the appellant) was 'pissed off' with him and decided to confront him and sort things out. He was angry, but could not remember a scuffle. He recalled speaking to some soldiers before his arrest but felt nothing when he was arrested. He said that he had previous convictions and had only recently completed a sentence for robbery.

Dr Partovi-Tabar, a Consultant Psychiatrist gave evidence that he considered the appellant's various adolescent difficulties, his relationship with his father, abusive experiences during his childhood and lifelong immaturity and 'unsettledness' had led to an anti-social personality disorder. He said that, apart from the personality disorder, there was no other evidence to suggest he was suffering from any other form of recognisable mental illness. The doctor observed that despite his lifelong unsettledness and inclination to impulsivity, the appellant had 'always avoided violence'. He considered that the sudden violent attack against a fellow worker with total amnesia for the event, which had not modified over the past month, indicated a diagnosis of automatism. He went on:

“Based on his childhood experiences and his lifelong relationship with his father, his recent conflict with his rather strong-willed partner and finally confrontation with a perceived humiliation from a fellow worker and specially under the mitigation of fatigue, alcohol and prescribed medication, the most likely diagnosis is obviously an insane automatism of psychogenic type.” (emphasis added)

The doctor stated it was also possible that the frontal lobes of the appellant's brain had been damaged as a result of a fall in the school playground when he was ten. Reports following EEG and CT scans showed activity within normal limits, but these were 'fairly insensitive investigations' and the possibility of an underlying organic condition in the form of temporal lobe epilepsy 'cannot be totally ruled out'.

Dr Jeffrey Roberts, a Psychiatrist, stated in evidence that, in his opinion, the appellant's actions were attributable to a combination of factors. He said that the appellant did not suffer from any recognisable ongoing mental illness which in itself could amount to a diagnosis of insanity, although it significantly affected the cognitive management of his behaviour in the form of poor impulse control to the extent that under certain circumstances the impulse to behave violently could overwhelm the cognitive process which might have controlled the impulse. He explained that an impulse could at times become irresistibly powerful, particularly in those who by virtue of their personality disorder and life experience were relatively inarticulate in areas of powerful feeling. In this case the offence was carried out by a man who had through his life many experiences of humiliation at the hands of larger and superior others and, on the night of the offence, his weakened cognitive processes were confronted with perceived provocation and humiliation from his victim. He also said that there was no evidence of any ongoing physical condition known to allow automatism or automatic behaviour to occur such as epilepsy, insulin dependent diabetes, or physical brain disease. However there were strong psychological factors pressing for expression in a man whose senses were impaired by fatigue, alcohol and prescribed medication at the time of the offence. In these circumstances he considered that the offence fell into the category of ‘insane automatism of psychogenic type’ as a result of a number of contributory factors, namely:

(1) Mixed Personality Disorder in the form of 'an inclination to impulsivity, but an avoidance of violence (he tells me)’;

(2) 'A long-standing learnt sensitivity to humiliation from aggressive authority figures';

(3) Impairment at the time of the offence of his 'higher mental functions' by a mixture of Carbamazepine (Tegretol) 200mg twice daily, Paroxetine (Seroxat) 20mg in the morning daily, two large vodkas and fatigue. The appellant had had life stresses in his previous six months consisting of a deteriorating relationship with his partner with outbursts of verbal violence and violence to their property and the failure of his plan to become a single parent family, perceived as sabotaged by his parents, particularly his father. His 'immediate stressor' was feeling abused and humiliated, ongoing for two days, by his (in his view) incompetent supervisor.

The opinion of Dr Roberts, as finally stated, was that the offence was committed in a state of 'Insane Automatism of Psychogenic Type' (cf the view of Dr Partovi-Tabar, paragraph 12 above). No internal physical factor known to be the sole agent in liberating an automatism was active on this occasion. Powerful non-conscious psychological events appeared to have overwhelmed a chemically impaired conscious 'psyche' in a man with a clear-cut personality disorder. The threshold for 'psychogenic automatism' was lowered by fatigue, prescribed medication and alcohol.

Contrary expert evidence was given for the prosecution by Dr Philip Joseph, a Consultant Forensic Psychiatrist who stated that the appellant's amnesia, even if genuine, did not necessarily denote his inability to form the required intention for the offence. He had found no evidence to support the proposition that the appellant had acted in a state of automatism of a psychogenic type. If he had been in such a condition, he would have been bewildered following his arrest; he would have had no memory of the incident. In Dr Joseph's view his behaviour was entirely consistent with someone who had become increasingly angry and had finally lost his temper, there being no evidence of psychotic medical illness.

Discussion of Directions

Upon the completion of the evidence, there was a discussion between the judge and Counsel concerning the directions to be given to the jury in relation to automatism. Mr Zeitlin, who appeared for the Crown submitted that the burden of proof of ‘insane automatism’ as diagnosed and advanced by Doctors Partovi-Tabar and Roberts lay upon the defence on the balance of probabilities in respect of which, if accepted, the jury should render a verdict of not guilty by reason of insanity. He referred the judge at length to the section in Archbold headed 'Insanity or Automatism?' now to be found at Archbold 2001, paragraphs 17-84 -17-95, in which the relevant authorities are referred to and extensively quoted. He submitted that it was clear from the evidence of the doctors that the appellant's state at the time of the attack was, in the words of this court in R v Burgess 93 Crim App.R.41 at 48:

“… an abnormality or disorder, albeit transitory, due to an internal factor whether functional or organic which had manifested itself in violence … [and] … and might recur …”

It was implicit in the judgment in Burgess that the danger of recurrence of a mental disorder manifesting itself in violence may be an added reason for categorising the condition as a disease of the mind, but the absence of such danger is not a reason for saying that it cannot be such a disease. Mr Zeitlin submitted that for the purposes of the McNaghten Rules the Mixed Personality Disorder spoken to by both defence doctors and specifically designated by them as giving rise to an 'insane automatism of psychogenic type' was, if accepted by the jury as having given rise to a state of automatism at the time, an abnormality or disorder due to an internal factor (whether functional or organic) which amounted to a disease of the mind for the purpose of the McNaghten Rules. He submitted that the evidence of Dr Joseph and, on analysis, indeed the evidence of Dr Roberts and Dr Partovi-Tabar, was no more than evidence of a so-called 'uncontrollable impulse' of a type which could not give rise to a defence of automatism but merely amounted to a plea of partial or reduced control in a volatile personality.

Mr Jordasch who appeared for the defendant as well as in this court, submitted for his part that Dr Roberts and Dr Partovi-Tabar had laid a sufficient evidential basis to show that there was disease of the mind and that it led to a state of automatism, rather than so called uncontrollable impulse. However, he also submitted that it would be open to the jury upon the evidence of the two defence doctors to come to the conclusion that the alcohol and the prescription medicine had acted in combination with the appellant’s personality disorder to bring about a state of automatism, so giving rise to a defence of 'non-insane' automatism, which would entitle the defendant to an acquittal, as opposed to a verdict of 'guilty but insane'. In this respect he relied upon the authority of R v Quick [1973] QB 910, a case in which the accused raised a defence of automatism by reason of an imbalance of insulin which, as a diabetic, he was taking on prescription. In particular, Mr Jordasch relied upon the following passages at p.922:

“In our judgment, the fundamental concept is of a malfunctioning of the mind caused by disease. A malfunctioning of the mind of transitory effect caused by the application to the body of some external factor, such as violence, drugs, including anaesthetics, alcohol and hypnotic influences cannot fairly be said to be due to disease...

Quick's alleged mental condition, if it ever existed, was not caused by diabetes, by his use of insulin prescribed by his doctor. Such malfunctioning of his mind as there was, was caused by an external factor and not by a bodily disorder in the nature of a disease which disturbed the working of his mind. It follows, in our judgment, that Quick was entitled to have his defence of automatism left to the jury...”

Mr Jordasch also relied upon the decision of this court in R v Burns 58 Crim.App.R.364, a case in which the appellant had a history of clinical alcoholism and periods of amnesia, both in the sense of not remembering past events and 'in the sense that the thing does not register at the time because the brain function is impaired at the time'. When asked about the appellant's probable condition at the time of the alleged events, his expert medical witness said he had probably taken a quantity of alcohol, possibly made much worse by a quantity of Mandrax and his state was such that he was not aware of himself. He also stated that the appellant had brain damage, which needed drink or drugs in addition to cause him to act in the way that he had. This court held the trial Judge to have been in error because he did not leave the issue of automatism to the jury as a separate issue from insanity, being of the view that the doctor's evidence showed that, if automatism had occurred, it must have been due to disease of the mind. The court considered that, in the light of the decision in Quick, automatism should have been left as a separate issue because there was some evidence 'that other factors were operating upon a disease of the appellant's mind', which factors appeared to have been the consumption of alcohol or Mandrax in a situation where the appellant did not appreciate the effect which they might have had. Mr Jordasch submitted that, similarly in this case, the effect of the medical evidence for the defendant was not that the defendant's personality disorder was the cause of his inability to control his actions, but that it was the effect of drink and drugs acting upon the personality disorder.

The Summing-up

The judge stated that he would consider the submissions overnight and would be grateful for the opportunity to have further discussion as to what he proposed to do before the jury returned. In the event, he did not do so. However, from the form of his summing up, considered in the context of the earlier submissions made to him, it is plain that, while he considered that the evidence was sufficient to raise a case of automatism, it was one that should only be left to the jury on the basis of automatism which merited a verdict of guilty but insane (in respect of which the burden of proof lay upon the defence on the balance of probabilities), coupled with a direction upon the specific intent required for a Section 18 offence and the necessity for the prosecution to prove intent beyond reasonable doubt. When summing up, he dealt with it in the following manner.

First, he gave a full and clear direction upon the meaning of intent to do grievous bodily harm and the necessity for the need for proof beyond reasonable doubt. Then, he referred to the unusual situation that the defendant did not dispute the facts or raise a plea of self-defence, but said that he did what he did in a state of automatism. He said:

“It is the defendant's case that he did not possess the necessary intent that I have just explained to you because he was, he says, acting in a state of automatism. Automatism means in law that the defendant's state of mind was such that at the time of the assault his ability to exercise voluntary control was totally destroyed. Impaired or reduced or partial control was not enough. To establish the defence, the defendant has to satisfy you that his voluntary control was totally destroyed.

Again unusually, the law requires that since it is the defendant who claims that he was in this state proves this aspect of his defence to you and the standard is on the balance of probabilities. What that means is that you find it more likely than not that he was in a state of automatism. If you do so find, then you do not return a simple verdict of not guilty, but a special verdict of not guilty by reason of insanity. If you do not find the defendant has proved to you that it is more likely than not that he was in a state of automatism, your verdict will be one of guilty as it follows that you have not (sic) found that he carried out the wounding with the necessary intent.

There are in fact three possible verdicts open to you. If you are satisfied so that you are sure that he intended to do grievous bodily harm and you are satisfied that he carried out the act, then your verdict will be one of guilty. If you feel it more probable than not, more likely than not, that he carried out the act in a state of automatism, that is that his voluntary control over his actions was totally destroyed, then your verdict should be one of not guilty by reason of insanity. It is an old-fashioned expression. It does not mean that he is in a psychotic state or raving mad, but it indicates your finding in a legal way. If it is your finding that he was not in a state of automatism, but he simply did not have intention to do grievous bodily harm, then your verdict would be one of not guilty.”

The judge then reviewed the factual evidence. He went on to summarise the medical evidence. He encapsulated the evidence of Dr Roberts much as we have done above. However, he dealt with the incidence of drink and drugs as spoken to by Dr Roberts in the following manner:

“He was also of the view that at the time of the offence his higher mental functions would have been impaired by a mixture of the two medications, two large vodkas and tiredness. We know that he had two prescribed medications, Carbemezepine - which is a mood stabiliser, it can be given in the treatment of epilepsy but all agree that the defendant does not suffer from epilepsy, he had it as a mood stabiliser - and Paroxatine which as a trade name in the form given to the defendant of Serotax, which is an anti-depressant. He thought that the combination of those, together with the vodkas that the defendant had taken and tiredness at the end of the day had led to a degree of stress.”

The judge then referred to the various other factors of stress, which had built up in the defendant's life and said of Dr Roberts:

“...He thinks that the trigger was feeling abused and humiliated by Mr Salman who he may have seen as another authority figure which set off what Dr Roberts thought was an automatism of a psychogenic type.”

[It will be noted that the judge did not speak of 'insane automatism’ in that respect, as Dr Roberts had.]

Having briefly summarised the evidence of Dr Partovi-Tabar much along the lines we have already described, the judge stated:

“So he is in effect saying that there may be some damage to his brain, to his frontal lobes, but in addition there is a personality disorder and the congruence of the various factors that he lists brought him to a state that Dr Partovi described as an insane automatism.”

Having then given an account of the evidence of Dr Joseph, the judge turned to various points, which he suggested the jury might wish to consider. He stressed the agreement of the doctors on the absence of any 'psychotic mental illness', the absence of any evidence of brain damage or treatment for epilepsy and commented upon the question of medication and drink in this way:

“The medication that he was prescribed in Cornwall is for calming his violent behaviour and for stabilising his mood and certainly you may think is not going to, if I may use the vernacular expression, wind him up. It is not going to send him flying off to stab someone because of the medication that he has consumed. You may wish to bear in mind the amount that he had had to drink and he demonstrated with one of our little brown plastic mugs by indicating a drop at the very bottom of that when he held it up to show how much he had drunk, and he said that he had had two vodka's during the course of the afternoon and evening.”

He then turned to the question of automatism as follows:

“So you are going to have to consider whether the doctors are in fact all talking of the same thing when they speak of automatism and whether or not Dr Bartovi or Dr Roberts are talking of a blind rage where there is simply a loss of self control and the defendant goes off and does something which he regrets, whether they are suggesting that he is in some form of a trance but again no question of epilepsy...

I remind you again so that you bear fully in mind that although the doctors had different views of what automatism is, the law only has one view, that the defendant's state of mind was such that at the time of the assault, his ability to exercise voluntary control was totally destroyed. Impaired, reduced or impartial control is not enough. You are going to have to consider the difference between amnesia and automatism. It perhaps matters not very greatly that the defendant says that he can't remember what happened. That does not necessarily mean that he was in a state of automatism. Sometimes, you may think, one does something very shocking that one regrets bitterly and the mind blanks it off, but that doesn't mean that at the time it happened that you weren't aware of what you were doing and couldn't if you had really put your mind to it, stopped doing it.”

After a few more observations, the judge directed the jury to weigh the whole of the evidence looking at it objectively and then to consider one of three verdicts: -

“... Guilty if you are satisfied so that you are sure the defendant wounded and intended to do really serious bodily harm and that he was not in a state of automatism. If you feel on the balance of probabilities that he may have been in a state of automatism when he carried out the attack, then your verdict is not guilty by reason of insanity; only if you are satisfied that he carried out the attack but on all the evidence he was not necessarily in a state of automatism but he did not intend to do any really serious harm, would your verdict be not guilty.”

The Grounds of Appeal

There are four grounds of appeal in this case, which we now recite and to which we will turn in somewhat different order than that taken in the grounds of appeal. First, it is complained that the judge failed to leave to the jury an issue of 'non-insane' automatism in respect of which, once it had been raised in the evidence, no burden of proof lay upon the defendant. Second, it is complained that, in various respects, the summing up of the defence evidence and arguments was inadequate in that:

a) The judge failed to remind the jury of the evidence of two eyewitnesses, which was supportive of the defence case;

b) the judge further made adverse comments on the evidence of the defence psychiatrists; and

c) the judge failed to highlight the defence evidence in a fair and balanced manner.

Third , the judge failed to discharge the jury upon the application of the defence following certain adverse remarks which he made about the appellant's dishonesty, having previously ruled that his criminal record in respect of offences of dishonesty should not go before the jury. Fourth, the judge failed to give a Lucas direction concerning lies allegedly told by the appellant. We can say at once that we consider there is nothing in that fourth ground of appeal and we propose to say no more about it.

Automatism

The task facing the judge in this case was not an easy one. In Quick (Supra), the nature of the defence of automatism and, in particular, whether the raising of such a plea amounts to a defence of insanity was described at p.922 as a:

“...quagmire of law, seldom entered nowadays save by those in desperate need of some kind of defence. It will always be for the judge to decide whether the evidence given in a particular case is sufficient to raise an issue of automatism sufficient to be left to the jury. If there is, the further problem which faces the court is whether the underlying condition or illness raised by the defendant as the reason for his involuntary or unconscious act at the time of the offence charged amounts to, or is no more or less than a 'disease of the mind' which raises a defence and attracts a verdict of insanity or, is due to a lack of mens rea by reason of some failure of the mind or consciousness not due to disease. The practical importance of this is not merely the difference in the form and effect of the verdict, but because, in relation to the issue of insanity, the onus is on the defence to prove insanity on the balance of probabilities, whereas the issue of 'simple' automatism not due to a disease of the mind, once raised, leaves the onus upon the prosecution to negative automatism upon the usual criminal standard: see Batty v A-G [1963] AC 386 and R v Burns 58 Crim. App.R. 364. In the former case, the 'automatism' of the appellant being due to psychomotor epilepsy was held to amount to disease of the mind, whereas in the latter case there was, in addition to an underlying disease (clinical alcoholism) evidence that the triggering mechanism of the defendant's impairment of brain function at the time was consumption of alcohol, possible made much worse by a quantity of Mandrax. That being so, the Court of Appeal held that the judge should have left the issue of (non-insane) automatism to the jury as a separate issue from insanity.”

Similarly, in Quick at p.922, it was stated:

“Our task has been to decide what the law means now by the words 'disease of the mind'. In our judgment, the fundamental concept is of the malfunctioning of the mind caused by disease. The malfunctioning of the mind of transitory effect caused by the application to the body of some external factor such as violence, drugs, including anaesthetics, alcohol and hypnotic inferences cannot fairly be said to be due to disease..... from time to time difficult borderline cases are likely to arise.”

In our view, on consideration of the medical evidence for the defence, this was such a borderline case.

As already indicated, the prosecution urged upon the judge that what the doctors for the defendant placed before the jury amounted to 'Insane Automatism'; indeed it was so described by them in the phrase 'insane automatism of psychogenic type', the 'disease of the mind' which they identified being described as 'mixed personality disorder'. On the other hand, as the defence urged upon the judge, the opinion of the doctors plainly accorded a causative role to the contributory factors of alcohol and prescribed medication in combination with the personality disorder and fatigue. As counsel rightly submitted to the judge and, indeed, Mr Zeitlin for the Crown conceded before him, whatever the psychiatric definition adopted by the doctors in their reports and evidence and their reference to ‘insane automatism’, the legal definition of automatism allows for the fact that, if external factors are operative upon an underlying condition which would not otherwise produce a state of automatism, then a defence of (non-insane) automatism should be left to the jury.

Upon careful consideration, it seems to us that such was the position here. Mr Zeitlin's argument before the judge and before us was not one in which he challenged the principles we have just described. The substance of his argument was a challenge to the adequacy of the evidence that contributory factors of drink and/or drugs had in fact exacerbated the defendant’s underlying condition. In sum, he pointed out to us that the drugs which were referred to were prescribed for the purpose of having a 'damping' effect upon the anger and volatility of the defendant and that the evidence of the defendant as to the amount of drink which he had consumed did not justify the description ‘two large vodkas’. So, he submitted, the combination of external factors alleged to have tipped the defendant over the edge when combined with his personality defect and general fatigue, could effectively be dismissed. Mr Zeitlin further argued that, in any event, the approach adopted by the judge of leaving the automatism defence to the jury simply upon the basis of a verdict of insanity, followed immediately by his direction on the need for prosecution to prove intent, was sufficient to cover the issue of ‘non-insane’ automatism. There was thus no 'lacuna' to the disadvantage of the defendant.

We regret that we cannot agree with Mr Zeitlain in that respect. It seems to us clear from the nature of the summing up that the jury were directed in terms which must have driven them to assume that the burden of proof for the defence of automatism lay upon the defendant upon the balance of probabilities and that if this burden was not discharged, then the defence of automatism must fail. Whilst logically it may be argued that a direction to the effect that the jury must in any event be satisfied upon the question of intent covers the issue of non-insane automatism, we do not think that the judge’s direction made it sufficiently clear to the jury in the context of his overall summing up. In the first relevant passage of the direction quoted by us at paragraph 19 above, it will be noted (a) that, in the second paragraph, the meaning of the last sentence is obscure (b) that, in the third paragraph, the judge dealt with the third verdict open to the jury by reference to a positive finding on their part that the defendant was not in a state of automatism', rather than in terms of the burden of proof which lay upon the prosecution. Similarly, in the passage we have quoted at paragraph 25 above, the direction again lacked clarity as to the overall burden of proof upon the prosecution.

Failure to mention certain evidence

Having found as we do, we can deal with the remaining two grounds of appeal shortly. We do indeed consider that it was a failing on the part of the judge to omit reference to two witnesses upon whose evidence the defence relied as providing some, albeit ambivalent, support for the evidence of the defence experts that the defendant had not realised what he was doing, in the face of Dr Joseph’s clearly stated opinion that the appellant’s amnesia was assumed. The first of the witnesses, Philip Thomas, an army musician posted to the Royal Tournament and present during the assault, accepted in cross examination that the appellant stopped attacking Salman because it appeared he had a sudden realisation of what had happened and dropped the knife; that the appellant had approached him and other soldiers behaving as if nothing had happened and appearing unconcerned; and that the appellant following the assault appeared to 'snap out of something'. Similarly, Silvia Forrani stated that after the attack 'when they separated I saw the white man laugh and joke as if nothing had happened.' Mr Jordasch, following the jury's retirement, raised the judge’s omission, but the judge declined to correct it.

Character

The third ground of appeal arises from an unfortunate intervention by the judge in the course of the re-examination of Dr Partovi-Tabar. It came about in this way. The appellant had a bad record for offences of dishonesty. Early in the trial, the prosecution had raised with the judge the extent to which the evidence of the defendant put his character in issue on the basis that, if the defendant was proposing to call psychiatric evidence as to the nature of his propensity for violence, his character became an important matter and could not be explored by the doctors without knowing his antecedents. It was clear that the defence doctors had been aware of a conviction for robbery on the part of the defendant and had taken it into account when coming to their conclusions. However, it was not clear that they were aware of other offences. In the event, the judge ruled that it would be proper to allow evidence of the appellant's previous convictions only insofar as they were offences of violence or related to the possession of drugs.

Mr Zeitlin had submitted to the judge that it would be unduly narrow to exclude the appellant's many convictions for dishonesty because, as he submitted, his credibility would be very much in issue not merely in respect of his account given to the jury, but in respect of what he had told the psychiatrists as the basis for their evidence. In response, the judge indicated that the psychiatrists should simply be challenged on the lines that they had made their assessment on the basis of what the defendant told them, which might or might not be true. Mr Zeitlin indicated that he wished to ascertain whether they were aware that the appellant had a large number of convictions for dishonesty. However, the judge headed him off in this respect, saying that a question along the lines 'he has a large number of criminal convictions, do you know about them?' should be sufficient. The judge said that he did not think that questioning should go further than that. Mr Zeitlin thereafter acted in accordance with that ruling.

In the course of the evidence of Dr Partovi-Tabar he was challenged by Mr Zeitlin on the basis that, although he had been aware of the appellant's involvement in a robbery, he had not been aware of another conviction for violence in 1984 and incidents of violence with the appellant's father. The doctor so conceded. In re-examination, Mr Jordasch took up those matters with Dr Partovi-Tabar, asking him whether, in the light of his knowledge of those matters now, it caused him to alter his opinion. At that point the judge intervened and the following exchange took place:

Judge: I think you are still giving a partial view to the doctor. What about your client's general honesty?

Mr Jordasch: I am not with you, your Honour.

Judge: Well, you are simply putting selected convictions. He has many convictions for dishonesty. If the doctor had known that he is a very dishonest man, would he have viewed this evidence any differently?

Mr Jordasch: Your Honour, a matter of law. Can I address you in the absence of the jury?

Judge: No, dealt with that. Finish your witness, please.

Mr Jordasch then persuaded the judge to permit him to deal with the matter in the absence of the jury. He protested to the judge that, despite his earlier ruling, the jury had now been made aware, as a result of the judge's own observations, of the appellant's convictions for dishonesty. The judge observed that, as well as ruling that the convictions should not go before the jury, he had also ruled that the question of the honesty of the appellant was a matter that the doctor was going to have to consider. He stated to Mr Jordasch that he (Mr Jordasch) was now being selective in asking about the assaults and said:

“What if he [i.e. the doctor] had known about the general honesty of your client?”

When Mr Jordasch replied that the judge had ruled that the Crown were not to cross-examine on the offences for dishonesty, the judge simply observed 'Well, there we are'. Mr Jordasch then protested that, because the Crown had cross-examined the doctors on the basis that they did not know about incidents of violence only and had made no points on the question of convictions for dishonesty, he (counsel) felt unsure about how he could now deal with the matter without highlighting the issue for the jury. The judge observed 'That is a matter for you' and called for the jury to return. Mr Jordasch then completed his short cross-examination without further reference to the question of dishonesty.

We must say that we are surprised that the judge saw fit to intervene as he did. It seems plain that his intervention ran contrary to the previous indication he had given as to the exercise of his discretion. He appeared to see some criticism as attaching to Mr Jordasch for the way he re-examined, but we are unable to support any such criticism, nor did Mr Zeitlin feel able to do so. If the judge had felt, as he plainly did, that, as matters turned out, it was important for the question of the appellant's convictions for dishonesty to be put to the psychiatrists, he should have indicated in the absence of a jury that he was minded to reconsider and to re-exercise his discretion in the matter; there could then have been proper discussion as to the appropriate procedure to follow in that respect. It seems to us that, not only was the judge's intervention prejudicial to the defence of the appellant, but it was made in a way which appeared to indicate a clear expression of his own opinion to the jury that the appellant was 'a very dishonest man'. It further contained the implication that this had been concealed from the defence doctors, thereby undermining their evidence. In this respect, defence counsel was put in an extremely embarrassing position at a late stage, in which he had no opportunity to retrieve the position in a considered way.

Other matters of less importance have been raised by counsel in support of the assertion that, when summing up, the judge made various comments at the expense of the defendant and his psychiatrist which amounted to a failure to deal with the defence evidence in a fair and balanced manner. It is not necessary in our view to deal with these matters, save to say we see little in them. We consider that the matters with which we have already dealt under Grounds 1-3 are sufficient to give us concern for the safety of the conviction. We come to that conclusion with some reluctance, because as it seems to us, the case was a strong one and was almost certainly decided upon the basis of a strong preference by the jury for the evidence and assessment of Dr Joseph, rather than that of the defence doctors. However, the matters which we have mentioned plainly went to the success or otherwise of the plea of automatism which the judge had considered was sufficient to be left to the jury. In our view, they amounted to irregularities which cumulatively render the conviction unsafe. In those circumstances this appeal will be allowed and the conviction quashed.