Armstrong v Chief Constable of West Yorkshire Police

The Chief Constable of West Yorkshire Police has permission to appeal against the ruling by Mr Recorder Myerson QC, delivered on 8 May 2008 in Leeds County Court, that there were no reasonable grounds for the arrest of the respondent, Mr Bernard Daniel Armstrong, on suspicion of raping a young woman I shall call Miss X.

It is necessary to rehearse the details of the appalling attack upon Miss X. In the early hours of 5 February 2002, she left a nightclub in Leeds. She was drunk and therefore particularly vulnerable. A man was seen speaking to or getting a light from one of the doormen of the club. She jumped into a taxi and, uninvited, the man joined her. He claimed he was a student, and suggested that they share the taxi to their halls of residence. The man knew enough about the area of Leeds where she lived to give the taxi driver directions, and instructed him to drop them off a short distance from the halls of residence in a dark, dirty area. Miss X walked away from the man, but he followed her and struck her a number of times to the head with a large stone. During the attack, she was mercifully unconscious for part of the time. She asked her attacker what she should call him; he said Daniel. When she shortened his name to Dan or Danny, he objected. At the end of the assault, the rapist emptied her handbag of any valuables, telling her that he was usually a thief. He made good his escape on foot, and Miss X reported the rape.

During the course of the ensuing day, a team of police officers was assembled and tasked with the investigation. The senior investigating officer was Detective Superintendent Gregg, but it was his deputy, then Detective Inspector, now Superintendent Whitehead, who led the investigation. He kept a policy file of decisions taken. The officer in the case was a Detective Sergeant Malthouse, and the team included a rape victim liaison officer, PC Hodgson. She took an initial statement from X, a copy of which we have been given today, which contained a description of the rapist to this effect:

“A white male. 19 or under. Over 6-foot. Slim. Light brown hair which was cut short under a dark baseball cap with a white symbol on it. Navy blue hooded sweatshirt with big white symbols on the front. Hood down. Clean shaven. Big staring eyes. Dark jeans and a Leeds accent.”

The description of the rapist rang a bell with Police Constable Hodgson. She told Detective Sergeant Malthouse of someone whose description might match; namely, the respondent. She knew the respondent, and her 20 year-old daughter knew the respondent. DS Malthouse passed the information to DI Whitehead. With Superintendent Gregg’s approval, DI Whitehead decided to arrest the respondent. The note of the information upon which his decision to arrest was based is at page 94 of the policy file. Page 94 reads as follows:

“Informed by OIC that PC Hodgson (RVLO) knows the above who is about 6ft and generally fits the description of the offender. He frequents the city clubs in Leeds and has possible been a bouncer although he may be slim/med build having been involved with drugs and theft. He resides near to the attach site within walking distance. PC Hodgson’s daughter 20 years saw above in January this year when he was wearing a black baseball cap with white motif and a black hooded top as described by the complainant. He is referred to as Daniel and didn’t like Bernard or Danny.”

We are told that the reference to PC Hodgson’s daughter’s age had been checked to ensure she was not a child who might have been mistaken in her description.

The extent of the information passed on by PC Hodgson was contentious at trial, but nothing now turns upon it. What matters is that the contents of page 94 created in the mind of DI Whitehead a genuine suspicion that the respondent might be the rapist. He believed there were seven or eight respects in which the respondent possibly matched the offender. He frequented the clubs of Leeds City Centre; he might have been a doorman, and therefore known to other doormen; he had two previous convictions for theft, and he might therefore have described himself as a thief; he lived near the attack; he fitted the offender’s general description; he had worn recently clothing of the kind described by the victim; and he was known as Daniel and not Dan or Danny.

On that basis, at 20.55 on 5 February the decision was taken to arrest the respondent. DI Whitehead explained in evidence that any attempt to arrest him that night might fail, and so alert the suspected rapist to police interest. The decision was therefore taken to make the arrest the following morning, so as to avoid any chance of the suspect absconding or destroying evidence. The next day, Mr Armstrong was arrested at his home. It must have been a very unpleasant experience for him.

It is his description which is of particular relevance to this appeal. He was described as 6 foot or over; he was not clean-shaven, although the extent of his facial hair was a matter of debate; he was 24, about to become 25 years of age; and he was of large build. As Mr Stanage on his behalf properly reminded us, he was not, as the victim appeared to be describing, a clean-shaven teenager of slim build. Mr Armstrong was interviewed, detained in a cell, and intimate samples taken from him. He was released some hours later. Further investigation exonerated him. He sued for battery, false arrest, false imprisonment, trespass to land and goods. With the benefit of a jury’s findings on the evidence, it fell to the trial judge, Mr Recorder Myerson, to decide whether or not the police had reasonable grounds for suspecting the respondent of rape. The parties in the court below and before us have focused on the grounds in the mind of Detective Inspector Whitehead, I assume because it was accepted he had briefed the junior officers who actually made the arrest. The Recorder held that, albeit the investigating officers genuinely believed the respondent was a suspect, they did not in fact have reasonable grounds for their suspicion.

The question for this court is whether the Recorder applied too strict a test. During the course of his judgment, the Recorder conducted a rigorous and painstaking forensic analysis of Detective Inspector Whitehead’s seven or eight points. He picked them off one by one, comparing them with what the officers knew about the respondent by the time of his arrest. He decided, for example, that the fact the respondent frequented the clubs in the centre of Leeds was irrelevant, matching as it did a large number of young people. He rejected the assertion that the respondent might properly be regarded as a thief on the basis of two relatively minor convictions some years before. He excluded from consideration the suggestion that the respondent may once have been a doorman, saying it took the investigation no further; one does not need to be employed as a doorman to talk to a doorman or to obtain a light from one, as the rapist apparently did. The Recorder went further, and observed that the police had not in fact even investigated from the doorman whether the man they spoke to was one of their colleagues.

The Recorder spent much of his ruling considering and comparing the description of the rapist and his clothing, and the respondent’s actual description and the colour of his clothing. Having analysed the information in considerable detail, in his view it came to this: the respondent was called Daniel, and lived within walking distance of the attack. This, he said, was not enough to justify an arrest. He also spent some time analysing the thoroughness of the investigation and the police’s failure, as he saw it, to check their facts overnight on 5/6 February. He made clear findings that the officers had failed to carry out a proper investigation before arresting the respondent, and that they did not have reasonable grounds for so doing.

The Recorder then turned to the law. He asked himself this question in paragraph 36: “is there anything in the law that means that I must in some sense modify the findings of fact?”With respect to the Recorder, it might have been helpful had he directed himself on the appropriate test to be applied at the beginning of his ruling, and then made his findings in the light of that test. He would then have had in the forefront of his mind the agreed and essential principles of law relating to lawful arrest. Those principles include the definition of suspicion affording sufficient grounds for arrest, as provided by Lord Devlin at page 948 of the judgment of the Privy Council in Hussein v Chong Fook Kam [1970] AC. Lord Devlin said this:

“Suspicion in its ordinary meaning is a state of conjecture or surmise where proof is lacking: "I suspect but I cannot prove". Suspicion arises at or near the starting point of an investigation, of which the obtaining of prima facie proof is the end. When such proof has been obtained, the police case is complete; it is ready for trial and passes on to its next stage. It is indeed desirable as a general rule that an arrest should not be made until the case is complete. But if arrest before that is forbidden, it could seriously hamper the police…”

Further on at page 949B, Lord Devlin added:

“There is another distinction between reasonable suspicion and prima facie proof. Prima facie proof consists of admissible evidence. Suspicion can take account matters that could not be put in evidence at all.”

The most recent exposition by the House of Lords of the principles to be applied is to be found in O’Hara v Chief Constable of the Royal Ulster Constabulary [1997] AC 286. Lord Steyn, in giving one of the two leading speeches at page 293C-E, said this:

“Certain general propositions about the powers of constables under a section such as section 12(1) can now be summarised. (1) In order to have a reasonable suspicion the constable need not have evidence amounting to a prima facie case. Ex hypothesi one is considering a preliminary stage of the investigation and information from an informer or a tip-off from a member of the public may be enough:Hussien v. Chong Fook Kam [1970] A.C. 942, 949. (2) Hearsay information may therefore afford a constable a reasonable ground to arrest. Such information may come from other officers: Hussien's case, ibid. (3) The information which causes the constable to be suspicious of the individual must be in existence to the knowledge of the police officer at the time he makes the arrest. (4) The executive ‘discretion’ to arrest or not as Lord Diplock described it in Mohammed-Holgate v. Duke [1984] A.C. 437, 446, vests in the constable, who is engaged on the decision to arrest or not, and not in his superior officers.”

Lord Hope, who gave the only other reasoned speech, agreed, saying at page 298C-E:

“This means that the point does not depend on whether the arresting officer himself thought at that time that they were reasonable. The question is whether a reasonable man would be of that opinion, having regard to the information which was in the mind of the arresting officer. It is the arresting officer's own account of the information which he had which matters, not what was observed by or known to anyone else. The information acted on by the arresting officer need not be based on his own observations, as he is entitled to form a suspicion based on what he has been told. His reasonable suspicion may be based on information which has been given to him anonymously or it may be based on information, perhaps in the course of an emergency, which turns out later to be wrong. As it is the information which is in his mind alone which is relevant however, it is not necessary to go on to prove what was known to his informant or that any facts on which he based his suspicion were in fact true. The question whether it provided reasonable grounds for the suspicion depends on the source of his information and its context, seen in the light of the whole surrounding circumstances.”

In O’Hara, the arresting officer gave as his reasonable grounds for arresting a suspected terrorist the briefing he had received from a senior officer, upon whose orders he was acting. He was not cross-examined at trial on the contents of that briefing. The question certified for their Lordships’ House was whether an order by a superior officer was itself sufficient to afford reasonable grounds for suspecting O’Hara of a terrorist offence. The House held it was not. Nevertheless, the appeal to the Lords failed, on the factual basis that the “scanty briefing” and “sparse materials” available to the arresting officer were sufficient from which to infer the existence of reasonable grounds for suspicion.

Mr Recorder Myerson described O’Hara as authority for but one proposition, namely:

“…providing on the facts of the case a reasonable investigation has been undertaken, reasonable suspicion will be demonstrated, but it follows from that that [a [reasonable] investigation is required.”

If by this statement Mr Recorder Myerson meant that O’Hara is authority for the proposition, absent a life and death emergency, reasonable grounds may only be only be inferred where there has been a reasonably thorough investigation, I beg to differ. That is not what O’Hara says. It was this gloss on O’Hara that meant the Recorder felt able to distinguish the decision in Cummings v Chief Constable of Northumbria Police [2003] EWCA Civ 1844 per Latham LJ at page 41 from the present case. Cummings was one of several suspects arrested on suspicion of an offence which could only have been committed by one or two of them. By analysing work rotas, skill sets and connections with a known offender, the police narrowed the field of their suspects to six employees, each of whom had the opportunity to commit the offence. They were arrested rather than asked to assist the police in their enquiries, simply because the police wanted to put maximum pressure on the guilty party to confess. Nevertheless, the arrests were held to be lawful. The Recorder, however, noted that in Cummings the police officers had done, as he put it, “absolutely everything they could short of arrest”.

For my part, I accept the proposition that the thoroughness of an investigation may well be relevant as part of the whole surrounding circumstances as described by Lord Hope in O’Hara. There may be circumstances, provided there is no urgency, which makes it incumbent upon an officer to make further enquiries before “suspicion could properly crystallise” (for which see paragraph 17 of the judgment of the then Simon Brown LJ in Hough v Chief Constable of the Staffordshire Constabulary [2001] EWCA Civ 39). However, it is important to remember, in my view, that an arrest may be effected very early on in an investigation, and it is nonetheless lawful for that. It will not always be possible or indeed desirable to carry out further enquiries before making an arrest; for example, see again the case of Hough. Mr Hough was arrested simply on the basis that a computer check suggested the occupant of the car in which he was travelling may be armed. The Recorder attempted to distinguish the facts here from Hough on the basis if the police could wait for ten hours to arrest the respondent, there was no emergency.

Finally on the law, this court has had the benefit of the most recent decision of the Court of Appeal judgment delivered after the ruling in this case in Commissioner of the Police of the Metropolis v Mohamed Raissi [2008] EWCA Civ 1237. The facts and the issues in Raissi are not relevant to the instant appeal; however, paragraph 20 of the judgment of the court given by Sir Anthony Clarke MR provides a useful restatement of the proper test to be applied when assessing the requirement for reasonable grounds for suspicion. He said this:

“On the other hand it is important to have in mind that, as the judge held at [47], the threshold for the existence of reasonable grounds for suspicion is low: see eg Dumbell v Roberts [1944] 1 All ER 326 per Scott LJ, where he said at page 329A-B that "the requirement is very limited"; Hussienper Lord Devlin at pages 948G to 949A; and O'Haraper Lord Steyn at page 293C and per Lord Hope at page 296D-E.”

The Recorder may not have had the benefit of that summary, but the cases referred to therein were put before him and he was addressed upon them. In my view, he fell into error by not bearing them sufficiently in mind. As a result, he has pitched the level of the threshold for suspicion too high. His natural sympathy for an innocent man taken from his home on suspicion of a dreadful offence seems to have unduly coloured what was meant to be an objective consideration of whether of not the officers had reasonable grounds to suspect the respondent.

I take Mr Stanage’s point, which is well made, that this court should be slow to interfere with the careful, reasoned findings of a trial judge, who has purportedly applied the correct test in law. However, to my mind, although the Recorder may have stated the test in law correctly, indeed repeated it correctly, he has not here in fact applied it correctly. He has applied it too strictly, with a large degree of hindsight, and he has not looked at the whole surrounding circumstances, as he was obliged to do.

First, in my view, he erred by failing to put DI Whitehead’s decision into proper context; this was an investigation into a vicious sexual assault, there was a rapist on the loose who might strike again at any time, and dispose of evidence. The protection of the public was uppermost in the minds of the police officers, and DI Whitehead, an experienced police officer, genuinely believed, as he put it, “he had got his man”. He had to act swiftly if he was to catch the rapist, prevent another attack, and preserve evidence. The situation in this case may not have been as much of an emergency as presented to the arresting officer in Hough, but it was nevertheless urgent. The fact that DI Whitehead decided to wait for ten hours, for good reason, does not detract significantly from that urgency.

The context and the whole surrounding circumstances here also includes the source of the information. As Mr Skelt submitted, it is important to note that the source of the information was a trained police officer, PC Hodgson, and her adult daughter. As far as the source is concerned, at paragraph 30 of his judgment the Recorder said this:

“In my judgment, what ought to have happened is that Superintendent Gregg ought to have been told, ‘PC Hodgson’s daughter tells PC Hodgson, who tells Detective Sergeant Malthouse, who told me that this is the position’, and had it been written down like that … the police would have been forced to confront reality, which in the end they were not.”

It is not clear to me what reality the police would have been forced to confront. If it was the reality that this was hearsay information, the authorities make plain that hearsay information will suffice. Nor, I confess, is it clear to me why the information, which appeared reliable, that the respondent likes to be known as Daniel, not Dan or Danny, that he fitted the general description, lived within walking distance, and had recently worn dark clothing of a distinctive nature similar to that worn by the rapist, had, as the Recorder described it, “insufficient value”. It was important information from an apparently reliable source. The trained police officer had taken the victim’s statement, and both she and her adult daughter had personal knowledge of the respondent. The fact that PC Hodgson referred to the possibility of Mr Armstrong’s having a slim build, and the possibility of his having once been a bouncer, making that part of the information “conditional” as the Recorder described it, did not undermine the fact that the respondent appeared significantly to match the man for whom the police were looking.

Next, to my mind, the Recorder fell into error by excluding from his consideration some of Detective Inspector Whitehead’s seven to eight points on the basis that he, the Recorder, felt they did not take the investigation much further. DI Whitehead based his decision to arrest on the cumulative effect of all the strands of information before him. It was, therefore, the cumulative effect that the Recorder should have borne in mind. He should not have fallen into the trap of overcompartmentalising the various pieces of information. Individually, some of them may not have appeared particularly significant; but that is often the case. Every strand which matched the suspect narrowed the field down to some extent.

Further, in paragraphs 21 to 25, in my view the Recorder has carried out far too subjective an analysis of the victim’s description of her attacker and how it compared to Mr Armstrong. Midway through paragraph 21, for example, he asked himself the question: “Can a description of bulk be unreliable?”. This passage followed:

“In my judgment, it cannot in this case because the claimant saw the man in the club, in the taxi and attacking her, and he laid on top of her. She, fortunately, protected from identity by calling her ‘X’, makes it clear in the witness statement I have seen that this was not her first sexual encounter. That is a relevant point, in my judgment, because it means that she is able or has experience in judging weight and build in the context of an intimate encounter, and so when she says here ‘slim’, once one thinks it though it must be an accurate description, or at least be taken to be accurate, so, in my judgment, Mr Armstrong’s build then becomes very important.”

I cannot accept the validity of that analysis for three reasons. (1) Given the protection now offered to rape victims in relation to their previous sexual experience, the propriety of this line of reasoning is, at least, questionable. (2) In any event, whether or not a woman who has experience of sexual intercourse would be in a good position to judge the build of her clothed rapist, during an attack at night in which she was knocked conscious and described herself as being “beside herself with terror”, must be a matter of opinion. (3) The Recorder has ignored the evidence of the experienced detectives. They tried to explain to him that they felt unable to place much reliance on some aspects of the victim’s description, because in their experience physical descriptions from traumatised victims may not be accurate. Yet for my part I can find in the Recorder’s ruling no real acknowledgment of this obvious truth. On the contrary, the Recorder placed heavy reliance on the physical description provided by the victim and criticised the police for not doing the same, saying they should have used the contraindications to rule out Mr Armstrong as a subject.

In my view, a proper objective analysis would have proceeded as the police officers proceeded, from an acceptance of the proposition that the human memory in these circumstances may be all too fallible. Eyewitnesses, even sober eyewitnesses, un-traumatised eyewitnesses close to an offender may provide descriptions which are very wide of the mark in terms of an offender’s age, colouring, height, build and physical features. Build, for example, is but a matter of impression. The officers in this case were entitled to bear very much in mind that the victim was drunk, knocked unconscious, terrified, and then traumatised by a brutal sexual assault. She may not have been the most reliable historian on the subject of her attacker’s physical description.

Nevertheless, the Recorder returned repeatedly to the details she provided; for example, the colour of the clothing. The respondent’s clothing was said to bear the same distinctive markings as the rapist, but Ms Hodgson’s daughter had described it as black, not dark blue. Given the circumstances of the attack, at night, with the closeness in colour of black and navy blue, I do not understand the basis for the Recorder’s finding in paragraph 25 that the complainant had shown “the ability to distinguish, contemporaneously, between dark and navy blue” so that the police were obligedto ignore the fact that the respondent had dark clothing with the same distinctive markings.

Further, in my view the Recorder was wrong to find that the arrest should have been called off when the arresting officers saw Mr Armstrong for themselves and were in a position to compare the victim’s description with the respondent’s appearance and voice.

There were indeed contraindications to Mr Armstrong’s being the rapist. He was not, as Mr Stanage emphasises, a slim, clean-shaven teenager with a broad Leeds accent. However, if one accepts as I believe one must the problems with eyewitness descriptions to which I have already referred, it cannot be said that the differences between the respondent’s physical appearance and voice and the rapist’s appearance and voice were so glaring that the arresting officers should have questioned their orders. Second, I consider it unrealistic to suggest, as the Recorder did in paragraph 34, that, having gone to the respondent’s home and found what they thought was their man, because the respondent appeared in court “polite, collected and responsible”, as the Recorder described him, the police should, therefore, have asked him to accompany them to the police station. The police were not investigating a relatively trivial, non-violent crime. They suspected the respondent was a violent rapist, and they wished to detain him, search him, take samples and secure evidence.

Finally, I should mention that I have had considerable difficulty in following the Recorder’s approach to the respondent’s name. At paragraph 26 he considered the fact that a police national computer check showed the respondent’s name as Bernard Daniel Armstrong, with an alias of Bernard Armstrong. In the last few lines of paragraph 26, the Recorder said this:

“Unless it was part of the police’s belief that during previous arrests he had deliberately called himself a false name, it must follow that he referred to himself habitually, to the police’s own knowledge, as Bernard. The police are not required to resolve those two hypotheses in order to act reasonably, in my view. They are, however, required to deal with them and on the evidence before me they did not.”

I confess, I do not follow that passage. But later, the Recorder referred to the fact that the respondent’s girlfriend called him Bernard or Bernie in the presence of the arresting officers. The suggestion, as I understand it, is the police should have known then that they had the wrong man. Again, I cannot accept that proposition. The question was not what the respondent’s girlfriend would call him, or what he called himself when arrested by the police in the past; the question was: did the police have reasonable grounds to suspect that, like the rapist, he preferred on occasion to be called Daniel in its unshortened form? They did. In my view, that was in itself a very significant piece of information, yet the Recorder appears to have left it out of his considerations.

Thus, principally for those reasons, and indeed for others that I do not need to rehearse, I am driven to the conclusion that the Recorder was clearly wrong and erred in law to find that the police had no reasonable grounds for suspecting Mr Armstrong of the offence. I emphasise the word “suspecting”. From an apparently reliable source, the police understood that Mr Armstrong generally fitted the description of the rapist: he dressed like the rapist in similar distinctive clothing; he was a local man who lived within walking distance of the attack, like the rapist; he frequented the city clubs, like the rapist; he might have known the doormen to stop and talk to them, as the rapist may have done; he had been a thief in his time, which is how the rapist described himself; and significantly, like the rapist, he preferred to be called Daniel in unshortened form. Put in the context of an investigation into a very grave crime which called for urgent action, on any truly objective analysis the police here had reasonable grounds to arrest without making further enquiries.

For those reasons, I would allow the appeal. The findings of fact have been made, and in my view, when the proper test in law is applied to those facts, the result must be that the judgment in the court below must be reduced to the agreed sum of ₤50, which was the sum awarded to Mr Armstrong for a separate matter of battery.

I agree.

I have not found this an easy case, though it has been fairly and properly argued by both counsel. Mr Armstrong was an entirely innocent man. There was, undoubtedly, a serious interference with Mr Armstrong’s liberty, and liberty is a matter to be jealously protected by the courts. But lawful arrest is an exception to the guarantee for the right to liberty and security contained in Article 5 of the European Convention on Human Rights. An interference with liberty is justified, therefore, if it is pursuant to a lawful arrest.

For the reasons given by my Lady, Lady Justice Hallett, I agree that on the authorities and in the circumstance of the alleged offence, and the evidence available to the police at the time, there was in the circumstances of this case a lawful arrest. In terms of the authorities, I refer in particular to the helpful recent summary of the authorities in paragraph 20 of the decision in Raissi, to which my Lady has referred.

Moreover, I consider that the Recorder disabled himself from reaching the right conclusion by adopting an overcompartmentalised approach to the individual items of evidence, and that amounts in my judgment to a misdirection in law. There are also instances, as I see it, in which the Recorder overstates the position. For instance, after the police effecting the actual arrest arrived at Mr Armstrong’s home, he contemplated that they might have rung back to Detective Inspector Whitehead and given a description of Mr Armstrong which did not, as my Lady has explained, to all intents follow that which the complainant had given in all respects. He contemplated that they would say that Mr Armstrong did not begin to answer the description in terms of his appearance, his physical bulk or his voice. That puts the matter very highly indeed and fails to take account of the other matters to which my Lady has referred, in particular his address and his clothing, which it was forethought that he might have had. In addition, it seems to me that the Recorder erred in law in failing to give proper weight to his finding, at the very start of his drawing of inferences from the primary facts, that Mr Armstrong fitted the general description of the offender as described by the victim.

Mr Stanage’s most powerful point, as it seems to me, is that when the police arrived at Mr Armstrong’s home, they found that he was, if I may say so, not slim or clean-shaven. But as I have explained, there were other factors connecting Mr Armstrong to the offence and, as my Lady has said, the description given by a victim may, in traumatic circumstances such as occurred to her in this case, not be infallible. His name and his address and what was still thought to be his clothing still were potential links to the offence. There was also, for the reasons that my Lady, Lady Justice Hallett, has given, an urgent situation; and this was not one, therefore, in which there was time to make more of an investigation and more enquiries in addition to those which had already been made.

For these reasons, and those given my Lady, Lady Justice Hallett, I would allow the appeal. It is agreed that a judgment of ₤50 should be substituted, on the footing that the respondent was placed in handcuffs immediately before, and not after, his arrest.