Johnson, R (On the Application Of) v Crown Prosecution Service

The Claimant is the widower of Mrs Lauren Johnson who died on 28 October 2016 having been struck by a car driven by the Interested Party Mr George Steele. The Claimant seeks to challenge the decision of the Crown Prosecution Service not to prosecute Mr Steele for causing death by dangerous driving.

On 9 November 2018, HHJ Pelling QC, sitting as a Deputy High Court Judge, considered the claim on paper. He granted an extension of time but refused permission to claim judicial review. The Claimant renewed his application for permission before me. The Claimant also seeks disclosure of medical evidence relating to Mr Steele. I was invited to, and have, considered those applications together. I should also note that, as is usual practice, I have considered the case afresh.

The accident was an unimaginable tragedy for Lauren Johnson’s family. She left behind two very young children. It touched the wider community and concerns about the case have spread well beyond the immediate family. Mr Johnson was supported at court by a large group of people. Everyone concerned conducted themselves with dignity and listened closely to the legal argument throughout.

This claim for judicial review is of the utmost importance to the family. I regard it as significant and sensitive. Although only at the permission stage, I considered it appropriate to allow Mr Sachdeva QC the opportunity to develop the submissions on behalf of the Claimant in far more detail than would normally be the case. Fortunately, my list allowed time for that. I am grateful to Mr Sachdeva QC and Mr Godfrey and their instructing solicitors for the care and attention given to preparing and presenting the case so as to allow for thorough consideration of the issues arising. I am also grateful to Mr McGuinness QC, who appeared for the Defendant, for his concise and well-focused submissions.

Throughout, I have been conscious that although I have heard a significant amount of argument, this remains an application for permission and no more. I have reminded myself that I am not determining substantive arguments at this stage but only considering arguability. That is a relatively low bar, but the permission stage remains important. The court cannot grant permission unless satisfied that the claim is properly arguable on public law grounds. In truth, however strongly the family wish

At this stage, I will deal with the factual circumstances only briefly. There is, of course, much evidence. As with most road traffic accidents, not all the evidence is wholly consistent. This is not the forum or the time for a detailed factual analysis.

The accident was both terrible and unusual. The car hit a van at speed in a residential road. It then continued without slowing, through a junction and mounted the pavement hitting Mrs Johnson from behind. Crash data from the vehicle’s airbag system showed that the car was travelling at 53.4 to 54.7 mph for 4 seconds prior to the collision. The speed limit was 20 mph. On the face of it then this was a clear piece of dangerous driving which caused the death of Mrs Johnson.

However, there is evidence that Mr Steele was “in a daze” or unresponsive at the time of the accident. A paramedic who attended him shortly after the collision suspected that he had suffered some sort of “neurological deficit”. This gives rise to a potential defence of automatism.

This is an area of criminal law that is not entirely straightforward. I note that in 2013, the Law Commission suggested that “the case law on insane and non-insane automatism is incoherent and produces results that run counter to common-sense.”

The family of Mrs Johnson were unhappy that there had been no prosecution. In November 2017, they had a meeting with the Assistant Chief Constable of Merseyside Police. Having been informed that Mr Steele was not to be prosecuted, they had a meeting with the CPS in December 2017. Thereafter, they requested a formal review of the charging decision under the Victims’ Rights to Review Protocol.

On 22 February 2018, a review upheld the decision not to prosecute. By way of very brief summary, the reviewer concluded that the defence of non-insane automatism was available, which could not be disproved. The reviewer referred to medical evidence from Dr Smith, a consultant neurologist. She concluded that the medical evidence did not establish that Mr Steele was suffering a disease of mind such as would be required to give rise to the defence of insane automatism. This decision has subsequently been superseded and is no longer relied upon. Mr McGuinness QC frankly conceded that the February 2018 decision was not defensible in law.

A further review was conducted by Mr James Boyd, a specialist prosecutor from the

Mr Boyd’s legal analysis of the case differed from that of the previous reviewer. However, the outcome was the same in that he supported the decision not to prosecute Mr Steele.

The Code for Crown Prosecutors sets out the Defendant’s policy to be applied when making decisions about prosecutions. The Code in force at the time was that published in 2013. A new Code came into effect in October 2018. Were the decision to be reviewed now, the 2018 Code would apply. Unless expressly stated otherwise, I am referring to the 2013 Code, although it may be that there is little practical difference so far as this case is concerned. There is one relevant point to which I will return shortly.

The Full Code Test has two stages: the evidential stage, followed by the public interest stage.

Mr Boyd concluded that the test at the evidential stage of the test was satisfied but that it was not in the public interest to prosecute where “the most likely outcome is that the jury would return a special verdict of not guilty by reason of insanity.”

Within the reasons for his order, HHJ Pelling QC considered that arguably the analysis in the decision letter was not correct on the basis that a special verdict is not a conviction and therefore the conclusion that the evidential stage was satisfied was questionable. HHJ Pelling QC noted that it was not in the Claimant’s interests to contend that this analysis was flawed. If a conclusion that a special verdict was likely meant that the evidential stage was not satisfied, the public interest test would not be engaged. That does not assist the Claimant’s case. The judge also noted that the Defendant did not seek to uphold the decision on this alternative basis.

In support of the approach taken by Mr Boyd, Mr McGuinness QC referred to the 2018 Code. The evidential stage requires prosecutors to be satisfied that there is sufficient evidence to provide a realistic prospect of conviction. A footnote in the new code provides that for the purpose of the Code “conviction” includes circumstances where someone is likely to be found not guilty on the grounds of insanity. Although Mr Sachdeva QC objected to consideration of the new Code, the reality is that this is not a material change. The footnote merely clarifies how the policy should be applied. Further, it is on proper consideration an entirely rational approach. As Mr McGuinness QC pointed out the prosecution of someone who had an insanity defence to a murder charge could otherwise not proceed so as to seek the necessary hospital order, since the Code mandates not proceeding if the test at the evidential stage is not met. Such an interpretation plainly makes no sense. Mr Boyd’s approach to the two-stage test was therefore correct.

In reality, there was broad agreement between the parties as to the legal framework and both sides agree that Mr Boyd was right to find the test at the evidential stage was satisfied and to move on to the public interest stage. It is his finding that it was not in the public interest to prosecute that is the focus of this claim.

At trial, the burden of proof in relation to the defence of insanity rests on the defence. A defendant must establish the defence on a balance of probabilities. By virtue of s.1 of the Criminal Procedure (Insanity & Unfitness to Plead) Act 1991, the jury cannot return a special verdict of not guilty by reason of insanity except on the written or oral evidence of two or more medical practitioners, at least one of whom is duly registered.

In the circumstances, the CPS had to consider whether a reasonable jury properly directed would be more likely than not to accept, on a balance of probabilities, that the defence is made out. It is acknowledged that a jury could not reach such a conclusion unless there were at least two medical opinions in favour of the defence.

Mr Sachdeva QC took me through the relevant authorities in relation to judicial review of prosecutorial decisions at some length. As Sir John Thomas PQBD said in R(L) v Director of Public Prosecutions [2013] EWHC 1752 (Admin), the relevant law is very clear and uncontroversial. In the context of decisions not to prosecute, the relevant authorities have recently been carefully summarised by the Lord Chief Justice and Jay J in R (Monica) v Director of Public Prosecutions [2018] EWHC 3508 (Admin) and by Hickinbottom LJ in R (Campaign against Antisemitism) v Director ofPublic Prosecutions [2019] EWHC 9 (Admin). It is not necessary, nor would it be helpful for me to include a detailed exposition of the authorities here. The family have seen the care with which Mr Sachdeva QC took me through the caselaw and will know that I therefore have the relevant principles in mind.

The threshold for judicial review of prosecutorial decisions is a high one. This is for good reason; decisions as to prosecutions are vested in the Defendant, a specialist body occupying a special constitutional position. The power of this court to review a decision is to be used sparingly and the authorities make it plain that it will be very rare for judicial review of a decision by the Defendant to succeed. The Right to Review Scheme introduced in 2013 has made it even less likely that this court will be required to intervene. In short, this is an area in which the court will exercise considerable restraint and will certainly not substitute its own judgment for that of the specialist prosecutor. However, decisions of the Defendant remain amenable to judicial review on strict public law grounds, as set out in R v DPP ex parte Chaudhury [1995] 1 Cr. App R 136, namely if the decision was arrived at by (1) pursuing an unlawful policy; (2) failing to act in accordance with the policy set out in the Code or (3) if the decision was perverse. The right to seek judicial review remains particularly important in cases involving decisions made on review not to prosecute since there is no alternative remedy. By contrast, a decision to prosecute may be challenged through the trial process.

At this stage, all that the Claimant need show to justify the grant of permission is that he has an arguable claim with a realistic prospect of success. The bar is no higher than that. Of course, in considering whether the claim is arguable, I must have in mind the appropriate threshold for a claim of this nature to succeed at the substantive stage.

In relation to the application for disclosure, the starting point is that disclosure is not required in claims for judicial review unless the court orders otherwise. Public authorities have a duty of candour to the Court and to the Claimant. In R vLancashire County Council ex parte Huddleston [1986] 2 All ER 941, Parker LJ said:

“… when challenged they should set out fully what they did and why, so far as is necessary, fully and fairly to meet the challenge. In doing so, they will, in my view, be making full and fair disclosure and putting the cards face up on the table …”

The test as set out in Tweed v Parades Commission for Northern Ireland [2006]

Before turning to the specific grounds raised by the Claimant, it is likely to be helpful to consider what lies at the heart of Mr Boyd’s decision.

Mr Boyd considered all the available evidence in relation to the accident itself. This included eyewitness accounts; “dash cam” footage and a forensic collision investigation report. He also considered evidence about Mr Steele’s presentation immediately after the collision, including accounts from the emergency services. From this review and leaving aside the medical evidence, he concluded that there was strong support in the evidence for Mr Steele having no control over his driving from moments before the first collision.

That is a conclusion with which the Claimant fundamentally disagrees. Within Ground 4, the Claimant advances a number of points which he asserts point the other way.

It is important to stress that this court, dealing with a claim for judicial review, is not tasked with weighing up all the evidence and reaching its own conclusions. I understand that further evidence has been obtained within the civil claim for damages under the Fatal Accident Act. Nothing within this judgment should be taken to reflect upon such a claim which will involve an entirely different assessment. It may well be that, in due course, a judge considering all the evidence then available will reach a different conclusion. That does not mean that Mr Boyd’s conclusion would have been shown to be wrong, still less unlawful. Mr Boyd was required to consider what a jury was likely to make of the relevant evidence in a criminal trial. Mr McGuinness QC makes the valid point that the parties to such a trial would be the Prosecution and Mr

For the purposes of this claim, Mr Boyd reached a factual conclusion that he was entitled to reach on the evidence. That cannot properly be challenged before this court.

Having addressed the issue of loss of control first, Mr Boyd then considered the cause of the loss of control. He directed himself correctly that the burden of proving insane automatism rests on the defence and that an accused can only discharge that burden at trial on the evidence of two medical practitioners. He recorded that he was alive to the reality that this type of defence was susceptible to being used fraudulently.

Mr Boyd considered the available medical evidence, in the form of expert opinion evidence of Dr Smith, consultant neurologist, and an MRI scan of Mr Steele’s brain. He noted that the scan showed evidence of hardening of the arteries and infarcts of the brain. This hardening of the arteries is the most common cause of seizures in people of Mr Steele’s age. Mr Boyd concluded that the most likely outcome was that Mr Steele would be found not guilty by reason of insanity at trial. In expressing himself in that way, Mr Boyd has answered the ultimate question: Would a reasonable jury, properly directed, be more likely than not to accept on a balance of probabilities that the defence was made out?

Given Mr Boyd’s express recognition of the burden of proof and the need for two medical opinions to discharge that burden, I entirely agree with Mr McGuinness QC’s suggestion that Mr Boyd’s conclusion represents his view that the opinion of Dr Smith could be relied upon by Mr Steele in support of his defence and that it could be anticipated that at least one other doctor would be likely to say the same. Making an assessment of that nature fell squarely within the role entrusted to the Defendant.

This highlights that the evidence of Dr Smith played an important part in the decision. However, Mr McGuinness QC is right to say that it is not the case that the entire decision turned upon this medical evidence. The medical evidence has to be viewed in the context of the other evidence and the conclusion on loss of control.

The structure of the decision-making, approaching the evidence as to loss of control first and then considering the cause of that, cannot be challenged. It was a perfectly rational approach to take.

Having identified the importance of Dr Smith’s evidence, I turn to the Claimant’s application for disclosure. He seeks disclosure of Dr Smith’s medical report(s), any other relevant medical report and letters of instruction on the basis that “The disputed matters surrounding this report are central to the resolution of the Claimant’s claim for Judicial Review.”

Although I am dealing with this application before addressing the question of permission, I have had the benefit of hearing the arguments on both applications so that I reach my conclusions on disclosure knowing the issues to be addressed.

There is guidance for the police and the Defendant about handling requests for disclosure from third parties where there is a possibility of civil litigation after a road traffic collision. While making it clear that each case is to be considered individually, usually witness statements and the forensic collision investigator’s report will be disclosed. However, reports and material provided by medical practitioners should not be disclosed. The disclosure provided to the Claimant accords with this guidance.

It is tempting to think that the evidence from Dr Smith should be disclosed to enable the Claimant’s representatives and the court to form their own view on the interpretation of the medical evidence. However, that is to misunderstand the function of the court and the nature of judicial review.

What is required of the Defendant is identification of the facts and the reasoning process leading to the decision. Mr Sachdeva QC relied upon R (Hoareau andBancoult) v Secretary of State for Foreign and Commonwealth Affairs [2018] EWHC 1508 (Admin) at [24] for the proposition that “where a public body relies on a document as “significant” to its decision, it is ordinarily good practice to exhibit the document”. However, the test remains as set out in Tweed (see above).

In R (S) v Crown Prosecution Service [2015] EWHC 2868 (Admin), Sir Brian Leveson, PQBD said [25]:

“Where (as here) the issue is whether the decision of the CPS was one open to a reasonable prosecutor and the decision-maker has provided evidence of the basis for her decision, the interests of justice do not require further disclosure in order to assess the reasonableness of the decision.”

The Defendant has provided evidence of the basis for the decision here. The decision letter explains the reasoning based upon the medical evidence. The medical evidence was then summarised in greater detail in a letter from the CPS to the Claimant’s solicitors dated 2 August 2018. That letter sets out the key points emerging from the medical evidence. It includes a direct quotation as follows:

“It is hard to be certain about the nature of the event but there can be no doubt that the patient’s loss of consciousness at the wheel was likely to be as a result of a medical episode.”

There is no reason to doubt the summary provided by Mr Boyd. Mr McGuinness QC provided additional reassurance that the duty of candour had been complied with. He confirmed to me that he had seen the medical evidence and was able to say that it was a fair summary.

Mr Sachdeva QC objected that this letter amounted to ‘ex post facto reasoning’. It does not. Mr Boyd had explained his reasoning with reference to the medical evidence in his original decision. In light of the request for disclosure and in compliance with the duty of candour, he provided clarification and further detail of the contents of the medical evidence. That was what was required.

I am entirely satisfied that the Defendant has complied with the duty of candour and that disclosure of the medical evidence itself is not necessary to resolve the matter fairly and justly. I therefore refuse the application for disclosure.

The Claimant seeks to advance four grounds. It is claimed that the Defendant erred in:

Grounds 2, 3 and 4 overlap to a significant extent. The essential question is whether

I have already said that Mr Boyd was entitled to reach the conclusion that he did in relation to loss of control and that it was rational to consider that first before considering the medical evidence.

I have listened carefully to Mr Sachdeva QC’s submissions that the decision was “deeply flawed” as a result of misinterpretation of the medical evidence. He put those submissions most forcefully. However, the sort of detailed forensic analysis of the

I have looked very carefully at the whole of the decision letter. Care must be taken not to take individual phrases and passages out of context.

Mr Sachdeva QC is very critical of the use made of the MRI scan. Mr Boyd said:

“The significance of this evidence is that not only does the eye-witness and forensic evidence provide strong support for S having lost control, but the medical evidence has also identified physiological support for S having lost control.”

Mr Sachdeva QC says that the MRI is not in fact supportive evidence, still less cogent supportive evidence. It is in fact neutral.

Looked at in isolation, that is undoubtedly right. All the MRI showed was that there were signs that were consistent with those that may be seen in a man of Mr Steele’s age who has experienced seizures. There is evidence of strokes, but no way of saying whether such had been symptomatic. Of course, the MRI could not possibly have demonstrated that Mr Steele had suffered a seizure or loss of consciousness at any time, still less at the moment in question. However, it is another piece of evidence that had to be put into the balance.

I do not accept that it is arguable that Mr Boyd’s reasoning was irrational or that it was based on a misunderstanding of the medical evidence. On the contrary, his careful analysis of the totality of the evidence was an entirely rational approach. With

For this reason, the Defendant was not required to ask whether there was any medical evidence, taken alone, that Mr Steele was unconscious at the relevant time immediately prior to the collision. All the evidence had to be pieced together to ask whether it was likely that a jury would find that the loss of control had a medical basis.

The conclusion that Mr Boyd reached was one that was properly open to him on the basis of all the evidence, including the medical evidence as summarised in the letter of 2 August 2018.

As I have already indicated, there are arguments that might be pursued in another context to counter a finding of loss of control. There are undoubtedly also arguments that might be made about the medical evidence and the weight to be given to it.

It follows that I conclude that grounds 2, 3 and 4 are not properly arguable.

The effect of this is that Ground 1 must be considered on the basis that Mr Boyd was entitled to conclude, as he did, that the likely outcome at trial would be a special verdict of not guilty by reason of insanity. His decision at the public interest stage was made in that context.

In the decision letter, Mr Boyd said that “the sole public interest consideration that arises is that the public be protected from a recurrence of S’s dangerous driving”. Taken out of context, that might seem wrong. The Code identifies factors that are to be considered at the public interest stage and states that prosecutors should consider each of the questions set out in paragraphs 4.12 a) to g). Those questions include the seriousness of the offence; the level of culpability (including consideration of mental ill health); the harm caused to the victim; the impact on the community; proportionality and other factors not of direct relevance here.

It is right that Mr Boyd did not go through each of the questions in a mechanistic way. However, he was not required to. The Code makes it clear (at paragraph 4.10) that the questions are not exhaustive and that not all questions will be relevant in every case. It continues:

“The weight to be attached to each of the questions, and the factors identified, will also vary according to the facts and merits of each case.”

In this case, the public interest test was being applied in an unusual context, because it fell for consideration even though it had been concluded that a “conviction” in the true sense was unlikely. Properly, the Claimant does not challenge the view that a court would not impose a hospital order or supervision order in this case and therefore an absolute discharge would follow a special verdict.

It is perfectly plain that the Defendant did not ignore the seriousness of the offence or the impact on the family and wider community. But for the likelihood of a special verdict, there would have been no doubt that prosecution was in the pubic interest. In those particular circumstances, it was reasonable for Mr Boyd to focus his consideration of the public interest in the way that he did. He was right to think about the protection of the public and to address the question of disqualification. It might be thought unfortunate that the law does not provide for disqualification following a special verdict in circumstances such as this. However, Mr Boyd could only operate within the framework of the law and he did so appropriately.

The law does not require public authorities to adopt a mechanistic approach to decision-making. As the Lord Chief Justice said in Monica v DPP (supra) at [46(3)]:

“decision letters should be read in a broad and common sense way, without being subjected to excessive or overly punctilious textual analysis.”

I conclude that Ground 1 is not arguable.

Looked at in a broad common-sense way, the decision contained in the letter of 7 June 2018 was rational and lawful. The conclusion that prosecution was likely to result in a special verdict of not guilty by reason of insanity and an absolute discharge was one that was reasonably open to the Defendant on the evidence. The Defendant has complied with the duty of candour and further disclosure is not necessary to dispose of this matter fairly and justly. The Claimant’s grounds are not properly arguable.