Spicer v The Commissioner of Police of the Metropolis

The article complained of in this libel action was headed “Two guilty of killing a woman while racing their cars”. In the body of the article, the reader was told that the claimant was one of the two racers, and that he had been found guilty by a jury. But the reader was also told that whilst both men had faced a charge of causing death by dangerous driving, as well as one of causing serious injury by dangerous driving, the claimant had been acquitted of both those charges, and convicted of careless driving. It was the other man, Farid Reza, who was convicted of causing death and causing serious injury by dangerous driving. The reader was told that, for his offending, Reza was sentenced to five years and three months’ imprisonment for the killing and three years’ concurrent for the serious injury, as well as being disqualified from driving. The claimant, it was reported, was fined £1,000, given nine penalty points, and ordered to pay costs of £500.

This has been the trial of the issue of meaning, as a preliminary issue in the action. The main issue for resolution, putting it broadly, is whether the headline encapsulates or reflects the natural and ordinary meaning of the article, read as a whole. Put another way, does the article mean that the claimant was one of two found “guilty of killing a woman while racing their cars”? The answer is that it does not. Established legal principle holds that the meaning of a published article or statement must be collected from the article or statement as a whole. The law does not permit a claimant to sue for damages in respect of a headline, however defamatory, if the headline and article are mismatched, and the impact of the headline is contradicted or neutralised by the remainder of the article. That is this case.

The claimant is William Alexander Spicer, described in the Particulars of Claim as a

The defendant is the Commissioner of Police for the Metropolis. She is sued on the basis that she is responsible in law for the operation of the website located at www.news.met.police.uk. This, as its name suggests, is the news website for the Metropolitan police. Judging by the article complained of, it appears to operate in a similar way to an online newspaper. I shall call it the “Met Website”.

The article complained of was published on the Met Website on 26 January 2017. According to the Particulars of Claim, it was also re-published by the defendant, by tweeting links to it.

I have set out in the Appendix to this judgment the full text of the article, which runs to 36 paragraphs in all. The paragraph numbers have been helpfully added by the parties.

The claimant has selected words for complaint “including in particular the headline”. In addition to the headline, the claimant complains of 7 paragraphs of the body of the article. I have underlined all the words complained of.

The case for the defendant is that not only the headline but all the first 23 paragraphs of the article have some relevance. The concluding section of the article, paragraphs [25] to [36], sets out in great detail the content of a victim personal statement made to the Court by the father of the deceased woman. Nobody has suggested that this has any bearing on the meaning of the article for present purposes, but it is right to include it in the Appendix nonetheless.

I have been provided with screen shots of the article as it appeared on the website, with the illustrative photographs. It is clear to any reader of the article that it is a publication made by or on behalf of the Metropolitan police. The screenshots time the publication at 17:21. This appears to have been a short time after the verdicts of the jury, and the sentencing decisions of the Judge, all of which were delivered earlier that same day.

The action was begun by claim form issued just within the limitation period, on 25 January 2018. The Claim Form and Particulars of Claim were served on 24 May 2018, just within the four-month limit provided for by the CPR. On 12 July 2018, Master Kay QC made an order by consent that there be tried as a preliminary issue “the question of what meaning the words in the publication complained of at paragraph 5 of the Particulars of Claim bear of the claimant.” It has been agreed that I should determine what natural and ordinary defamatory meaning is borne by the words.

The meaning attributed to the words complained of by the claimant was first set out in a letter dated 13 November 2017, complaining that he had been very seriously libelled by the Article in the following meaning:

“William Spicer and Farid Reza have been found guilty by a jury of unlawfully killing a young woman pedestrian, Hina Shamin, at about 9pm on 31^st March 2015 while recklessly racing their high performance BMWs against each other at more than 60 mph in a 30 mph zone in Kingston town centre. Spicer also failed to stop and illegally drove on. He was later found by the Police, charged and arrested.”

This is the meaning complained of in the Particulars of Claim, with the immaterial change that the spelling of Mr Reza’s name has been corrected.

The meaning advanced by the defendant in correspondence, and in the application for the determination of the issue at this hearing, was that:

“There are reasonable grounds to suspect that [the Claimant] was involved in racing his high-performance car against another similar vehicle at more than 60 mph in a 30 mph zone in Kingston Town Centre. The other vehicle struck a young woman, Hina Shamin, who was killed instantly, and its driver was convicted of causing death by dangerous driving. The Claimant failed to stop at the site of the collision, and then made an illegal right-hand turn. He was arrested the following day and charged with causing death by dangerous driving, and causing serious injury by dangerous driving. He was found not guilty of those two offences, but guilty of careless driving, and was given nine penalty points on his licence, a £1,000 fine, and ordered to pay £500 costs.”

In the course of argument at this trial, and in response to questions from me, Mr Sterling has expanded on the claimant’s case in relation to arrest and charge. He has submitted that the implication of the report of arrest and charge is that there were reasonable grounds to suspect the claimant of the offences in question. Mr de Wilde has not sought to quarrel with that approach.

The common law test of what is defamatory is not controversial. I summarised it recently in Allen v Times Newspapers [2019] EWHC (QB) [19]:

“(1)

At common law, a statement is defamatory of the claimant if, but only if, (a) it imputes conduct which would tend to lower the claimant in the estimation of right-thinking people generally, and (b) the imputation crosses the common law threshold of seriousness, which is that it “[substantially] affects in an adverse manner the attitude of other people towards him or has a tendency so to do”: Thornton v Telegraph Media Group Limited

[2010] EWHC 1414 (QB) [2011] 1 WLR 1985 [96] (Tugendhat

J).

(2)

“Although the word ‘affects’ in this formulation might suggest otherwise, it is not necessary to establish that the attitude of any individual person towards the claimant has in fact been adversely affected to a substantial extent, or at all. It is only necessary to prove that the meaning conveyed by the words has a tendency to cause such a consequence”: Lachaux v Independent Print Limited [2015] EWHC 2242 (QB) [2016] QB 402 [15(5)].”

In this case, all the meanings proposed by the parties are agreed to be defamatory meanings, by these criteria.

The Court’s function at a trial such as this is to identify “what is the natural and ordinary meaning of the Article, as it relates to the claimant”: Allen v Times Newspapers Ltd [2019] EWHC 1235 (QB) [39]. The principles to be applied are conveniently collected in the recent judgment of Nicklin J in Koutsogiannis v The Random House Group Limited [2019] EWHC 48 (QB) at [11-13]. It is unnecessary to rehearse them all here. They are not controversial in any way. My task is to apply them to the facts of the particular case.

But in every case, some principles will be more pertinent than others. Here, I have had in mind, in particular, principles (viii) and (xi) in the Koutsogiannis canon. Principle (viii), the “bane and antidote” principle, was summarised in this way by Nicklin J:-

“The publication must be read as a whole, and any “bane and antidote” taken together. Sometimes, the context will clothe the words in a more serious defamatory meaning (for example the classic “rogues’ gallery” case). In other cases, the context will weaken (even extinguish altogether) the defamatory meaning that the words bear if they were read in isolation (e.g. bane and antidote cases).”

The time-honoured language of “bane and antidote” has an antiquarian ring to it, but I think the sense of it is still clear enough. It is explained in the classic statement of the principle, that of Baron Alderson in Chalmers v Payne (1835) 2 Cr. M. & R 56, 159:

“[If] in one part of the publication something disreputable to the plaintiff is stated, but that is removed by the conclusion, the bane and the antidote must be taken together.”

Headlines commonly feature in bane and antidote arguments. Experience shows that there is quite often a disconnect between a headline and the body of an article. One reason for that may be that many headlines are written by editors or sub-editors, who aim for something eye-catching and may be less familiar with the nuance of the text than its author(s). A headline can create a libel, even if the text contains none: see Gatley on Libel and Slander 12^th ed para 3.30, text to n 349. That is especially so, when one bears in mind the (reasonable) tendency of ordinary readers to give weight to that which is most prominent, and most negative. But there are cases in which the text neutralises what would otherwise be a libel in the headline - the headline being the poison, to which the body of the article provides the antidote.

The clearest and most famous instance of this last category is the article which was the subject of Charleston v News Group Newspapers Ltd [1995] 2 AC 65. The article, referring to two characters in the TV series “Neighbours”, was headlined “Strewth! What’s Harold up to with our Madge?”, and featured what appeared to be photographs of the one having sex with the other. Later in the article it was made clear that “Harold and Madge’s faces are added to porn actors’ bodies” in a scene from a game. It was conceded that, read as a whole, the article was not defamatory. The claimant sought to recover on the basis that the headline and photographs, considered together but in isolation from the remainder of the article, were defamatory. The issue before the House was whether this could be a legitimate argument. Their Lordships were unanimous that this was impossible, as a matter of principle. Lord Nicholls referred to the single objective standard by which meaning is to be assessed and said (at 74):

“I do not see how, consistently with this single standard, it is possible to carve the readership of one article into different groups: those who will have read only the headlines, and those who will have read further. The question, defamatory or no, must always be answered by reference to the response of the ordinary reader to the publication.

This is not to say that words in the text of an article will always be efficacious to cure a defamatory headline. It all depends on the context, one element in which is the layout of the article. Those who print defamatory headlines are playing with fire. The ordinary reader might not be expected to notice curative words tucked away further down in the article. The more so, if the words are on a continuation page to which a reader is directed. The standard of the ordinary reader gives a jury adequate scope to return a verdict meeting the justice of the case.”

The present case is well on the other side of the borderline.”

Lord Bridge also emphasised the fact-sensitive nature of the bane and antidote principle, observing (at 72-73) that:

“Whether the text of a newspaper article will, in any particular case, be sufficient to neutralise the defamatory implication of a prominent headline will sometimes be a nicely balanced question for the jury to decide and will depend not only on the nature of the libel which the headline conveys and the language of the text which is relied on to neutralise it but also on the manner in which the whole of the relevant material is set out and presented.”

As these passages show, the House of Lords was addressing the bane and antidote issue in the context of trial by jury, at a time when that was the default mode of trial in defamation. The question before the House was whether a reasonable jury, properly directed, might hold the the words complained of to be defamatory. The question of whether a defamatory sting was removed by an “antidote” elsewhere in an article was always regarded as quintessentially a jury question, which would only rarely be resolved in favour of the defendant before trial. Charleston was very much the exception that proved that rule.

Today, thanks to the effective abolition of jury trial by the Defamation Act 2013, a Judge can decide the “jury question”. This does not diminish the need for caution when deciding whether a defamatory headline has, to the ordinary reasonable reader, been neutralised by other elements of an article. But it does mean that the Judge at a trial of meaning is a fact-finder, determining the single meaning of the words complained of, rather than that of a legal gatekeeper setting boundaries on the range of decisions that others might make.

To carry out that fact-finding role, the Judge must apply the established principles to the facts of the case, placing himself or herself in the position of the hypothetical ordinary reasonable reader. I have adopted the standard modern practice of reading the article complained of once to myself and (in the light of these principles) forming a provisional view on meaning, before looking at what the parties have said about it. This has recently been approved by the Court of Appeal as “the correct approach for a judge at first instance”: Tinkler v Ferguson [2019] EWCA Civ 819 [9].

Koutsogiannis principle (xi) concerns the characteristics of the reasonable reader. It was put this way by Nicklin J:

“The hypothetical reader is taken to be representative of those who would read the publication in question. The Court can take judicial notice of facts which are common knowledge, but should beware of reliance on impressionistic assessments of the characteristics of a publication’s readership.”

There has been some debate about the characteristics of the representative reader of the article, on the Met Website. Mr de Wilde has suggested that readers of that Website would be professional journalists treating it as a form of “wire service”, or others with a particular interest in criminal justice. He has invited me to treat the representative reader as someone with an informed interest, professional or otherwise, in the workings of criminal justice, capable of picking out the essential components of the story. Mr Sterling argues that this was a story of obvious human interest, and particular interest to the residents of Kingston and the surrounding areas. The readership may well have included a wide range of others, such as friends, relatives, or fellow students of those directly involved, as well as members of the wider community in which the tragic events unfolded. As is typical in such cases, I have no evidence other than the wording of the article and the screenshots, which are agreed items of evidence. The only safe approach is to assume that the readership includes lay people with no special knowledge that would affect the way they read the words complained of.

In arriving at my final conclusions, I have taken full account of the other written and oral submissions of the parties, for which I thank both Counsel. As I made clear in the course of argument, I have deliberately left out of account any consequences my decision might have for the defensibility of the claim (and I have not read any of the relevant correspondence or other written materials that touch on that question).

In brief summary, the argument of Mr Sterling is that the words complained of are clearly defamatory of the claimant in the meaning pleaded on his behalf. The ordinary reasonable reader would, inevitably, understand that both Mr Reza and Mr Spicer had unlawfully killed Ms Shamin. This is the first and lasting impression on reading the article, he submits. This bane is to be found primarily in the headline, but also in paragraph [1], which refers to the car race “leading to the death of a young woman” and to two men being “found guilty by a jury”. The article contains no satisfactory antidote to these sentences; there is nothing which is sufficient to remove their sting. Indeed, Mr Spicer’s conduct is presented as so closely associated and intertwined with that of Mr Reza that a reasonable reader would conclude that Mr Spicer was at fault for the accident and that he and Mr Reza were equally to blame for causing Ms Shamin’s death. Mr Sterling submits that the impression of gross misconduct by the claimant in regard to the accident is also underlined by the suggestion that he failed to stop but drove on and was only later found, arrested and charged.

Mr Sterling submits that paragraph [4] does not clearly identify either the offences of which the claimant was acquitted, or the offence of which he was convicted. Taking account of the headline and paragraph [1], the reader would take it that the claimant was not just convicted of careless driving, but of causing death by doing so. The article contains a number of additional elements, going beyond the attribution of fault in connection with the death: racing, failing to stop after the accident, arrest and charge. These are separate and independent elements of the single meaning, and all are defamatory, smack of misconduct, and make a contribution to the first strand of meaning complained of.

For the defendant, Mr de Wilde contends that the claimant’s argument suffers from a number of flaws. First, it involves complaint of a meaning which relates to both the claimant and Mr Reza, which is wrong in principle given that a claimant can only complain of an imputation that defames him. But the main defect is said to be that the pleaded meaning “ignores the effect on the Article’s meaning of those words within it which are not complained of”. It is submitted that the excerpts selected for complaint give an impression of the Article which is internally incoherent. Shorn of its context, paragraph [4] contains nothing to identify the two offences of which the claimant was acquitted. Read in context, as it should be, the answer is clear: the claimant was acquitted of both the offences of which Mr Reza was convicted.

Mr de Wilde argues that, given the relatively specialised nature of the website in question, the hypothetical reader would understand that it was reporting on the outcome of criminal proceedings. Any assessment of its meaning would include the key facts material to any such proceedings: the identities of the defendants, the offences with which they were charged, the verdicts of the jury, and the penalty imposed: paragraphs [2-4]. These points apply, he submits, whether or not he is right in his submission that I should attribute particular sophistication to the readership. A meaning which disregards or misrepresents those facts as they are presented by the article is “hopelessly artificial and strained.” This, like Charleston, is a case which is well on the other side of the borderline. The article cannot reasonably be read as suggesting that the claimant was convicted of involvement in the killing of the victim. Its real focus, so far as the claimant is concerned, is on his involvement in racing. It also presents him as having failed to stop, as having been arrested, and having been charged. Those are all separate defamatory elements of the message conveyed.

I focus on what the article means about the claimant, rather than Mr Reza, although the two cannot be entirely separated. I have no hesitation in rejecting the contention that this article means that the claimant was “found guilty by a jury of unlawfully killing” Hina Shamin. In my judgment, no reader of the whole article could reasonably draw that conclusion. This line of argument can only have any traction if passages are artificially selected from the article as a whole, which is what the claimant’s team have done.

There is nothing improper or wrong in principle in selecting words from an article or statement, and complaining of those without setting out the remainder of the published statement. Sometimes it is a few paragraphs or a few words only that refer to a claimant, or have any bearing on the meaning that a reader would take away about the claimant. In this case, the long concluding section of the article is concerned exclusively with the victim personal statement of the father of the deceased. It really has nothing to do with the meaning issue that I am concerned with. In any case, pleaders are understandably wary of setting out more than they need to by way of complaint. None could suppose that artificial selectivity will go undetected.

But four points in particular are clear on the facts of this case. The first is that no reader could identify the claimant as a person defamed by this article unless and until the reader had got as far as paragraph [4]. That is the first time the claimant’s name appears, and it is the only time he is referred to by name as having been convicted of anything. So, it would be impossible for the claimant to complain of the headline and first paragraph in isolation. He has to include complaint of paragraph [4] to get his claim off the ground.

Secondly, paragraph [4] would not and cannot be read or interpreted in isolation from its immediate context, namely paragraphs [2] and [3]. Thirdly, if paragraphs [2-4] are read together, they make clear that Mr Reza and the claimant were both charged with and tried for causing death and serious injury by dangerous driving; that Mr Reza was convicted of both charges; but that the claimant was acquitted of both. Fourthly, there is no reason for the reader to conclude that the claimant was convicted of causing death or serious injury carelessly, or at all. Quite the contrary. The article makes clear it was Reza’s car that caused the death and injury. The claimant’s conviction is said to be for careless driving. That is the ordinary and natural meaning of the words used. That is reinforced by the modest sentence of a fine and penalty points. The disparity with the sentence of five years and three months’ imprisonment imposed on Mr Reza is stark. The reader would not suppose an Old Bailey Judge considered a fine to be a suitable sentence for a careless killing, carried out by racing another car at speeds of nearly 70mph in an urban area.

This is a reasoned approach, but I do not believe it is overly-analytical. It corresponds to my impression on first reading the article, unaided by analytical submissions or unduly careful analysis of my own. It obviously means that the headline jars with the body of the article. As Mr de Wilde acknowledged, on this view, the headline is inconsistent with the text. It is not true that two men were “guilty of killing a woman”. But ordinary reasonable readers would not allow the headline to distort the meaning they took from the article as a whole. Readers are familiar with discordance between headlines and articles. Here, the more detailed explanation of events was in the article itself. It clearly stated that one man had been convicted of killing and another acquitted. Reading the headline and article together a reasonable reader would conclude that the headline was wrong. Some readers might think this odd, or sloppy, and move on. Discordance between headline and text is hardly unknown. But whatever conclusion (if any) the reader drew about the reasons for it, I am confident the reader would not see the claimant as a convicted killer.

The rule that the Court must arrive at a meaning by reference to the whole of an article has been described as artificial, but at the same time necessary to do justice to the defendant: see Cruddas v Calvert [2013] EWHC 1427 (QB) [93] (Tugendhat J). It is artificial, because experience tells us that in practice some people, perhaps many, do not read the whole of a published article. Some readers may in fact take the meaning of an article from the headline alone or the first few paragraphs. If that legal principle can be unfair to claimants, it is not unfair in this case, for the reasons given: the only readers who could identify this claimant as a subject or target of the article would also know the fuller picture presented elsewhere in the article. The law is in step with the realities.

I can deal more shortly with the remainder of the meaning. It is common ground that the article suggests (a) something about the claimant’s participation in racing cars; (b) something about failing to stop after Mr Reza struck Ms Shamin; and (c) something about the claimant’s arrest and charge. The parties’ positions on these elements of the meaning are nowhere near as far apart as their positions in relation to the claimant’s primary meaning.

I accept that the article, read fairly, spells out what happened in a logical and chronological sequence. In my judgment, the defendant’s meaning underplays what the article suggests about racing. The article would leave the reasonable reader in no doubt that the claimant was a participant in a hazardous road race at speeds over twice the legal limit, and that he did so to show off. Both matters are presented as fact: see the headline, paragraphs [1] and [7]. This is bolstered by the reported allegations in paragraphs [10] and [21], which are tantamount to direct allegations, given the absence of any balancing or contrary assertions. I do not agree with Mr de Wilde that the sting of those paragraphs is effectively neutralised by the report of the claimant’s acquittal of dangerous driving. The “racing” allegation (as Mr de Wilde has labelled it) is thus at Chase level one, not Chase level two. The article clearly suggests that the claimant failed to stop but drove past the scene of the accident and away. Finally, I accept the modification of the claimant’s meaning advanced by Mr Sterling. A Chase level two imputation, of reasonable grounds to suspect guilt of causing death and serious injury by dangerous driving, is clearly implicit in the report of the claimant’s arrest, charge, and acquittal.

My conclusion, after reading and hearing argument, is that the article complained of bore the following defamatory natural and ordinary meaning about the claimant’s behaviour on and in relation to 31 March 2015:-

This is close to the meaning I jotted down on my preliminary reading of this article, before reading and hearing argument. The main effect of the argument has been to tease out and refine the separate strands of defamatory imputation.

Costs and any other consequential matters will be dealt with by way of written submissions for which I will set a timetable.

The issue of costs budgeting was not addressed at the time of the consent order for this trial in July 2018, as is now the established practice (see Price v MGN Ltd (Practice Note) [2018] EWHC 3395 (QB) [2019] 1 WLR 1464). A further order to that end was agreed and submitted to the Court, but not until 7 May 2019, some ten months after the order for a preliminary trial. Budgets were submitted. But that did not, in the event, leave time for the Court to address that issue in advance of the trial.

Budgeting cannot be carried out retrospectively, of course so my task when it comes to costs will be to assess the reasonable costs which should be paid by one party to the