George (Respondent) v Cannell and another (Appellants)

LORD LEGGATT (with whom Lord Hodge and Lord Richards agree):

Introduction

1. We are concerned on this appeal with a tort with many names. It embraces actions which have variously been called slander of title, slander of goods, disparagement of goods and trade libel. By the turn of the twentieth century these actions were coming to be seen as examples of a more general wrong, for which Sir John Salmond coined the name “injurious falsehood”: see Salmond, The Law of Torts: A Treatise on the English Law of Liability for Civil Injuries, 1st ed (1907), p 417. That name is still used by many legal writers; but courts in England and Wales have generally preferred the term “malicious falsehood”, which I will use. Whatever name is used, the nature of the wrong is not in doubt. As stated by the Court of Appeal in the leading case of Ratcliffe v Evans [1892] 2 QB 524, 527:

   “an action will lie for written or oral falsehoods … where they are maliciously published, where they are calculated in the ordinary course of things to produce, and where they do produce, actual damage …”

1. Equally clear is that “actual damage” in this context means pecuniary damage - that is, loss that can be estimated in money (rather than merely being compensated by an award of money). A more modern term which I will use to mean the same thing is “financial loss.” Because financial loss is the basis or “gist” of the tort, malicious falsehood is generally classified as an “economic tort”: see eg Hazel Carty, An Analysis of the Economic Torts, 2nd ed (2010), p 1.

1. In this case the trial judge found that the first defendant maliciously published falsehoods about the claimant to two individuals; but also that neither publication caused the claimant any financial loss. The claimant asserts that the publications nevertheless caused injury to her feelings for which she is entitled to compensation. Admittedly under the common law such a claim cannot be maintained. But the claimant argues that a statutory modification of the tort made by section 3(1) of the Defamation Act 1952 changed this. She contends that this statutory provision applies here and, where it applies, enables damages to be recovered for injury to feelings even when no financial loss has been sustained.

1. If section 3(1) of the 1952 Act has this radical effect, it had gone unnoticed for 70 years. The Court of Appeal, however, accepted that it does. They decided, first, that section 3(1) applies here but that, as the claimant suffered no financial loss, only nominal damages could be awarded on that account. I agree with this. They then held that it would be open to the judge, on an assessment, to award substantial (as opposed to nominal) damages to the claimant for her injured feelings. With that, I cannot agree. As I will explain, it was neither the intention nor the effect of the 1952 Act to transform an economic tort into one which protects the claimant’s emotional wellbeing.

The facts

1. The second defendant, LCA Jobs Ltd (“LCA”), is a recruitment agency owned and operated by the first defendant, Linda Cannell. The claimant, Fiona George, was employed by LCA as a recruitment consultant. She resigned after eight months and then got a job at another recruitment agency called Fawkes & Reece. Her contract of employment with LCA did not prohibit her from soliciting business from LCA’s clients after her employment ended - although she gave Linda Cannell verbal assurances that she would not do so.

1. Fiona George started her new job with Fawkes & Reece at the beginning of January 2019. Right away she began actively targeting LCA’s clients. Linda Cannell soon found out. On 21 January 2019 Ms Cannell emailed the claimant threatening to take legal action against her for breach of “your post-employment obligations under the terms of your employment, not to solicit business from LCA clients.” Ms Cannell also said that she would be writing to the claimant’s employer and contacting LCA’s clients “to advise them of your actions and your violation of the terms of your post-employment obligations.”

1. The trial judge found that, when she sent this email, Linda Cannell knew that the claimant’s contract of employment contained no restriction on soliciting business from LCA’s clients. But together with her legal adviser she decided to assert that there was such a legal obligation. Linda Cannell believed that Fiona George did not have the handbook containing the terms of her employment with LCA and hoped that she would not discover the reality of the situation.

1. No claim for defamation or malicious falsehood could be based on the email sent to the claimant because it was not published to any third party. But immediately before and after sending the email, Linda Cannell made similar statements to two other people.

1. The first such statement was made to an individual called Matthew Butler who worked for a client of LCA. Mr Butler’s firm had been approached by the claimant to use her services to search for new staff. During a telephone call Linda Cannell told Matthew Butler that in doing this Fiona George was breaking her contract with LCA under which she had agreed that she would not approach LCA’s clients.

1. Linda Cannell also sent an email to the claimant’s line-manager at Fawkes & Reece called Graeme Lingenfelder. The email said that Fiona George had been approaching LCA’s clients for new business in breach of “her legal obligations under the terms of her employment with LCA, not to solicit business from our clients and candidates (and Fiona’s absolute assurances that this is something she would never do).” Ms Cannell asked for assurances that this would stop immediately.

1. Very shortly after receiving this email, Mr Lingenfelder spoke to the claimant. They discussed the allegation of breach of contract. Contrary to Linda Cannell’s belief, Fiona George did in fact have a copy of the handbook containing her terms of employment with LCA. She showed this to Graeme Lingenfelder so that he could see that, in reality, there was nothing in her contract to prevent her from soliciting LCA’s clients. He accepted that this was so.

1. A few days later, on 27 January 2019, the claimant resigned from Fawkes & Reece, despite efforts by Mr Lingenfelder to persuade her to stay. She did so because she supposed (wrongly, as it turned out) that Linda Cannell had carried out her threat to contact other clients of LCA. Fiona George felt that this made her position untenable. She quickly found another job with a recruitment agency operating in a different sector.

The proceedings below

1. Fiona George sued Linda Cannell and LCA for both defamation and malicious falsehood. The claim for defamation failed at trial because the judge, Saini J, found that the statements made by Linda Cannell to Mr Butler and to Mr Lingenfelder had not caused serious harm to the claimant’s reputation, as is now required to establish that a statement is defamatory by section 1 of the Defamation Act 2013: see [2021] EWHC 2988 (QB); [2021] 4 WLR 145, paras 133-136. There was no appeal from that decision.

1. On the claim for malicious falsehood the judge found that the statements made by Linda Cannell to the effect that the claimant had breached post-employment obligations in her contract with LCA were false and were made maliciously, as Linda Cannell did not honestly believe that they were true: see paras 160-163. He also found that the statements had not caused any financial loss at all to the claimant. What Linda Cannell had said to Matthew Butler had no financial impact because Mr Butler had in fact already decided not to deal further with the claimant due to an unrelated issue (involving a disagreement about commission). Equally, no loss flowed from the email sent to Graeme Lingenfelder because, straight after receiving the email, he saw for himself that the claimant’s employment contract contained no relevant legal obligation. Any restrictions that he imposed on the freedom of the claimant to contact LCA’s clients were therefore his own decision and were not affected by the false statement: see paras 180-181.

1. The judge was persuaded that, in the light of his findings that the statements complained of did not cause any financial loss, the claim for damages for malicious falsehood failed under the common law and that section 3(1) of the 1952 Act did not apply. He therefore dismissed the claim.

1. On appeal the Court of Appeal (Warby LJ, with whom Underhill LJ and Snowden LJ agreed) decided that the claim does fall within section 3(1) with the consequence that the claimant is entitled to a judgment for damages to be assessed: see [2022] EWCA Civ 1067; [2023] QB 117, paras 72-73. They also decided that, even though the publications complained of caused the claimant no financial loss, she is not limited on that account to an award of purely nominal damages but is entitled to recover compensation for injury to her feelings: paras 74-79. The Court of Appeal ordered that the case be remitted to the High Court to assess these damages.

The issues on this appeal

1. On this further appeal the defendants dispute each step of the Court of Appeal’s reasoning. Their case is put in two alternative ways. They first argue that, on the facts found by the judge, and in particular his finding that the two publications caused no financial loss to the claimant, the claim must fail altogether. Alternatively, they argue that, if the claimant is entitled to a judgment in her favour, it can only be for nominal damages and not for damages for injured feelings. The first of these arguments turns on the proper interpretation of section 3(1) of the 1952 Act, to which I now turn.

Section 3(1) of the 1952 Act

1. Section 3(1) of the 1952 Act states:

   “In an action for slander of title, slander of goods or other malicious falsehood, it shall not be necessary to allege or prove special damage –

   1. if the words upon which the action is founded are calculated to cause pecuniary damage to the plaintiff and are published in writing or other permanent form; or

   1. if the said words are calculated to cause pecuniary damage to the plaintiff in respect of any office, profession, calling, trade or business held or carried on by him at the time of the publication.”

1. In construing section 3(1), section 2 is also relevant. This states:

   “In an action for slander in respect of words calculated to disparage the plaintiff in any office, profession, calling, trade or business held or carried on by him at the time of the publication, it shall not be necessary to allege or prove special damage, whether or not the words are spoken of the plaintiff in the way of his office, profession, calling, trade or business.”

1. These provisions are largely unintelligible to anyone untutored in the technicalities of the common law of defamation and malicious falsehood. The term “special damage” is not one used in ordinary speech. Nor, unfortunately, does it even have a uniform legal meaning. As Bowen LJ pointed out in Ratcliffe v Evans, at p 528, the term “special damage”, although “found for centuries in the books, is not always used with reference to similar subject-matter, nor in the same context”; see also JA Jolowicz, “The Changing Use of ‘Special Damage’ and its Effect on the Law” [1960] CLJ 214. For that reason, the drafter of the Act might have done well to follow the example of the Court of Appeal in Ratcliffe v Evans in avoiding the term because of its potential “to encourage confusion in thought” (see p 529).

1. There is another difficulty in the drafting of these provisions. Both section 2 and section 3(1) state that, where they apply, “it shall not be necessary to allege or prove special damage.” Yet they do not state for what purpose this shall be unnecessary.

1. These two provisions of the 1952 Act only make sense when interpreted against the background of the common law which the Act was designed to modify. So I need to outline the relevant background to these provisions before I examine their meaning and effect.

The common law background: defamation

1. The term “special damage” derives its original meaning from a feature that is fundamental to the common law of defamation. This is the distinction between statements actionable “on proof of special damage” and statements actionable “without proof of special damage” (also known as statements actionable “per se”). Understanding this distinction requires some knowledge of its history, as its existence and significance are largely a historical accident. The following summary of the relevant history is drawn from the valuable accounts given by: David Ibbetson, A Historical Introduction to the Law of Obligations (1999), pp 112-125; Paul Mitchell, The Making of the Modern Law of Defamation (2005), chs 1 and 3; and John Baker, An Introduction to English Legal History, 5th ed (2019), ch 25.

1. In mediaeval England remedies for defamatory words could be obtained only in the ecclesiastical courts, which could order penance but not damages. The common law courts began to permit a general action on the case for words in the first two decades of the sixteenth century. The essence of the action was not injury to reputation as such but the effect of the words in causing quantifiable economic loss.

1. The earliest actions were brought for accusations of theft and other criminal offences. The next situation where economic loss regularly gave rise to such actions was where the words affected the claimant’s income from a profession, trade or calling. A third special category of actions which developed in Elizabethan times concerned imputations of French pox (syphilis) and was later extended to other contagious diseases. Where an allegation fell within one of these three categories, the courts would presume damage from the nature of the words used and it was then left to the jury to decide the amount of damages. In such cases the words were said to be actionable in themselves or “per se.” In other actions on the case for words “special damage” had to be proved.

1. The distinction was cemented by the Limitation Act 1623. Section 3(4) of this Act introduced a special limitation period for actions for slanderous words of two years from when the words were spoken. This was interpreted by the courts as applying only if the words were actionable per se. If proof of special damage was required, time did not start to run until the damage occurred, and the normal six-year period applied: see Saunders v Edwards (1662) 1 Sid 95; Littleboy v Wright (1662) 1 Sid 95.

1. In the eighteenth century the courts introduced a new restriction: that for words to be actionable they should not only cause loss but should also be capable of bearing a defamatory meaning. The test adopted, taken from the criminal law of libel, was that the words must expose the claimant to “hatred, contempt or ridicule.” The action for “words” thus became the action for “defamation.”

1. Originally the publication of defamatory words in writing was regarded simply as a form of slander. The distinction between slander and libel appears to have developed from an undefined fourth category of words actionable per se by reason of their being particularly malignant or widely disseminated. Widespread distribution of printed words became the paradigm case. Eventually written words were seen as constituting the entire category. The distinction between slander and libel was entrenched by the decision of the Court of Common Pleas in Thorley v Kerry (1812) 4 Taunt 355. Giving the judgment of the court, Sir James Mansfield CJ said that he could not, in principle, see any difference between written and spoken words that would allow an action to be maintained for written words when an action could not be maintained if the words were spoken; but that the distinction was too well established by authority for it now to be repudiated.

1. In the nineteenth century the categories of defamatory statements actionable per se came to be seen as closed to further judicial development so that only Parliament could alter them. This attitude was encapsulated by Pollock CB in Gallwey v Marshall (1853) 9 Exch 294, 300, when he said that “we ought not to extend the limits of actions of this nature [ie where proof of special damage is not required] beyond those laid down by our predecessors.” Parliament did create one new category. The Slander of Women Act 1891 provided that spoken words “which impute unchastity or adultery to any woman or girl shall not require special damage to render them actionable.” (This Act was repealed by the Defamation Act 2013.)

1. A limitation in the category of statements actionable per se relating to a person’s fitness in their profession, trade or calling was highlighted by the decision of the House of Lords in Jones v Jones [1916] 2 AC 481. The claimant was the headmaster of a school who was orally accused of adultery with the wife of a school cleaner. Although the allegation was calculated to damage the claimant’s reputation in his profession and, as the jury found, put him at risk of being dismissed from his employment, the claimant did not in fact lose his job and no special damage was shown. Following a trial, judgment was entered for the claimant for damages assessed by the jury in a sum of £10. But on appeal the judgment was reversed on the ground that the words spoken were not actionable without proof of special damage. The House of Lords held that the relevant category of slander actionable per se was restricted by precedent to defamatory words spoken of someone “in the way of” their profession or calling. As the allegation made did not relate to the claimant’s work as a school master but to his private life, this requirement was not satisfied.

1. All the law lords said that any change in the law could be made only by the legislature: see p 493 (Viscount Haldane), p 499 (Lord Sumner), p 506 (Lord Parmoor), and p 508 (Lord Wrenbury). Ironically, the very illogicality of the law was seen as a reason why the courts could not change it. Viscount Haldane, at p 489, approved this passage in the judgment of Lord Herschell in Alexander v Jenkins [1892] 1 QB 797, 801:

   “When you are dealing with some legal decisions which all rest on a certain principle, you may extend the area of those decisions to meet cases which fall within the same principle; but where we are dealing with such an artificial law as this law of slander, which rests on the most artificial distinctions, all you can do is, I think, to say that if the action is to be extended to a class of cases in which it has not hitherto been held to lie, it is the legislature that must make the extension and not the Court.”

   The legislature did eventually extend this category of slander actionable per se by section 2 of the 1952 Act.

1. The ability to recover damages for any written and some spoken words tending to injure the reputation of the claimant without having to prove special damage allowed the gist of the tort in such cases to come to be seen as injury to reputation rather than financial loss. Until comparatively recent times, damages were assessed by juries and the courts were very reluctant to interfere with a jury’s award. In practice, therefore, juries had a broad discretion to award whatever amount of money they thought fit compensation for a form of injury (to reputation) which is inherently non-financial.

1. By contrast, the “special damage” which a claimant must show to found an action for slander where the words are not actionable per se is still limited - as it has always been - to financial loss: see Lachaux v Independent Print Ltd[2019] UKSC 27; [2020] AC 612, paras 5, 15, 18. In some old cases the expressions “temporal” or “material” loss were used, but those expressions were synonymous with financial loss as they also meant that the loss was capable of being estimated in money: see Chamberlain v Boyd (1883) 11 QBD 407, 412 (Lord Coleridge CJ); McGregor on Damages, 21st ed (2021), para 46-003; Gatley on Libel and Slander, 13th ed (2022), para 6-002. In Lachaux, para 5, Lord Sumption summarised the position in this way:

   “The interest which the law protects in cases where a defamatory statement is actionable per se differs from that which it protects in other cases. The gist of the tort where the statement is not actionable per se is not injury to reputation but … wrongfully inflicted pecuniary loss.”

Lord Sumption added that it is an open question, which has given rise to conflicting dicta, whether non-pecuniary damage is recoverable at all when proof of special damage is required.

The common law background: malicious falsehood

1. The earliest actions for malicious falsehood were framed as actions for slander. A false allegation that the claimant did not have a valid title to land, which prevented him from selling or leasing it, was called “slander of title.” False statements about the claimant’s wares which caused loss of custom were also characterised as a type of slander. When slander became confined more narrowly to words which defamed the claimant’s person or character rather than their property, these actions were seen as having a different basis and as requiring proof of actual malice as well as special damage.

1. The question whether it was necessary to show special damage even when the publication was in writing was authoritatively decided in Malachy v Soper (1836) 3 Bing NC 371; 132 ER 453. Tindal CJ, giving the judgment of the Court of Common Pleas, observed that the publication in that case was “one which slanders not the person or character of the Plaintiff, but his title [to certain shares in a mine].” It was therefore not an action of a type in which “no special damage need either be alleged or proved.” After examining earlier cases, he concluded, at pp 383-384:

   “We hold, therefore, on the authority of these cases, that an action for slander of title is not properly an action for words spoken, or for libel written and published, but an action on the case for special damage sustained by reason of the speaking or publication of the slander of the Plaintiff’s title.”

1. The question of what constituted special damage for the purpose of the tort of malicious falsehood was closely examined in Ratcliffe v Evans [1892] 2 QB 524. The claimant in that case was an engineer and boiler-maker. He sued the owner of a newspaper for falsely and maliciously publishing an article stating that he had ceased to carry on his business. At the trial the claimant proved a general loss of business since the publication but gave no specific evidence of the loss of any particular customers or orders. Judgment was given for the claimant for damages of £120. The defendant appealed on the ground that no special damage had been alleged and proved.

1. The Court of Appeal dismissed the appeal. The impressive judgment of the court, delivered by Bowen LJ, carefully disentangled the different meanings of the term “special damage.” Bowen LJ noted that the term had been used in various senses, and that its meaning depended on the context. In its original meaning “special damage” referred simply to the damage which a claimant had to show to found a claim in cases where damage was the basis or gist of the action. As Bowen LJ explained, at p 528:

   “… where no actual and positive right (apart from the damage done) has been disturbed, it is the damage done that is the wrong; and the expression ‘special damage’, when used of this damage, denotes the actual and temporal loss which has, in fact, occurred.”

   As with slander actionable only on proof of special damage, such “actual and temporal loss” meant pecuniary loss: see Gatley on Libel and Slander, 13th ed (2022), para 22-021; McGregor on Damages, 21st ed (2021), para 48-011.

1. Bowen LJ went on to explain how the term “special damage” came also to be used in a narrower sense. This derived from the insistence of the courts, in cases of slanders not actionable per se, “where actual damage done is the very gist of the action”, that “the actual loss must be proved specially and with certainty” (p 531). The same applied in actions for malicious falsehood (where again the damage done was the gist of the action). Bowen LJ said, at p 532:

   “The necessity of alleging and proving actual temporal loss with certainty and precision in all cases of the sort has been insisted upon for centuries …”

   As a result of this requirement, the term “special damage” came to have the connotation of damage which is precisely specified. This was associated with a rule of pleading requiring “special damage” in the narrower sense of damage capable of exact calculation to be pleaded with particularity.

1. The conclusion of the Court of Appeal in Ratcliffe v Evans was that, although in an action for malicious falsehood it is necessary to allege and prove “special damage” in the original sense of pecuniary loss, it is not always necessary to plead and prove such loss with exactitude. Rather, the degree of precision required depends on the nature and circumstances of the publication. As Bowen LJ memorably put it, at pp 532-533:

   “As much certainty and particularity must be insisted on, both in pleading and proof of damage, as is reasonable, having regard to the circumstances and to the nature of the acts themselves by which the damage is done. To insist upon less would be to relax old and intelligible principles. To insist upon more would be the vainest pedantry.”

   The Court of Appeal held, at p 533, that where, as in that case, the statement falsely and maliciously published about the claimant’s business is one “which in its very nature is intended or reasonably likely to produce, and which in the ordinary course of things does produce, a general loss of business, as distinct from the loss of this or that known customer”, evidence of such general loss of business was admissible and could support the action.

1. In two later cases the House of Lords reiterated the need for proof of special damage in an action for malicious falsehood. In White v Mellin [1895] AC 154 the House of Lords laid down that words calculated to injure a person in their trade, even if false and maliciously published, were not actionable without proof of special damage if the words were disparaging of the claimant’s goods without being defamatory of his character. In The Royal Baking Powder Co v Wright, Crossley & Co (1900) 18 RPC 95 the complaint was again that the defendant had maliciously made false statements which were calculated to injure the claimant in its trade but which were not defamatory. While different views were expressed about whether the evidence was sufficient to establish liability, all four law lords who gave reasoned judgments emphasised that damage was the gist of the action and three of them said that proof of special damage or “specific money damage” was required: see p 99 (Lord Davey); p 101 (Lord James of Hereford); p 103 (Lord Robertson). Lord Robertson stated that:

   “the essential ground, the gist, of the action … is special damage, done maliciously. Unless the Plaintiff has in fact suffered loss which can be and is specified, he has no cause of action. The fact that the Defendant has acted maliciously cannot supply the want of special damage, nor can a superfluity of malice eke out a case wanting in special damage.”

Lord Robertson distinguished Ratcliffe v Evans on the ground that the claimant had not attempted to show a general loss of business.

The Porter Committee report

1. It was against this legal background that a Committee on the Law of Defamation chaired by Lord Porter was appointed in 1939 “to consider the law of defamation and to report on the changes in the existing law, practice and procedure relating to the matter which are desirable.” The Committee treated the law of defamation as including certain “actions on the case” said to comprise slander of title, slander of goods and “other false statements made maliciously and calculated to cause and actually causing pecuniary damage”: see report, para 20. After an interruption of its work during the Second World War, the Porter Committee’s report was finally published in 1948 (Cmd 7536/48).

1. One subject discussed in the report was the distinction between statements actionable per se and statements actionable only on proof of special damage - taken by the Committee to mean “actual pecuniary loss” (see para 33). The report emphasised the difficulty of proving special damage, so that actions for slander actionable only on proof of special damage and “actions on the case” for false but non-defamatory statements made maliciously and calculated to cause damage were seldom brought (paras 33-35).

1. The report nevertheless recommended that the categories of slander actionable per se should be left unchanged, with one exception. The Committee criticised the requirement that, to be actionable without proof of special damage, words naturally tending to injure or prejudice the reputation of the claimant in his office, profession or trade had to be spoken of the claimant “in the way of” his office, profession, or trade (paras 47-49). The facts of Jones v Jones (without mentioning the case by name) were given as an example of how this requirement could cause injustice. The Porter Committee recommended that the common law definition should be amended by statute to abolish this restriction.

1. With regard to “actions on the case” (ie for malicious falsehood), the report stated, at para 51:

   “The necessity of furnishing proof of special damage has rendered this type of action rare in the extreme; but statements of these kinds may cause very serious damage which, owing to technical rules of evidence, it is impossible to prove strictly as special damage. In the result, the injured person is left without any remedy for the loss which he has suffered. In our view, this constitutes an injustice which should be righted by an amendment of the existing law.”

The 1952 Act

1. In response to these (and other) recommendations made by the Porter Committee, Parliament enacted the 1952 Act, described in the preamble as “[a]n Act to amend the law relating to libel and slander and other malicious falsehoods.” The sections of the Act relevant on this appeal are sections 2 and 3(1), which I have quoted at paras 18 and 19 above.

1. I will take section 2 first because it helps to clarify the meaning of section 3(1). Even without the aid of the Porter Committee report, it is evident that section 2 is intended to remove the defect in the law exposed by Jones v Jones. The clear purpose and effect of section 2 is to extend the category of slander actionable per se where the claim is based on words tending to injure a person in his or her calling etc. It does so by abolishing the restriction that the words must be spoken of the claimant “in the way of” that calling.

The effect of section 3(1)

1. Turning to section 3(1), its clear underlying purpose is to make it easier for claimants to recover damages in actions for malicious falsehood. It is less clear exactly how the statute is intended to achieve this. I must consider three submissions made by counsel for the defendants about what legal consequences follow when the words complained of fall within either (or both) of conditions (a) and (b) of section 3(1).

1. First, they submit that the words “it shall not be necessary to allege or prove special damage” dispense with the requirement under the common law to allege and prove loss with certainty and precision. The defendants acknowledge that in Ratcliffe v Evans the Court of Appeal had already relaxed that requirement by allowing a claimant to rely on a general loss of business, without having to prove the loss of specific customers. But they submit that section 3(1) goes further than this by allowing the court to draw an inference that the publication caused financial loss simply from the likelihood of such loss even without evidence of a general loss of business.

1. Second, the defendants submit that the presumption created where section 3(1) applies is rebuttable. They argue that there is nothing in the wording of section 3(1) or in principles of common law to prevent a defendant from relying on specific facts established at the trial to negative the inference of financial loss to which section 3(1) gives rise and thereby defeat the claim.

1. Third, the defendants submit that there is nothing in section 3(1) to suggest any intention to discard the basic principle that financial loss is the gist of an action for malicious falsehood.

The nature of the presumption

1. I agree with the first and third of these submissions but not the second. Although I initially thought otherwise, I have been persuaded that the effect of section 3(1), where it applies, is to create a presumption of law that the publication of the words complained of has caused financial loss. This presumption is irrebuttable as far as liability is concerned but does not necessarily lead to an award of substantial damages. If the court concludes, as the judge did here, that no financial loss has actually been caused, the claimant will be entitled only to nominal damages.

1. As I now see it, this conclusion is inescapable when section 3(1) is read together with section 2. Unless there is some compelling reason to conclude otherwise, it is to be assumed that language in a statute is used consistently and that the same term bears the same meaning wherever it occurs. That inference is reinforced here by the fact that sections 2 and 3(1) not only both contain the term “special damage” but also follow a similar pattern. Each provides that, if the words on which the action is founded are “calculated to” have a specified effect, then “it shall not be necessary to allege or prove special damage.”

1. In section 2 the intended effect of this formulation is plainly to make words falling within the scope of the provision actionable per se in the sense discussed above. That is to say, all the claimant need show to establish a cause of action is that words of the relevant kind were spoken by the defendant to a third person. If this is shown, the damage necessary to found the action is presumed by law and therefore need not be proved: see Jones v Jones [1916] 2 AC 481, 500 (Lord Sumner); Gatley on Libel and Slander, 13th ed (2022), para 5-002. It is then a matter for the jury (or nowadays the judge) to assess the amount of damages. Thus, if Jones v Jones had been decided after the 1952 Act was enacted, the judgment for the claimant in that case would have been upheld.

1. It follows that, when the same formulation (“it shall not be necessary to allege or prove special damage”) is used in section 3(1), this does not mean merely that damage need not be alleged and proved with certainty and precision. Rather, these words are to be understood as having the same meaning as they do in section 2. The term “special damage” must therefore likewise refer to the damage which under the common law a claimant would have to allege and prove to establish a cause of action: that is to say, actual pecuniary loss. As with slander actionable per se, where the words published fall within the scope of the statutory provision the damage necessary to found an action is presumed by law and so need not be alleged or proved.

1. There is another aspect of the relationship between the two sections which reinforces this conclusion. The combined effect of conditions (a) and (b) in section 3(1) is to limit the scope of section 3(1) as it applies to spoken words to words calculated to cause pecuniary damage to the claimant “in respect of any office, profession, calling, trade or business held or carried on by him at the time of the publication.” This language also tracks the language used in section 2. It can be inferred that the reason for this limitation is to avoid inconsistency between the scope of defamatory statements actionable per se and the scope of malicious falsehoods which are similarly actionable per se. If section 3(1) had applied to all words calculated to cause pecuniary damage to the claimant, whether written or spoken, a claimant could have sued for a malicious falsehood spoken orally without the need to prove special damage in circumstances where such an action would not lie for a defamatory statement spoken orally because the words do not fall within one of the categories of slander actionable per se. The legislative aim of avoiding such an inconsistency is corroborated by the Porter Committee report, paras 52-54. This feature of section 3(1) further confirms that the curtailment of the need to allege and prove special damage is intended to operate in the same way and have the same effect in section 3(1) as in section 2.

1. I add that, if the defendants’ interpretation were right, it is hard to see when time would begin to run for an action falling within section 3(1). In actions for defamation it remains the law that, for the purposes of section 4A of the Limitation Act 1980 (as under the Limitation Act 1623), the cause of action accrues on publication if the words are actionable per se, and on the occurrence of special damage in other cases (because until then, the cause of action is incomplete): see Lachaux, para 18. Where in an action for malicious falsehood, the claimant relies not on evidence of actual damage but on the presumption created by section 3(1), then the time for bringing an action must run from the moment of publication. As in such a case it is unnecessary to allege or prove actual damage, there is no other possible date. Proof at trial that no loss subsequently occurred can preclude an award of substantial damages but cannot retrospectively extinguish a cause of action that has already accrued.

Financial loss remains the gist of the action

1. Despite the parallels between sections 2 and 3 of the 1952 Act, there is an important difference between claims based on defamatory statements actionable per se and actions for malicious falsehood based on words that fall within section 3(1). This difference lies in the interest which the law protects. As discussed above, where defamatory statements are actionable per se, the gist of the tort is injury to reputation. That is clearly not so in cases of malicious falsehood. Once the distinction was decisively drawn in Malachy v Soper (see para 35 above) between slander of someone’s person or character and “slander” of their title or goods, it was impossible to characterise actions for slander of title, slander of goods or other malicious falsehood as concerned to protect reputation. Even when a false statement is both defamatory and maliciously published, and so gives rise to claims under both heads, the two causes of action are distinct and damages for injury to reputation cannot be recovered in an action for malicious falsehood. As Sir Donald Nicholls V-C said in Joyce v Sengupta [1993] 1 WLR 337, 348:

   “The history of malicious falsehood as a cause of action shows it was not designed to provide a remedy for such injury …”

1. It equally cannot be said that the cause of action for malicious falsehood was ever designed to provide a remedy for injury to the claimant’s feelings. Throughout its history, the only interests protected by the various actions which coalesced into the tort of malicious falsehood have been economic interests: see generally, Clerk & Lindsell on Torts, 24th ed (2023), paras 22-03, 22-09.

1. Section 3(1) has not altered this. Indeed, it impliedly confirms that the cause of action for malicious falsehood is designed to provide a remedy only for financial loss. That is because section 3(1) can only apply when the words upon which the action is founded are calculated to cause “pecuniary damage” to the claimant. This point was well made in Mayne and McGregor on Damages, 12th ed (1961), para 947 - the first edition of the work to be edited by Harvey McGregor:

   “There remains the question of non-pecuniary loss. Although s.3 of the Defamation Act 1952 continues to reflect the association of injurious falsehood with defamation because the statements singled out as not requiring allegation and proof of special damage are very similar to those defamatory statements which are actionable per se, there is still an interesting difference between such injurious falsehoods and such defamatory statements in relation to non-pecuniary loss. Whereas with defamatory statements actionable per se the principal head of damage is the injury to reputation which is non-pecuniary loss, with those similar injurious falsehoods which fall within s.3 of the Defamation Act 1952 the principal, and perhaps the only, head of damage is the pecuniary loss that the claimant has suffered. Section 3, by dispensing in these cases with the requirement of allegation and proof of specific pecuniary loss for success in the action, thus impliedly confirms the view that for all injurious falsehoods only pecuniary loss will ground the action …”

This passage has been repeated in all later editions: see now McGregor on Damages, 21st ed (2021), para 48-012.

The meaning of “calculated to cause”

1. The next question is whether the words found by the judge to have been false and maliciously published by Linda Cannell were “calculated to” cause pecuniary damage to the claimant. It is common ground that in this context the phrase “calculated to” does not mean “intended to” but “objectively likely to”; and that the test of likelihood is “more likely than not.” The dispute is about what facts are to be taken into account in assessing this probability. The defendants put their case in two alternative ways.

The “historical” approach

1. Their primary case is that the question whether the words complained of “are calculated to cause” pecuniary damage to the claimant is to be answered taking into account all relevant facts before, at and after publication, as established at the trial. So if, on the evidence adduced at the trial, it is found to be more probable than not that no pecuniary damage was caused to the claimant, section 3(1) does not apply. This may be called the “historical” approach. In contrast, a “forward-looking” approach limits the inquiry to facts existing at the time of publication.

1. The historical approach suffers from the double disadvantage of being inconsistent with both the language and the purpose of section 3(1). Section 3(1) does not say that there is no need to allege or prove special damage if the words on which the action is founded “probably have caused” pecuniary damage. The test is whether the words “are calculated” (ie are likely) to cause pecuniary damage. Such an assessment can be made only by looking forward from when the words were published without considering any information about what has subsequently happened and whether any pecuniary damage did in fact result from the publication. The language is inherently forward-looking.

1. The historical approach is also inconsistent with the purpose of section 3(1). As discussed, the aim of the provision is to make it easier to bring actions for malicious falsehood by removing the need to allege and prove (on the balance of probability) that special (ie pecuniary) damage has actually been caused. It would completely defeat that purpose if, to avoid the need to allege and prove that a malicious falsehood - more probably than not - has caused pecuniary damage, it were necessary to prove that it probably has caused pecuniary damage. Such an interpretation would make section 3(1) vacuous.

1. The historical approach also conflicts with the conclusion I have reached above that, where section 3(1) applies, the cause of action accrues at the time of publication of the words complained of and does not depend on anything that happens subsequently.

1. It is not suggested by counsel for the defendants that any court has decided that the historical approach is correct. The most that can be said is that judges have sometimes used language which is either equivocal or suggests that they may have assumed that it was relevant, in considering whether section 3(1) applies, to have regard to events that occurred after publication. In other cases, however, the court clearly treated the operative time for assessing the likelihood of loss as the time of publication. A good example is Tesla Motors Ltd v British Broadcasting Corpn [2013] EWCA Civ 152 which, before the present case, was the leading modern authority on section 3(1). I will return to this case but mention now that, as the defendants acknowledge, the claim in Tesla was advanced and addressed by both the judge and the Court of Appeal on the basis that whether the publication complained of was calculated to cause financial loss was to be assessed from the perspective of the time of publication.

1. At the trial of this action the judge received little assistance on this point, which was not clearly raised by the defendants until after the hearing had ended and was addressed only in additional written submissions. In accepting the historical approach as correct, the judge did not consider its compatibility with the language and purpose of section 3(1) and referred to only one authority as support for the proposition that facts occurring after publication can be taken into account: see paras 205-206. This was the decision of the Court of Appeal in Sallows v Griffiths [2001] FSR 15. There the relevant question was whether allegations of dishonesty made to the claimant himself and two other individuals were calculated to cause financial loss to the claimant so as to engage section 3(1). The claimant succeeded on this point before the judge; but an appeal was allowed on the basis that, apart from damage resulting from the claimant’s wrongful dismissal, which had been separately compensated:

   “there was no evidence before the judge that the [claimant] would have been likely to suffer any other damage from publication to any of the three persons relied upon.” (para 17 - emphasis added)

1. As Warby LJ pointed out, at para 58, this passage does not support the contention that what actually happened after publication may be considered. Rather, the language used shows that the Court of Appeal was looking at the facts at the time of publication and deciding that there was no evidence that it was likely at that time that the claimant would suffer any loss (other than loss for which he had already been compensated). In other words, the approach adopted was forward-looking.

1. I conclude that the Court of Appeal was right to reject the historical approach and to hold that the test is whether, viewed at the time of publication, the words complained of were likely to cause financial loss to the claimant.

The “forward-looking” approach

1. Counsel for the defendants recognised the difficulty - I would say impossibility -of justifying the historical approach. They took their main stand on their alternative argument that it makes no difference to the result here if the test is forward-looking from publication. What is critical, on their case, is that there is no restriction, other than relevance, on the facts existing at that date which can be considered. On the forward-looking approach information about what has actually happened after the date of publication must be left out of account. But this is only likely to lead to a different outcome from the historical approach if loss is avoided as a result of an occurrence which, at the time of publication, was unlikely.

1. The defendants criticise the Court of Appeal for applying what they call a “tendency” test of asking whether the words published had a “natural tendency” (or were “inherently likely”) to cause financial loss: see judgment, paras 65, 72. Counsel for the defendants make the point that words do not in themselves have a tendency to cause financial loss. Whether, viewed at the time of publication, words are likely to cause financial loss can be determined only if you know something at least about the factual context in which the words are published. On the facts of Ratcliffe v Evans, for example, to form the view that the words published were calculated to cause financial loss to the claimant, you would need to know at least that the claimant was carrying on the business referred to in the publications and had not ceased doing so. You would also need to know that the words were published to people who were actual or potential customers of the claimant.

1. The Court of Appeal accepted, at para 56 of the judgment of Warby LJ, that the question whether the words complained of are calculated to cause financial loss to the claimant cannot be approached in an entirely abstract fashion, detached from the circumstances of publication. Warby LJ thought it permissible to take account of the identity and essential characteristics of the claimant, the publisher and the publishee(s). The defendants submit that there is no principled basis for admitting evidence of these facts while excluding evidence of other facts which bear on the likely consequence of the publication. They argue that a realistic assessment of whether the words published are likely to cause financial loss to the claimant must take into account any relevant fact in existence at the date of publication, whether known to the defendant or not. Such facts include what the publishee(s) already knew or had been told about the subject matter of the publication and any steps which they would be likely to take in response to the publication.

1. An example of a case in which statements previously published were regarded as relevant is Tesla. Tesla sued the BBC alleging that statements broadcast on the programme “Top Gear” about the performance of its electric car were malicious falsehoods. Viewed on their own, the statements were calculated to cause financial loss to Tesla by putting off potential purchasers. Yet by the time of the broadcast complained of, the programme had been shown by the BBC on some 28 occasions. No claim could be made in relation to those earlier broadcasts because the time limit had expired. The judge decided that the action was an abuse of process because Tesla had no real prospect of showing that the actionable falsehoods were calculated to cause it any damage over and above that caused by the non-actionable statements. The Court of Appeal was not prepared to go quite that far, but upheld the result on the basis that Tesla had insufficient prospect of recovering substantial damages to justify the costs and the use of court time that would be involved in continuing the proceedings to trial: see [2013] EWCA Civ 152, para 49.

1. In Tesla the facts which showed that the publication complained of was unlikely to cause any financial loss to the claimant had occurred before publication. The defendants also point to cases where claims for malicious falsehood have failed because of the likelihood that, after the publication and in response to it, the publishee would take steps to verify the statement and would thereby discover the truth before relying on the statement.

1. In Stewart-Brady v Express Newspapers plc [1997] EMLR 192 a convicted murderer compulsorily detained in a secure psychiatric hospital complained of a newspaper article alleging that he had sexually assaulted a female visitor. The publication was said to be false and malicious and calculated to cause the claimant financial loss by inducing the hospital management to withdraw or reduce his discretionary weekly allowance. The judge held that there were no reasonable grounds for alleging that the publication was likely to have that result because the natural and probable response of the hospital management to the allegation would be to conduct an internal inquiry to determine whether it was true. If the allowance was withdrawn, it would be as a consequence of that inquiry and not of the publication.

1. In Fage UK Ltd v Chobani UK Ltd[2013] EWHC 630 (Ch); [2013] FSR 32, para 152, in the context of an allegedly false and malicious complaint to a regulator, it was found to be unlikely that the regulator would take steps of a kind likely to cause damage to a business without first investigating the validity of the complaint. Again, in Musst Holdings Ltd v Astra Asset Management UK Ltd[2021] EWHC 3432 (Ch), paras 632 and 640, the judge found that a false statement to a potential investor about the company’s financial position was not calculated to cause the company financial loss because it was more likely than not that the investor would make inquiries and ascertain the true position.

1. Applying such reasoning, the defendants submit that, even on a forward-looking test, the statements complained of here do not fall within section 3(1). Viewed at the time of publication, there was no prospect that the words spoken to Matthew Butler would cause financial loss to the claimant because Mr Butler had already decided not to do business with her. The same applies to Graeme Lingenfelder. He did not want the claimant to deal with LCA’s clients anyway and had warned her against doing so at the start of her employment. Nor was there any real likelihood that the false allegation about the claimant’s terms of employment with LCA made in the email sent to Mr Lingenfelder would cause her any financial loss. The defendants argue that, when the email was received, it was overwhelmingly likely that (as in fact happened) Mr Lingenfelder would discuss that allegation with the claimant, who would then show him her terms of employment with LCA, from which he would see that it contained no relevant post-employment obligation. So Mr Lingenfelder was bound quickly to discover (as he in fact did) that the allegation made in the email was untrue.

Relevance of the defendant’s knowledge

1. The validity of these arguments depends on the premise that, in judging whether the words complained of are likely to cause financial loss to the claimant, any relevant fact should in principle be taken into account if at the time of publication it existed or was objectively likely to occur. I do not agree that this premise is correct.

1. In my view, the defendants’ case misunderstands the purpose of the test in section 3(1). The aim is not to make the most accurate assessment possible of the likelihood of financial loss to the claimant. If that were the aim, section 3(1) would not have been enacted. The court would have been left to ask in all cases whether (on the balance of probability) the words published were not merely calculated to cause, but actually did cause, financial loss, which is the test under the common law. Starting from the defendants’ premise, the logical end point is the historical approach, which renders section 3(1) redundant.

1. The requirement that the words must be calculated to cause financial loss to the claimant was not introduced by the 1952 Act. It already existed under the common law. Under the common law it is necessary to show both that a false statement maliciously published is “calculated in the ordinary course of things to produce”, and that it does produce, pecuniary damage: see the authoritative statement of the law in Ratcliffe v Evans [1892] 2 QB 524, 527, quoted at para 1 of this judgment. Thus, the common law does not simply apply a test of actual causation. It also requires the words published to be calculated to cause financial loss to the claimant. If one asks why this is so, the answer must lie in the general principle of the law of tort that a defendant should not be held responsible for consequences of their wrongful act which are too remote in the sense that the defendant should not be blamed for failing to anticipate them. As Viscount Simonds stated in Overseas Tankship (UK) Ltd v Morts Dock & Engineering Co Ltd (The Wagon Mound) [1961] AC 388, 422-423:

   “It is a principle of civil liability … that a man must be considered to be responsible for the probable consequences of his act. To demand more of him is too harsh a rule, to demand less is to ignore that civilised order requires the observance of a minimum standard of behaviour.”

1. As the underlying question is whether the consequence is one for which the defendant ought to be held responsible, the assessment of probability must be based on such facts as the defendant knows or ought reasonably to know. It would defeat the purpose of the test if a defendant were to be held responsible for consequences which would be considered probable only if account is taken of facts which the defendant neither knew nor ought reasonably to have known.

1. If, judged on the facts known or which ought to be known by the defendant, a falsehood would be likely to cause financial loss to the claimant, it may nevertheless turn out that, for unexpected reasons, no loss is actually caused. Under the common law the defendant escapes liability in this situation. The effect of section 3(1) is to remove that escape route. What, if any, loss actually results remains relevant in assessing damages. But to establish liability the claimant need not allege or prove any actual financial loss.

1. Thus, I agree with Lord Hamblen and Lord Burrows that the appropriate test in determining whether the words complained of “are calculated to cause pecuniary damage” to the claimant is to ask whether, on the facts known or which should reasonably have been known to the defendant at the time of publication, it was objectively likely that the words published would cause financial loss. I also agree that, subject to this constraint, potentially relevant matters include prior publications, other statements made in the same publication, and steps which the publishee would be likely to take in response to the publication. Tesla and the cases mentioned at paras 74 and 75 above are all consistent with this test.

1. Here the matters on which the defendants rely to argue that the words published by Linda Cannell were not calculated to cause financial loss to the claimant are not matters which, at the time of publication, Linda Cannell knew or ought reasonably to have known. There is no suggestion that she knew or had reason to know of Matthew Butler’s private intention not to deal further with the claimant (because of a disagreement about commission). Nor that she knew or had reason to know of the conversation in which Graeme Lingenfelder had told the claimant at the start of her employment not to deal with LCA’s clients. In addition, an important factual finding made by the judge is that Linda Cannell was under the belief that the claimant did not have the handbook containing her terms of employment. This belief was based on the fact that, after leaving LCA, the claimant had asked Linda Cannell to send her a copy of her “contract”, which Linda Cannell had evaded doing. When Linda Cannell sent her email to Mr Lingenfelder, therefore, as well as her email sent a few minutes earlier to the claimant herself, it was not likely, based on the facts known to her, that Mr Lingenfelder would discover the truth: see Saini J’s judgment, paras 50-51, 61, 153, 155, 159, 163.

1. Judged, therefore, by reference to the facts known (or which ought to have been known) to Linda Cannell, her false statements made to Mr Butler and to Mr Lingenfelder were both calculated to cause financial loss to the claimant. The Court of Appeal was therefore right to conclude that both publications fell within section 3(1).

Article 10

1. I mention article 10 of the European Convention on Human Rights because the defendants set much store by it on this appeal. They emphasise that section 3(1) must be interpreted so that it does not interfere with the right to freedom of expression protected by article 10 in a way that is disproportionate to the legitimate aim of protecting the rights of others. They argue that this requires the court to adopt one of their proposed interpretations of section 3(1). This argument is supported in a written intervention from the organisation Media Defence which is dedicated to defending the rights of journalists and independent media to freedom of expression.

1. The argument made is not so bold as to advocate a right to publish false and malicious words that are calculated to cause financial loss to the person about whom the words are published. Instead, as arguments in this field commonly do, the defendants postulate a “chilling effect” on legitimate freedom of expression. It is argued that making it easier to establish liability for malicious falsehood would interfere disproportionately with the right to freedom of expression because it would have the consequence, even if unintended, of making it easier to use legal proceedings, or threats of such proceedings, to silence or deter the publication of allegations which are in fact neither malicious nor false.

1. Sometimes arguments of this kind carry real weight. Leading counsel for the defendants, David Price KC, in his reply submissions referred to the classic decision of the United States Supreme Court in New York Times Co v Sullivan (1964) 376 US 254. This held that the right to freedom of speech protected by the First Amendment to the United States Constitution precluded an award of damages to a public official for defamatory falsehood relating to his official conduct unless “actual malice” was proved, meaning that the statement was made with knowledge of its falsity or with reckless disregard of whether it was true or false. Brennan J, giving the judgment of the court, said that erroneous statement is inevitable in free debate, and that it must be protected if “the freedoms of expression are to have the breathing space that they need to survive” (see pp 271-272).

1. Conclusion on mental distress damages

1. In summary, it is our view that damages for proved mental distress/injured feelings can be and, unless ruled out on normal grounds such as remoteness, mitigation etc, should be awarded for malicious falsehood whether at common law or under section 3(1) (but, although we have heard no argument on this, not for loss of reputation as a non-pecuniary loss). In this case, this means that mental distress damages can be recovered by the claimant for, for example, upset and anxiety caused by the malicious falsehood that is actionable per se under section 3(1). As ordered by the Court of Appeal, the case should therefore be remitted to the trial judge for the assessment of such damages on the facts.

1. On this issue we therefore disagree with the judgment of the majority. Our reasons for doing so are set out above. The majority conclude that mental distress damages cannot be awarded because “financial damage is an essential element of the tort” (para 109). In our view that is inconsistent with the majority’s acceptance (at paras 51-56) that once the requirements of section 3(1) are made out the tort is actionable per se. If so, that means that the tort of malicious falsehood can be established (and, indeed, has been established in this case) without proof of pecuniary loss – ie it is not an essential element of the tort. It also follows that, contrary to what is said in the majority’s judgment at the end of para 109, our approach is consistent with the premise, shared with the majority, that non-pecuniary loss caused by a tort is recoverable subject to normal rules restricting or denying damages such as remoteness and mitigation.

1. Overall conclusion

1. The proper interpretation of section 3(1) of the 1952 Act is that, in the circumstances specified, the tort of malicious falsehood is actionable per se. Section 3(1) imposes a forward-looking test that requires the objective likelihood of the words used causing pecuniary loss to be judged at the time of publication by reference to all causally relevant facts and matters which are, or should reasonably have been, known to the publisher. Mental distress damages can be awarded for the tort of malicious falsehood under section 3(1) even though the claimant has suffered no pecuniary loss.

1. For all the reasons set out above, we therefore agree with the decision of the Court of Appeal so that, in our view, the appeal should be dismissed.