Majrowski v Guy's & St Thomas's NHS Trust

since the legislative history of the Act was to penalise conduct aimed at specific and identifiable individuals as distinct from groups or individuals, it cannot be assumed that Parliament intended vicarious liability to apply;

as a matter of construction of the Act, a claim only lies against an individual who is personally pursing a course of conduct amounting to harassment against another individual, and also, possibly against a corporation acting through someone as its “controlling mind”; and

Parliament did not intend to import the general principles of vicarious liability in tort into the remedy provided by section 3 since the Act was not intended to create another level of liability in employment law where the common law already provided adequate remedies:

“12 … Parliament did not intend simply to import the general law of tort wholesale into s. 3. That is apparent from the fact that the Parliament thought it right to confer a right under sub-section 2 to damages for anxiety which is not available to a claimant under the ordinary law tort. The essential question to be considered, putting it in a slightly different way, is whether the scheme and structure of the Act imports fundamental notions of vicarious liability into the remedy provided by s. 3(1).”

an employer is vicariously liable where his employee’s unlawful act in the course of his employment is sufficiently closely connected with the employer’s business;

there is no good reason of principle or policy to distinguish between statutory and common law civil wrongs when applying that form of vicarious liability;

for the same reason, it is immaterial that the act in question may also constitute a statutory criminal offence, since there may be secondary civil liability even though there may not be secondary criminal liability; and

unless the context and wording of the Act creating the statutory tort in question compels otherwise, an employer should normally be vicariously liable for his employee’s statutory tort committed in the course of his employment.

In holding that the acts of the employee may be attributable to his employer even where they are unauthorised and/or criminal in nature English courts had not followed Australian law;

Under English law an employer “stands in the shoes of his employee” so that he can be vicariously liable for his employee’s acts even if he, the employer, would not have been liable had he carried out the acts himself;

A civil claim lies against an employer for breach of statutory duty whether or not he can be held criminally liable for that breach; see Potts v Reid [1943] AC 1 HL.

because employers are more likely to introduce policies designed to reduce harassment in the workplace;

there must have been some reason for introduction of this statutory tort notwithstanding the availability of common law remedies against harassers and their employers for the same conduct;

one such reason may be to ensure the availability of a solvent defendant where a common law remedy is not available against the harasser or the employer;

another may be that the Act, in section 3(2), provides a wider range of damages than those recoverable at common law, in that it expressly includes provision for damages for “any anxiety caused by the harassment and any financial loss resulting from the harassment”; and

such an approach ensures that the law is consistent in its approach regarding secondary liability for different torts, and thus avoids anomalies that might otherwise bring it into disrepute.

As support for his argument that common law principles of vicarious liability cannot simply be “cut and pasted” into certain statutory contexts, he referred to the following passage from paragraph 40 of the speech of Lord Clyde in Lister v Hesley Hall:

“… some caution has to be exercised in looking for assistance from cases were the court is engaged in an exercise of statutory construction. The language and the purpose of the provision may call for an approach and a solution which may not exactly accord with the application of the rule of vicarious liability. … ”

He went on to give, as examples of legislation unreceptive to common law principles those concerning sexual and racial discrimination, citing Jones v Tower Boot Co Ltd. [1997] ICR 254, CA. However, that decision, which concerned the scope of the words “in the course of his employment” in section 32(1) of the Race Relations Act 1976, was given before the wider formulations by the House of Lords of that test in Lister v Hesley Hall and Dubai Aluminium. And, faced as it was, with a narrower linguistic construction than now, the Court had recourse to the clear purpose of the Act to protect employees from racial harassment of the sort found by the industrial tribunal, as a basis for affirming the tribunal’s entitlement to find for the claimant against the perpetrator’s employer.

Nevertheless, it should be noted that sections 32 and 33 of the 1976 Act, like, for example, sections 41 and 42, of the Sex Discrimination Act 1975 and sections 57 and 58 of the Disability Discrimination Act 1995, expressly render employers civilly liable for discriminatory acts of their employees in the course of their employment. Mr Platt contrasted the absence of any such express provisions in the 1997 Act, and submitted that their absence was powerful testimony to the intention of its draftsman not to provide for vicarious liability for this statutory tort.

Discussion and conclusions

As I have said, the 1997 Act neither expressly provides for vicarious liability nor expressly excludes it. Nor is there any sign of it, for example, by a special defence for employers of the sort provided in section 32(3) of the Race Relations Act 1976 as a counter-point to section 32(1) and (2) in their express attribution to an employer or principal respectively of acts done by a person in breach of the Act. However, unlike the 1997 Act which is enforced by the courts with common law principles such as vicarious liability at their disposal, claims under the 1976 Act and other cognate legislation fall to be litigated in tribunals without access to that resource unless imported in some adapted form into their statutory jurisdiction.

I also have in mind Lord Phillips’ observations in Thomas v News Group Newspapers Ltd. that I have set out above, in paragraph 5, to the clear effect that, while stalking may have been the prime mischief at which the 1997 Act was aimed, it was not the only one. The conduct was described in section 7 by reference to its consequences, not by reference to its nature, which he clearly regarded as capable of taking various forms, including on the pleaded claim in that case the publication of press articles publicising racist criticism of the claimant. Certainly, that is how the statute reads, as Lord Phillips, in the following passages acknowledged:

“15.

… the appellants concede that publication of press articles is, in law, capable of amounting to harassment albeit in only rare circumstances. This is an important concession, but I consider that it is a concession rightly made. The suggestion that journalistic articles were implicitly outside the ambit of the 1997 Act was, in my view, unarguable. In these circumstances, there can be no question of looking at Parliamentary material in order to resolve ambiguity of statutory interpretation. The issue is whether it is arguable that the publication in question constituted ‘harassment’ having regard to the effect that the respondent alleges they had on her.

16.

In this context, the appellants seek to place reliance on the ‘legislative intent’ by reference to debates in Parliament and other material. I do not propose to embark upon an elaborate analysis of the extent to which these are admissible as an aid to interpretation. I am prepared to proceed on the basis that that the appellants are correct in their contention that the mischief which led to the legislation was the practice of stalking. This much is common ground. Once it is conceded, however, that the 1997 Act applies to activities which go beyond stalking, the fact that the stalking was the principal mischief at which the Act was aimed, affords only limited assistance.”

I adopt a similar approach, but it is still necessary, in the context of this complaint, to consider any matters of policy that may assist on the question whether the 1997 Act should be construed so as not to apply to claims against employers of harassers. In doing so, I should keep in mind that the issue arises, not only where one employee, in the course of his employment, harasses another employee, but also where an employee, in the course of his employment, harasses an outsider, say a customer of his employer or some other third party with whom his work brings him or her into regular contact. It cannot, therefore, be disposed of, as the Judge appears to have done in paragraph 21 of his judgment, simply as a matter of employment law in the sense of claims by employees, on the basis that the Act was not designed,

“to create another level of liability in employment law where, as Lord Hutton pointed out in the Waters case, employees are adequately protected by the common law.”

I make no apology for returning in the context of this issue to Atiyah’s powerful reasoning on page 284 of his 1967, part of which I have already quoted:

“… there may be statutory provisions, such as s. 159 of the Mines and Quarries Act 1954, which make it quite plain that the employer is vicariously liable for a breach of statutory duty by his servants.

But in many cases, of course, the construction of the statute will not readily yield any simple answer to the question, and in these cases it is surely permissible to raise the question of policy. And from a policy point of view it is hard to find one good reason why the action for breach of statutory duty should constitute a solitary exception to the otherwise universal principle of the law that a master is vicariously liable for the torts of his servant committed in the course of his employment. In a great many cases it may be an accident of drafting precisely how the duties are formulated. Moreover, in many circumstances statutory duties are merely declarations by the legislature of what it regards as dictated by the requirements of reasonable care. … there are many statutory duties which overlap with the ordinary common law duty to take reasonable care, Where this is the case the statutory duties cannot normally be taken as superseding the common law duty, so an employer can still be held liable vicariously liable for negligence whatever the position may be with regard to the breach of statutory duty. But there may be cases in which the statutory duties do supersede the common law entirely, and in such circumstances it would be extraordinary if the effect the statute were to take away remedies against an employer which were formerly available.”

As Mr Langstaff observed, if regard is to be paid to the fact that the 1997 Act had its genesis in the need to curb stalking, the work-place is much closer to stalking than, say, the conduct accepted as covered by the Act in Thomas v News Group Newspapers Ltd. It is the very place where harassment is often encountered and from which its victim is often powerless to escape. It is thus often likely to be a risk incidental to employment, as, for example, recognised by Lord Hutton in Waters v Metropolitan Police Commissioner, in which the House of Lords held that a common law claim against the Commissioner in negligence for harassment of one of his officers by fellow officers was arguable. Why should the answer be any different because the claim is couched as breach of statutory duty under the 1997 Act? And it falls squarely within the two main policy considerations articulated by McLachlin J, giving the judgment of the Court in Bazely v Curry, at paragraphs 30 - 32, underlying the imposition of vicarious liability - risk created for others by the employment and the need, by deterrence, to minimise that risk.

As to Mr Platt’s “floodgates” argument in the field of employment law, there are, as Mr Langstaff pointed out, a number of safeguards for employers. First, there is the constraint provided in the Act itself, where, in section 1, it prohibits “a course of conduct” amounting to harassment; that is, it takes more than a single act of an employee in the course of his employment to engage the liability of his employer. Second, to succeed in claims under section 3 it is necessary for the claimants to establish to an objective standard at least that the conduct amounts to harassment, usually in the sense of it being likely to alarm or cause the claimant distress, as provided by sections 1(2) and 7((2) (see paragraphs 3 and 4 above). And, third, the conduct, looked at in the statutory context making it unlawful, must make it just and reasonable in the circumstances of the case to compensate the claimant by application of the “close connection” and/or reasonably incidental risk” criteria, as described by Lord Steyn in Bernard (see paragraph 35 above).

Of course, the courts have yet to develop some practical norms for determining in what circumstances it may be just and reasonable to fix an employer with vicarious, that is, absolute, liability without eliding that exercise with an enquiry as to common law negligence. As I have said, in so doing it will be vital to cling to the criteria of “close connection” and/or “reasonably incidental risk”, otherwise there would be a risk of removing the absolute quality of the liability which hitherto has flowed from the traditional formula of an employee’s act having occurred “in the course of his employment”. In the field of employment law, the reasonably incidental risk criterion may be of particular value when considering whether an employer should shoulder responsibility for, say, harassment by an employee whom he has put in a position of authority over other employees who may, for one reason or another, be vulnerable to such conduct, or in a position to exploit third parties such as needy customers, clients or claimants who may be similarly vulnerable.

Nowadays, employers, whether public or commercial undertakings, are expected to be alert to all sorts of discrimination and exploitation by their employees, which may include harassment, in or from the workplace, whether of fellow employees or third parties, and to establish good working practices and procedures to warn and/or guard against such abuse. If, for want of such practices or procedures falling short of negligence, or if, despite them, the nature of the employer’s undertaking and/or circumstances of a claimant’s exposure to his employees’ conduct are such as, in the view of the court, to render harassment in breach of the Act a reasonably incidental risk of the undertaking and/or employment, it may consider it just and reasonable in the circumstances to hold the employer vicariously liable.

As to Mr Platt’s argument contrasting the express provision for employer’s liability in other statutes concerned with discrimination and analogous conduct, it is, as always, a matter of looking at each statute in its own context. A useful example of such a trite proposition is Hilton International Hotels v Protopapa [1990] IRLR 316, a constructive dismissal case in which the Employment Appeal Tribunal rejected a submission that the absence of any provision for vicarious liability in the Employment Protection (Consolidation) Act 1978 indicated that the general rule that an employer is vicariously liable for his employee’s acts done in the course of his employment did not apply. Knox J, giving the judgment of the Tribunal, said at para 17:

“… We do not regard this argument as compelling because the context of the Sex Discrimination Act 1975 is quite different from the context of the Employment Protection (Consolidation) Act. The Sex Discrimination Act 1975 is dealing with a very wide multiplicity of different situations besides the master and servant employer-employee relationship. In relation to many of those relationships where there is no necessary contractual nexus, it was necessary in our view to define in general terms the circumstances in which employers were liable for their employees’ actions. Accordingly there being a different context which justifies the inclusion of the specific provisions in s. 41(1) we do not think it right to draw conclusions from the absence of such a provision in the different context of the Employment Protection (Consolidation) Act 1978. …”

More generally, it should be noted that the various pieces of legislation dealing with discriminatory and analogous conduct to which Mr Platt referred are for determination by statutory tribunals exercising the statutory jurisdiction expressly conferred on them. Unless the jurisdiction given to them expressly provides for some form of vicarious liability, it would not normally be open to them to import common law principles to like effect. By contrast, the 1997 Act is for application by the courts, exercising their mainstream common law, as well as relevant statutory, jurisdiction.

Whether a corporation is “a person capable of harassing “another” under section 1 of the 1997 Act.

Submissions

Mr Platt’s second main submission on this issue was that, as section 1 prohibits “… [a] person” from harassing “another”, a company cannot be “a person” who can harass in breach of this prohibition. He also relied on this proposition as a further reason for construing the Act as directed at the mischief of public order offences and so as to exclude from its application the common law concept of vicarious liability.

I have, already referred to the shortcomings of this argument as an approach to the question whether there can be vicarious liability under the Act. But in deference to Mr Platt’s arguments going to the narrower question whether, in any event, a corporation, as distinct from an individual may be vicariously liable under the Act, and the Judge’s reliance on it in striking out the claim, I shall express my views on that question.

Mr Langstaff submitted that, by the ordinary rules of construction, the word “person”, insofar as it applied to the harasser, as distinct from the person harassed, could encompass a corporation and that, if the moving spirit of a company caused one of its employees to pursue a course of conduct amounting to harassment of an individual, the company would have primary liability for the statutory tort. Equally, Mr Langstaff submitted, those terms have effect for the purpose of fixing a company employer with secondary – that is, true vicarious - liability for an employee’s breach of duty committed in the course of his employment, including those of which the employer may be unaware and which he may have forbidden or not authorised.

Discussion and conclusions

In my view, none of the three authorities cited by Mr Platt for this submission supports it. The first was DPP v Dziurzynski in which Rose LJ and Gibb J, sitting in the Divisional Court, held that a criminal charge of harassing a group of unidentified employees of a company was bad for duplicity as they did not constitute a clearly definable group. In the course of his judgment, Rose LJ observed obiter, at paragraphs 32-33, that the legislative history of the provision pointed away from the victim of the alleged harassment, being a corporation. Similarly, in the second and third authorities on which Mr Platt relied, Daiichi Ltd & Ors v Stop Huntingdon Animal Cruelty [2003] EWHC 2337 (QB), and Huntingdon Life Sciences Group PLC v. Stop Huntingdon Animal Cruelty & Ors, Owen J and Pitchers J respectively held, following the reasoning of Rose LJ in Dziurynski, held that that the word “another”, i.e. the alleged victim of harassment, in section 1 of the Act, did not on its proper construction embrace a corporate entity and that the Act was not intended to criminalise behaviour towards a limited company.

There is nothing in any of those authorities to suggest that the harasser or the employer of the harasser could not be “a person” for this purpose, or any policy reason why not. Even if there were some basis for that premise, I do not see why it should follow as a matter of policy or common-sense why a corporate employer should not have the same vicarious responsibility as any other employer for his employee’s breaches of this statutory duty causing harm of the sort specified in sections 3 and 7(2). The position is wholly different when one is considering, as the courts were in those cases, the victim rather than the perpetrator of the offence.

The thrust of the Act is plain, namely to protect individuals from a course of conduct amounting to harassment, regardless of who causes it. It is not, as Mr Platt suggested, an apt example of “What is sauce for the legislative goose should be sauce for interpretative gander”. Why? What possible dictate of policy or logic should protect an employer, whether corporate or not, from primary or secondary liability for the mischief of harassment of individuals at which the 1997 Act is directed? It is interesting to note that no such issue on liability was taken in Sharma v Wells and Medico-Legal Investigations Ltd, a successful application for a strike-out before Gray J in which, as can be seen, one of the defendants was a limited company.

As a post-script to this issue, I would distinguish Re Nelson v Bryon Price & Associates, to which Mr Platt referred the Court. In that case, as I have mentioned in paragraph 25 above, the British Columbia Court of Appeal held that no claim under section 17 of the Canadian Human Rights Code lay against the employer of the perpetrator of alleged acts of racial discrimination, since the provision referred to remedies against a “person who contravened this Act” carrying out such acts of discrimination, and did not on a natural construction include that person’s employer. Not only did this fall for discrete consideration as a part of that Code, but such claims were litigated at first instance before a statutory tribunal not a court exercising common law jurisdiction.

Whether there can be vicarious liability under the act for conduct that is both a criminal and civil offence

Submissions

Lastly, Mr Platt submitted that, as a course of conduct amounting to harassment under section 1 of the 1997 Act amounts, as he put it, “automatically” to a criminal offence under section 2, as well as to a civil offence under section 3, an employer can only be criminally guilty for the unauthorised offence of his employee if the employee is the governing mind of the company (see e.g Tesco Supermarkets Ltd v Nattrass [1972] AC 53, HL) or where it follows as a matter of statutory construction (see e.g. Meridian Global Funds Management Asia Ltd. [1995] 2 AC 500). Where neither of those avenues is open, so that an employer, which is a corporation, cannot be criminally liable, it cannot, he submitted, be civilly vicariously guilty for the same conduct under section 3.

Mr Platt took as his starting point that the 1997 Act creates a “statutory scheme of tortious liability”, which is coterminous with the creation of a criminal offence of harassment. This is not, he submitted, a breach of a statutory duty simpliciter since it is expressed as a criminal prohibition and not a positive obligation. Whilst acknowledging that a prohibition against doing something may well connote a positive obligation not to do the self-same act, he maintained that the distinction reinforced his submission that the Act was fundamentally concerned with personal criminal activity and not, given its context, some “unified tort of breach of statutory duty”. He pointed out that that there was no common law tort of harassment, as distinct from the tort of intentional infliction of harm, when the 1997 Act came into force, despite attempts made to persuade this Court that there was in Wong v Parkside Health NHS Trust [2001] EWCA Civ 1721. Accordingly, he submitted that the 1997 Act was “clean slate” legislation to which common law notions such as vicarious liability did not necessarily apply.

Conclusion

Mr Langstaff submitted in reply that that the fact that the employer could not be held criminally liable for such breach is no answer to the claim, since it is well established that an employer may be civilly responsible for his employee’s breach even though it constitutes a crime. – see e.g. Dyer v Munday; Morris v Martin (CW) & Sons [1966] 1 QB 716; and Lister v Romford Ice [1957] AC 555.

In my view, and quite apart from the fact that the standard of proof of the two forms of liability is different, there is abundant authority for the proposition that an employer may be vicariously liable in civil proceedings for his employee’s unauthorised criminal conduct, even though it could not be vicariously guilty of it in criminal proceedings. As Mr Langstaff observed, the 1997 Act, even without section 2, would make good sense. The civil remedy provided in section 3 is not predicated on there having been a criminal offence.

For all those reasons, I am of the view that the Judge was wrong to strike out the claim under CPR, Part 3.4(2)(a) on the basis that the 1997 Act does not permit the imposition of vicarious responsibility for breach of its provisions, and, therefore that Mr Majrowski had no reasonable grounds for bringing it against the Trust. First, it was not appropriate to strike out the claim as to the limits of vicarious liability in this statutory context, which, in the light of recently developed, and possibly still developing, jurisprudence, has established new and very broad principles the application of which is highly case and statute specific. Second, there is no clear basis in law for the limitations urged on behalf of the Trust that the Act was not intended to apply to the conduct of employees in the course of their employment, whether directed at fellow employees or third parties, or that it was not intended to import vicarious liability for such conduct in an appropriate case.

I stress the words “in an appropriate case” because, in the absence of any express provision in the Act, or imperative implication from its structure and context to exclude vicarious liability, the matter is best dealt with under the recently re-stated and broad general principles of vicarious liability. They are that the existence of vicarious liability for any common law or statutory wrong depends on whether, on the facts of the case, it is, by reference to the criteria of “close connection” and/or “reasonably incidental risk”, “just and reasonable” to hold the employer vicariously liable. And, for this purpose, the facts of the case have to be looked at in the context of the statute creating the civil offence. Therein lies the court’s control over any attempt at inappropriate extension of the Act to circumstances and fields of activity in which the imposition of vicarious liability would not be “just and reasonable”, not the imposition by the courts of a blanket exclusion of vicarious liability in respect of breaches of it regardless of their factual context (see paragraphs 35 and 56 above).

In my view, on the facts of the case as presently disclosed by the papers, the matter should go to trial. Accordingly, I would allow the appeal and would quash the Judge’s order striking out the claim.

Lord Justice May :

I agree that this appeal should be allowed for the reasons given by Auld LJ, whose account of the facts and circumstances of the appeal I gratefully adopt. As I shall explain, my reasons for agreeing that the appeal should be allowed are somewhat more directed to the specific facts on which the appellant seeks to rely. But this does not detract from my agreement with the principles which Auld LJ has discussed and adopted.

I have had the advantage of seeing in draft the judgment of Scott Baker LJ. He agrees, as I do, with Auld LJ’s exposition of, and conclusion on, the broad issue – whether an employer may be civilly vicariously liable for a breach of statutory duty imposed on his employee. I have nothing that I would wish to add to Auld LJ’s consideration of this issue in paragraphs 14-40 of his judgment.

Scott Baker LJ, however, reaches a different conclusion from Auld LJ on the narrower issue. He concludes that a purposive construction of the Protection from Harassment Act 1997 does not extend in any circumstance to making an employer vicariously liable for the civil consequences of harassment committed by an employee in the course of his employment.

He is persuaded by Mr Platt’s submissions that parliament cannot have intended to translate an essentially personal statutory liability into an added burden upon already burdened employers. He is concerned that a statute intended to control or prevent stalking should by an unintended side wind permeate the workplace. If Auld LJ’s approach is correct, every case will have to be examined on its facts. Employees already have a considerable bundle of rights and vicarious liability for an employee’s breach of the 1997 Act would considerably extend employers’ liability. Employers would have no defence to such claims of the kind provided, for example, by section 32(3) of the Race Relations Act 1976. There is under the 1997 Act a lower threshold for damages and a longer limitation period than apply, for instance, to stress at work claims. There would be a serious floodgates problem without an adequate control mechanism. Employers’ liability policies would not normally cover harassment claims. Parliament cannot in these circumstances have intended to impose a whole swathe of vicarious liability on employers.

These are powerful concerns, some, but not all, of which I share. There are, however, in my view, two clear control mechanisms, in the sense of conditions which a claimant must fulfil, or at least arguably fulfil, before a claim against an employer for statutory harassment by an employee, for which the employer was not personally responsible, might be entertained.

First, what is alleged has to constitute an arguable case of harassment within the statute properly construed. The statute does not define harassment, but the following provisions are relevant. By section 1:

“(1)

A person must not pursue a course of conduct –

(a)

which amounts to harassment of another, and

(b)

which he knows or ought to know amounts to harassment of the other.

(2)

For the purposes of this section, the person whose course of conduct is in question ought to know that it amounts to harassment of another if a reasonable person in possession of the same information would think the course of conduct amounted to harassment of the other.

(3)

Sub-section (1) does not apply to a course of conduct if the person who pursued it shows

…

(c)

that in the particular circumstances the pursuit of the course of conduct was reasonable.”

As Lord Phillips MR said in paragraph 30 of his judgment in Thomas v News Group Newspapers Limited [2001] EWCA civ 1223:

“The Act does not attempt to define the type of conduct which is capable of constituting harassment. “Harassment” is, however, a word which has a meaning which is generally understood. It describes conduct targeted at an individual which is calculated to produce the consequences described in section 7 and which is oppressive and unreasonable.”

There are other features of the 1997 Act which, in my view, tend to confine what a reasonable person would think amounted to harassment. Section 2 provides that a person who pursues a course of conduct in breach of section 1 is guilty of an offence punishable by imprisonment or a fine or both. This should colour any appreciation of conduct which amounts to harassment. It would, I think, reinforce the view of a reasonable person that harassment is serious conduct calculated to produce the consequences described in section 7(2) and which is oppressive and unreasonable. The reasonable person should also understand from section 3 that an actual or apprehended breach of section 1 can sustain, not only a claim for damages, but also an injunction, granted in the High Court or county court, restraining the defendant from pursuing any conduct which amounts to harassment. Alleged breach of an injunction may lead to arrest and, if the breach is established, the defendant is guilty of an offence. This again colours, so as to confine, any appreciation of conduct which amounts to harassment. Of course, a person who alleges harassment can pursue a civil claim for damages without recourse to criminal or injunctive remedies. But the same conduct sustains the criminal and injunctive remedies. In the civil context, the court will be alive to this fact when considering whether what is alleged really does amount to harassment. These considerations may not readily enable a court to strike out a claim without hearing evidence, but they do circumscribe the possible ambit of such claims.

Second, there is a control mechanism to be derived from the circumstances in which, as Auld LJ has explained, an employer may be vicariously liable for an employee’s statutory tort. Auld LJ has analysed the authorities, including Lister v Hesley Hall [2002] 1 AC 215, [2001] UKHL 22; the two Canadian authorities which the majority of the House of Lords found particularly illuminating, Bazley v Curry (1999) 174 DLR (4^th) 45 and Jacobi v Griffiths (1999) 174 DLR (4^th) 71; Dubai Aluminium v Salaam [2003] 2 AC 366, [2002] UKHL 48; and Benard v The Attorney General of Jamaica (Privy Council Appeal No 30 of 2003). Lord Steyn summarised the correct approach in Benard at paragraph 18, saying:

“The correct approach is to concentrate on the relative closeness of the connection between the nature of the employment and the particular tort, and to ask whether looking at the matter in the round it is just and reasonable to hold the employers vicariously liable.”

Lord Clyde said at paragraph 37 that the sufficiency of the connection may be gauged by asking whether the wrongful actings can be seen as ways of carrying out the work which the employer had authorised. He emphasised in paragraph 44 that consideration of the time at which and the place at which the actings occurred will always be relevant, but may not be conclusive. The acting may be so unconnected with the employment as to fall outside any vicarious liability.

“Acts of passion and resentment (as in Deatons Pty Ltd v Flew (1949) 79 CLR 370) or of personal spite (as in Irving v Post Office [1987] IRLR 289) may fall outside the scope of employment.”

Lord Hobhouse brought these two strands together in paragraph 59 where he said:

“Whether or not some act comes within the scope of the servant’s employment depends upon an identification of what duty the servant was employed by his employer to perform. (see Diplock LJ above). If the act of the servant which gives rise to the servant’s liability to the plaintiff amounted to a failure by the servant to perform that duty, the act comes within “the scope of his employment” and the employer is vicariously liable. If, on the other hand, the servant’s employment merely gave the servant the opportunity to do what he did without more, there would be no vicarious liability.”

It is always dangerous to postulate hypothetical facts as examples of what may or may not come within a general principle. But these passages indicate that the fact that there may be harassment at work is by itself insufficient to sustain vicarious liability in the employer. Thus, as Mr Langstaff QC accepted, if two employees, neighbours perhaps, who have a confrontational relationship outside and unconnected with their work, continue the confrontation at work so that one harasses the other, there would be no sufficient close connection between the nature of their employment and the harassment. By contrast, if a manager is alleged in the course of management to have harassed a subordinate, this may, depending on the facts, give rise to vicarious liability.

It is with these considerations in mind that I would allow the present appeal. Auld LJ has related the facts which the appellant alleges in paragraph 8 of his judgment. I acknowledge that Judge Collins made his decision upon his construction of the 1997 Act without material reference to the facts, which have not been investigated. Nevertheless, since the court is able and, I think, obliged upon application to strike out or dismiss a claim which as pleaded has no real prospect of success, I would have dismissed this appeal, if I had considered that, upon the facts pleaded, the claimant had no real prospect of establishing both harassment and the close connection with employment necessary for the vicarious liability claim.

I return to the question of the proper construction of the 1997 Act. Although preventing stalking may have been its first aim, it is not limited to stalking. I have already referred to the passage at paragraph 30 of Lord Phillips MR’s judgment in Thomas. The statute is silent as to the possibility of vicarious liability, but the general law, as Auld LJ has explained, is that an employer can be civilly vicariously liable for breach of a statutory duty imposed on an employee. The question therefore is whether the statute is to be construed as excluding the possibility of vicarious liability. Auld LJ has considered Mr Platt’s particular submissions to this effect, and I agree with his analysis of each of the points. The only points which seem to me to come close to true matters of construction are (a) the argument with reference to criminal liability, and (b) the contrast with, for instance, section 32 of the Race Relations Act 1976. I do not find the first of these persuasive. As to the second, as Auld LJ has explained, tribunals need a statutory structure which the common law does not need.

For the rest, the submissions are floodgates arguments with reference to one view of what is put forward as presumed parliamentary intention. Purposive construction needs to identify securely the purpose. The original purpose of the 1997 Act may have been to prevent stalking, but the statute was not drawn so narrowly. The purpose plainly was to provide criminal and civil remedies against harassment. I am unable to discern a purpose that this should not extend to the possibility of an employer being vicariously liable for harassment committed by an employee upon the principles as to vicarious liability which I have identified. On the contrary, I can see nothing difficult in principle with a parliamentary purpose, in legislating to create the statutory tort of harassment, that, if there is a sufficiently close connection between the nature of the employment and the harassment such that looking at the matter in the round it would be just and reasonable to hold the employer vicariously liable, the employer should indeed be liable. To put it in Diplock LJ’s more concrete terms, if the harassment is sufficiently closely connected with the employee’s job, why should it not have been parliament’s purpose that the employer should be vicariously liable in civil proceedings for the harassment?

Lord Justice Scott Baker:

I have had the advantage of reading in draft the judgment of Auld LJ. I agree with him as to the broad issue and gratefully accept his lucid analysis of the principles and authorities. I have, however, come to a different conclusion on the narrow issue namely whether an employer may be civilly vicariously liable under section 3 of the Protection from Harassment Act (“the 1997 Act”) for harassment committed by an employee in the course of employment. I adopt Auld LJ’s description of the facts, the relevant statutory provisions and also the abbreviations used by him.

My starting point is therefore that vicarious liability is not confined to common law claims and as a general principle an employer may be vicariously liable for breach of a statutory duty imposed on his employee. Whether he is or is not ultimately depends on the true intention of the statute imposing the duty. Further, the employee’s wrongful act must be so closely connected with what he is authorised to do that it would be fair and reasonable to regard it as a mode, albeit an improper one, of doing it.

The real question facing the court in this case is the true construction of the 1997 Act. Was it Parliament’s intention to impose civil liability on an employer for acts of harassment perpetrated by an employee, thus giving the victim the opportunity to recover damages from the employer as well as or instead of the employee? The victim may of course be any third party and may or may not be another employee of the perpetrator’s employer. The fact that an act of harassment may be committed within or from the workplace is purely incidental. The Act is concerned with harassment rather than the particular circumstances in which it takes place.

There is no express provision in the 1997 Act to say that an employer either is or is not vicariously liable for harassment by his employee. I do not, however, regard this as conclusive in either direction. One might have thought that if the draughtsman had intended to create vicarious liability in this instance he would have expressly said so, see e.g. section 159 of the Mines and Quarries Act 1954. I look therefore with some care at the structure of the Act. The long title describes it as:

“An Act to make provision for protecting persons from harassment and similar conduct.”

Section 1 sets out the prohibited conduct. This makes it clear that what is forbidden is a course of conduct rather than a single act. Also, intention is not relevant; the test whether a particular course of conduct amounts to harassment is objective. There are limited exceptions for conduct that could otherwise be described as harassment. The focus of the Act is on the effect of the harassment on the victim. The reaction of the victim is obviously important and this is something that is likely to be peculiarly within the knowledge of the harasser at whose continuing conduct the Act is aimed.

Section 2 renders criminally liable a person who pursues a course of conduct prohibited by section 1.

Section 3 is headed “civil remedy”. It gives a victim or potential victim the right to claim in civil proceedings for an actual or apprehended breach of section 1. Section 3(2) provides, importantly in my view, that damages may be awarded for, inter alia, anxiety caused by the harassment and any financial loss arising from the harassment. I shall return to this in a moment. The remaining subsections of section 3 make provision for an injunction against the harasser and what may happen in the event of a breach. There is nothing in the section to suggest that Parliament is looking beyond the perpetrator for a remedy.

Section 4 creates a parallel offence of putting a person in fear of violence. Section 5 gives a court power, when dealing with a person convicted of an offence under section 2 or section 4, to impose a restraining order on him.

Section 6 expressly disapplies the ordinary limitation period of three years that applies to personal injury actions. The limitation period is therefore six years.

Section 7 is an interpretation section. It describes harassing a person as including alarming the person or causing him or her distress (s.7(2)). A course of conduct must involve at least two occasions (s.7(3)) and conduct includes speech (s.7(4)).

As Lord Phillips MR pointed out in Thomas v News Group Newspapers and Another [2001] EWCA Civ 1233, para 16, the mischief at which the 1997 Act is aimed is stalking, but the Act is wider than that. Speaking of the nature of the harassment he said:

30 “The Act does not attempt to define the type of conduct that is capable of constituting harassment. “Harassment” is, however, a word which has a meaning which is generally understood. It describes conduct targeted at an individual which is calculated to produce the consequences described in section 7 and which is oppressive and unreasonable. The practice of stalking is a prime example of such conduct.”

It seems to me clear therefore that harassment can be committed in a variety of different ways and with different overtones, sexual and racial being examples. What is all important (per s.1(2)) is the effect on the victim. The different circumstances in which harassment may occur are many and various. Obvious examples are between neighbours, former partners, in the workplace both between employees at the same level and downwards, and even by an employee of his superior. The 1997 Act is by no means confined to nor even directed at harassment in the workplace. Many circumstances caught by the Act will have nothing to do with the workplace whatsoever. In some cases the workplace may be partly involved. For example it is possible to envisage circumstances where A is conducting a campaign of harassment against B and does so from primarily or partly outside the workplace, but from partly within it. An obvious example is by harassing telephone calls. The question is whether Parliament intended the employer to be statutorily liable for any harassment perpetrated by an employee in the course of his employment.

In the present case Mr Majrowski complains of bullying intimidation and harassment by his departmental manager in the course of her employment by the Trust. He too was employed by the Trust and this is a case where the conduct complained of was committed, so it appears, entirely within the workplace. Does the 1997 Act, without more, fix liability on the Trust for conduct of which it may have been completely unaware? At this stage the question is simply whether the judge was right to strike out Mr Majrowski’s claim against the Trust as disclosing no cause of action. If the approach proposed by Auld LJ is correct it follows that every case in which it is alleged the perpetrator of the harassment was acting in the course of his employment will have to be examined on the facts for the court to decide whether it would in all the circumstances be just and reasonable to impose vicarious liability on the employer. That, as Auld LJ points out, is a matter for the trial judge on the evidence. The alternative is that on its true construction the 1997 Act does not envisage the imposition of vicarious liability in any circumstances and the judge should, as Judge Collins did, dismiss the claim without embarking on any fact finding exercise at all.

An employee already has a considerable bundle of rights against his employer. First he has rights under his contract of employment and second he has rights in negligence. There is also a raft of employment legislation. Auld LJ at paragraph 41 draws attention to the possibility of establishing either primary or secondary liability against an employer for victimisation and/or harassment quite apart from vicarious liability under the 1997 Act. In Waters v Commissioner of Police of the Metropolis [2000] 1WLR 1607 Lord Slynn of Hadley said at 1611 A:

“If an employer knows that acts done by employees during their employment may cause physical or mental harm to a particular fellow employee and he does nothing to supervise or prevent such acts, when it is in his power to do so, it is clearly arguable that he may be in breach of his duty to that employee. It seems to me that he may also be in breach of that duty if he can foresee that such acts may happen and, if they do, that physical or mental harm may be caused to an individual.”

If there is vicarious liability for breach by an employee of the 1997 Act this is a considerable extension on the employer’s liability at common law. In particular there is a lower threshold for damages under section 3 of the 1997 Act. The 1997 Act is concerned with the effect of harassment on the mind of the victim. Any anxiety caused by the harassment qualifies for an award of damages. This is in sharp distinction to stress at work claims where the threshold for an award is identifiable psychiatric injury, which has to be foreseeable injury following from a breach of duty on the part of the employer. Statutory claims for harassment and common law claims for stress at work will often overlap, but a statutory claim will be much easier for a claimant to establish.

There are in place in stress at work cases a number of control mechanisms that limit successful claimants to claims that properly warrant compensation; see the sixteen propositions set out by Hale LJ, as she then was, in Hatton v Sutherland [2002] 2 ALL ER 1, repeated by the Court of Appeal in Hartman v South Essex Mental Health and Community Care NHS Trust [2005] EWCA Civ 06. If the 1997 Act enables claims to be made against employers for vicarious liability for breach of statutory duty it seems to me there will be a serious floodgates problem and I would not regard the just and reasonable test as an adequate control mechanism. This test is so fact specific that the courts are likely to be faced with numerous claims. Stress at work claims will be presented as claims for breach of statutory duty under the 1997 Act. It will not be possible to strike the claims out because they will have to proceed to the point where a judge can decide on the facts whether the imposition of vicarious liability would be just and reasonable. I would not regard that as a satisfactory state of affairs. Furthermore, I would not have expected Parliament to create such a substantial extension to an employer’s liability, as it were, by a side wind.

There are two other points that perhaps give some indication of the intention of the legislature. First the limitation period for civil cases under the 1997 Act is six years rather than three and second the Employers Liability (Compulsory Insurance) Act 1969 only applies to bodily injury or disease. Employer’s liability policies ordinarily offer liability on this basis. Claims for anxiety due to harassment would not, in the absence of special arrangements, be covered under the usual employer’s liability policies. Although neither point is in itself a matter of great significance, both are indicative of the additional burden actions under the statute would place on employers. It is to be noted that in the present case Mr Majrowski went off work because of stress and yet brought no claim against his employers for breach of duty of care, apparently because (i) the Trust was not aware of or a party to any harassment on the part of their employee, Mrs Freeman and (ii) the three year limitation period had expired.

The judge in his judgment said this at para 19:

“But bearing in mind that, firstly adequate remedies exist at common law and, secondly the legislative history of the Act to which I have already referred, it by no means follows that it should be assumed to have been Parliament’s intention that a whole additional swathe of vicarious liability should have been imposed upon employers by virtue of section 3 of the Act. Section 3 was designed to give a remedy where none had existed for actions which had not previously been actionable.”

As to the legislative history, it is reasonably clear, see e.g. the Government White Paper: “Stalking – The Solutions” and the discussion by Rose LJ in DPP v Dziurzynski [2002] EWCA Admin 1380 that the 1997 Act was passed to replace the inadequate provisions of sections 4 and 5 of the Public Order Act 1986. The 1997 Act invokes the assistance of both the criminal and civil law. Its purpose is, as its title makes clear, protection from harassment and similar conduct. Compensating the victim is, it seems to me, incidental to the main purpose of preventing the conduct from taking place. True it can be said that if employers are vicariously liable they will be encouraged to take steps to prevent harassment in the workplace, but there are already in my judgment sufficient incentives in place. Civil liability arises when there is conduct or apprehended conduct that would amount to the criminal offence of harassment. The structure of the Act (and nothing in the sections that relate to Scotland suggests otherwise) is that it is directed towards the victim and the perpetrator. Harassment is something of an essentially personal nature, usually between two individuals. There is nothing to suggest that strict liability should attach to the perpetrator’s employer for those acts done in the workplace. The 1997 Act creates a statutory tort of harassment and provides for compensation by the perpetrator for anxiety and any financial loss caused by the harassment. The financial loss does not have to have been foreseeable provided it is caused by the harassment and compensation is triggered by mere anxiety as apposed to injury which would be necessary in a negligence action. Compensation under the statute, which is unlikely to be very great in the absence of a separate claim at common law, is surely intended to be paid by the perpetrator rather than someone else who is treated by the law as standing in his shoes.

I agree that the weight of authority as described by Auld LJ supports the view that vicarious liability applies generally to statutory torts committed in the course of employment. But that does not mean that in every case where a statute is silent as to the imposition of vicarious liability it should be assumed that it applies. It is important to apply a purposive construction to the statute. It is necessary to examine closely the vice at which the particular statute is aimed and the nature of the remedy that it provides. The fact that the perpetrator of harassment may be doing so within or from the workplace seems to me to be entirely incidental to the underlying purpose of this legislation. The primary purpose of the Act, so it seems to me, is to stop the harassment and provide the machinery for doing so rather than to award compensation in those cases that would not otherwise be covered by the common law. There is nothing to suggest that Parliament was looking to the employer to provide a monetary remedy for a lower threshold of damage to the victim than would be the case apart from the statute.

It is also to be observed that if vicarious liability is imposed by the statute it is imposed without any limitation as to circumstances. No statutory defence is provided; the only control mechanism is the just and reasonable test outlined by Lord Steyn in Bernard v The Attorney General of Jamaica (see Auld LJ at para 35 above).

Mr Platt drew attention to Jones v Tower Boot Co Ltd [1997] I.C.R. 254 and sections 32 and 33 of the Race Relations Act 1976 which expressly impose liability on an employer for discriminatory acts by employees. He also pointed out that other comparable deeming provisions are to be found in the Sex Discrimination Act 1975 sections 41 and 42, the Disability Discrimination Act 1995 sections 57 and 58, Regulation 11 of the Part-Time Workers (Prevention of Less Favourable Treatment) Regulations 2000 and Regulation 12 of the Fixed-Term Employees (Prevention of less Favourable Treatment) Regulations 2002. While I agree with Auld LJ that Jones v Tower Boot is not entirely in point to the present debate and that these statutory provision are covering different situations from those in the present case they are indications of the legislature’s ability to spell out liability on the part of the employer for acts of his employees.

In conclusion, the 1997 Act creates a statutory liability that does not overlap with common law negligence as for example the Occupiers Liability Act 1957. Nor is it directed to an employment situation as for example the health and safety legislation. It is aimed at unconscionable behaviour essentially by one individual to another. I regard the statutory duty as personal in nature and not one in which, in the event that the prohibited conduct happens to occur in the workplace, the employer is to be treated as standing in the shoes of an employee perpetrator. I regard it as difficult to envisage circumstances in which it would be just and reasonable to hold an employer vicariously liable and I do not think, viewing the statute as a whole, that it was Parliament’s intention that an employer should be vicariously liable.

I would therefore dismiss the appeal