1. Introduction

It’s pretty rare for the mainstream media to include a pdf link to a High Court decision in a news article, and so I was impressed to see that an article in the SMH included an embedded copy of Elisha v Vision Australia Limited (‘Elisha‘) — a decision concerning a psychiatric injury sustained by an employee when they were dismissed by their employer. Specifically, Elisha concerned the availability of damages for psychiatric injury to an employee where the employee had been dismissed following a disciplinary hearing that was — as the primary judge described it, and as the High Court majority adopted the term — a ‘sham’.¹ In a separate judgment, but in agreement with the majority, Jagot J put the issues very tersely:

An employer botches its own disciplinary procedures by not informing an employee of key allegations that the employer ultimately considers in deciding whether to terminate the employee’s employment and, therefore, fails to give the employee any opportunity to respond to those undisclosed allegations or a real opportunity to respond to the allegations in fact notified to him. In so doing the employer breaches terms of the employment contract.²

As might be obvious, the appeal was brought on a breach of contract basis, but the action was also framed in negligence. Ultimately, however, the High Court majority found that the applicant had succeeded on the breach of contract grounds and it was therefore not necessary to make finding as to the claims in negligence. Steward J, who dissented, however, did consider the negligence claim — and I’ll get to his Honour’s reasons for rejecting that claim below. But first I will summarise the breach of contract claim and holdings.

The breach of contract claim, in short, put the following questions to the Court. First, did Vision Australia’s (Vision) contract of employment with the employee include as a term its disciplinary procedure (which were subsidiary to and were developed after the contract was signed)? Second, did liability for psychiatric injury caused by Vision’s breach of contract fall outside the scope of an employer’s contractual duties insofar as the breach concerned the manner in which an employee was dismissed? And, third, was liability for psychiatric injury too remote to qualify as a kind of damage compensable in law for breach of contract in the circumstances of this particular contract?

The majority answered each of these questions in favour of the appellant, Mr Elisha, finding that the employment contract did incorporate the disciplinary policy as a term; that psychiatric injury was not beyond the scope of the employer’s contractual duties with respect to breaches of contract regarding the manner of dismissing an employee; and, finally, that liability for psychiatric injury was not too remote, especially given that the breach of contract on the facts of this case was particularly serious. So what happened?

2. Factual background

The facts leading to the termination of Mr Elisha (‘Elisha’) were said to be disputed. Yet in essence they were as follows. Elisha was employed by Vision as a technology consultant. That role required Elisha to set up or assist with the set up of software and hardware for vision impaired people throughout various homes and workplaces across Australia. This inexorably led Elisha to travel. While travelling through rural Victoria in the course of his duties, Elisha found himself unable to sleep one night at the Bairnsdale International Hotel, where noise was emanating from outside his room. Elisha complained to the hotel manager at the reception of the hotel at around 12.30am and, within 30 minutes, Elisha was moved to another room. At a later date, other employees of Vision were staying at the same hotel. In the course of conversation with the hotel manager, these employees were told by the hotel manager of the incident involving Elisha. In short, the hotel manager alleged that Elisha had been aggressive and intimidating when he had complained about the noise; and he had similarly been unpleasant to the manager on checking out the following morning.

As background to these incidents, Elisha was being treated by a general practitioner and psychologist for anxiety and depression in the previous year. In evidence were the psychologist’s notes, which recorded that Elisha was sensitive to sounds and had experienced chronic workplace stress and interpersonal difficulties with staff members at Vision. Also in evidence was material demonstrating that Elisha’s relationship with his manager was strained.

Upon hearing about the incident involving Elisha’s noise complaints at the Bairnsdale International Hotel, Elisha’s manager wrote to the human resources (HR) manager of Vision and, after various communications between the manager, the general manager of Vision, and a member of the HR team, the HR team member took the initiative of interviewing the hotel manager. Following that interview, the HR manager sent a letter to the manager of the Bairnsdale International Hotel, apologising for Elisha’s alleged behaviour. All of this occurred while Elisha was on annual leave, which took the form of a holiday with his family.

When Elisha returned from leave, he was given a formal notice — known in the proceedings as a stand down letter. That letter required Elisha to attend a meeting and advised he would therein be asked to respond to the allegations regarding the hotel. The letter also enclosed a procedure document, titled the Vision 2015 Disciplinary Procedure (‘2015 DP’) said to be effective from April 2015. The letter also contained instructions regarding the confidentiality of the meeting and the matter in general, and warned that a violation of confidentiality would result in his summary dismissal. The meeting occurred and Elisha denied the allegations made against him by the hotel manager. Elisha also denied the general allegations that he was aggressive and that his conduct could be seen to amount to serious misconduct.

In a subsequent meeting between Elisha’s manager and other Vision staff it was determined that Elisha’s employment would be terminated based on what had been heard in the disciplinary meeting. Various emails were then sent and received over subsequent days among Vision staff members, some of them suggesting that the finding that Elisha had demonstrated a pattern of aggression required strong (or stronger) evidence, seemingly for the purposes of the dismissal records. Elisha’s termination was confirmed shortly thereafter by letter. In part, the letter stated the following:

Vision Australia considers that on the balance of probability you did behave in the manner described by [Ms Trch]. [Ms Trch] reported this behaviour caused her to feel intimidated and humiliated. Vision Australia considers this behaviour to be serious misconduct.

Following the termination, Elisha was diagnosed with major depressive disorder (MDD), adjustment disorder and depressed mood, and was assessed to be without capacity to work in the future.

3. Curial history

• Elisha commenced proceedings for breach of contract and negligence in the Trial Division of the Supreme Court of Victoria, where O’Meara J delivered reasons and made orders in favour of Elisha: Elisha v Vision Australia Ltd [2022] VSC 754.
• Vision then sought and was granted leave to appeal to the Victorian Court of Appeal. McLeish, Kennedy and Macaulay JJA delivered reasons in favour of Vision, and ordered that the damages awarded in favour of Elisha should be set aside: in Vision Australia Ltd v Elisha [2023] VSCA 265.
• Elisha then sought and was granted leave to appeal to the High Court.

4. The breach of contract claim

4.1. Was the 2015 DP a contractual term?

The first legal issue concerned the dispute about whether the 2015 DP was incorporated into the employment agreement signed between Vision and Elisha in 2006 (‘2006 Contract’). In short, the 2006 contract took the form of a letter of offer executed by Vision. On its last page, the letter of offer included a clause titled ‘Other Conditions,’ which specified that

Employment Conditions will be in accordance with regulatory requirements and Vision Australia Policies and Procedures. Breach of the Policies and Procedures may result in disciplinary action.

The High Court majority found, relying on Toll (FGCT) Pty Ltd v Alphapharm Pty Ltd,³ that a reasonable person in the position of the parties would understand the above language to create contractually binding obligations, especially given the seriousness of the consequence of those obligations being breached (‘may result in disciplinary action’). The majority also rejected the contention that the contract’s commercial object was to create a one-sided obligation on Elisha: ie, that only he would be required to comply with directions issued by Vision in accordance with the procedures. Instead, the majority found, consistent with the courts below, that it would ‘defy logic and common sense’ for Elisha to be bound by policies and procedures but Vision would not be so bound.

Oddly, Vision also contended that it should not be bound to comply with its own policies or procedures because that would give rise to uncertainty, since those policies or procedures were likely to change from time to time. The majority rejected this contention too:

But the loss of some certainty is not a reason to deny the clearly expressed intention of the parties that those things that answer the description of a policy or a procedure, from time to time, will have contractual effect. A reasonable person would understand the intention to incorporate into an employment contract those policies and procedures as they might change from time to time, instead of requiring formal amendment to the contract every time those policies or procedures are varied.⁴

The majority noted that issues of concern might arise if Vision were to impose ‘substantially new or onerous’ obligations on employees unilaterally and without formally amending the contract. However, nothing of that nature had occurred here, because the 2015 DP (1) imposed obligations on Vision, not Elisha; and (2) these obligations were not ‘qualitatively any different’ to the applicable disciplinary procedures to be found in the original 2006 Contract.

Vision also contended for a fallback argument. It submitted that if the Court found the 2015 DP to be a term of the contract, then such a term would not be understood by a reasonable person in the parties’ position as a binding term. On examination of the 2015 DP, the majority found that certain specific assurances and promises made in the document that stated that certain procedures will be followed were — despite the presence of some non-binding recitals in the document — binding on Vision.

The majority found the 2015 DP could not be reasonably understood as merely designed to ensure compliance with the employment agreement because the former ‘must have been intended by the parties to have separate contractual effect.’⁵ That was the case because the 2015 DP (1) was not substantially different to the original 2005 Contract; (2) makes no references to the 2005 Contract and the terms are expressed as independent promises; and (3) is not expressed to be confined to or co-extensive with the unfair dismissal regime under the Fair Work Act 2009 (Cth).⁶

4.2. Where Vision’s manner of termination breached the contract, were psychiatric injuries within its scope of contractual duties?

Vision contended that an employee is ordinarily ‘precluded’ from recovering damages arising from a breach of contractual duty concerning the manner of their dismissal. This was a contention about limiting the scope of an employer’s liability to those species of liability that the parties might fairly be regarded as having contemplated and been willing to accept. In short, if the parties did not agree to be liable for certain forms of liability or damage, then they could not be liable for them.

Vision contended that mental distress is an unrecoverable head of damage in breach of contract claims, other than in narrow circumstances where the contract contains an express provision to the contrary (such as where the breach is of an express duty to ‘provide enjoyment, relaxation or freedom from molestation’⁷). The majority acknowledged that the principle underlying this rule was that, unless there is a contrary provision, breaches of contracts leading to mental distress (even where that is likely) should not be considered to instantiate a form of liability that is ‘part of the business risk of the transaction.’⁸

To clarify the principle, the majority quoted Mason CJ’s dicta in Baltic Shipping.⁹ The Chief Justice said that ‘anxiety is an almost inevitable concomitant of expectations based on promises’ and therefore ‘a contracting part must be deemed to take the risk of it.’ In other words, one who enters into a contract must be understood to have hazarded the risk of becoming mentally distressed in the event that the other party breaches the contract. This does not mean that such feelings, should they come home, are compensable as part of the breach litigation. Instead, those feelings are virtually inevitable and therefore should be understood to have been considered an acceptable risk by the contracting party at the time of entering into the agreement.

Vision marshalled the ratio above — relating to ‘anxiety’ and ‘mental distress’ — to contend that there was a rule that damages for psychiatric injury were beyond the scope of any contractual term concerning the manner of termination. Indeed, the Victorian Court of Appeal had agreed with that point and Vision relied on the Appeal Judges’ holding, as well as the older holding of the House of Lords in Addis.¹⁰ In Addis, the House of Lords set aside an award of damages made by a jury to a plaintiff who had been wrongfully dismissed. Lord Loreburn had made the leading speech and had said

If there be a dismissal without notice the employer must pay an indemnity; but that indemnity cannot include compensation either for the injured feelings of the servant, or for the loss he may sustain from the fact that his having been dismissed of itself makes it more difficult for him to obtain fresh employment.¹¹

Interestingly, Vision used the ratio from Addis to contend that three kinds of psychiatric injuries were unrecoverable as heads of damage: namely, damages concerning the manner of dismissal; damages for hurt feelings; and damages for any loss associated with difficulties of finding new employment. However, the majority, in oral argument, observed that the law of the United Kingdom (‘UK’) and Wales with respect to the third point — losses relating to re-employment — had changed, such that Addis was not a binding authority there. It seems that the High Court was not prepared in those circumstances to apply UK common law where that authority had been overtaken; and yet, that was also beside the point, since in this appeal, the majority said, the claimant was not seeking damages for hurt feelings or for losses associated with re-employment. Rather, the claim for psychiatric injury was independent from a claim for hurt feelings; and there was no equivalent claim in Addis — a fact that the majority deduced not just from the arguments advanced in Addis but also from the characterisation of the claim by Lord James as one of ‘aggravation’ (ie, aggravated damages) and by Lord Atkinson as one having the ‘shape of exemplary damages.’¹²

The majority went on to discuss the diminishing relevance of Addis, noting that much ‘water has passed under the bridge’ since it was decided in 1909. Such developments noted by the majority included:

• The UK Parliament’s establishment of legislation to protect employees against unfair dismissal pursuant to recommendations made in the so-called ‘Donovan Report,’ which was the report of the Royal Commission on Trade Unions and Employers’ Associations 1965-1968 (1968) composed by Lord Terence Donovan;
• The subsequent developments in the common law recognising an implied contractual term that an employer will not (without reasonable cause) jeopardise the relationship of confidence between employer/employee;
• The cases that followed later that have focused on this relationship of confidence and the ‘boundaries’ of the implied duties of trust (as opposed to the scope of the decision in Addis)
• These included Johnson v Unisys Ltd, which, although Lord Hoffman applied the ratio in Addis as it related to mental distress, and created what became known as the ‘Johnson exclusion area’¹³
• However, it also included Eastwood v Magnox Electric plc, wherein Lord Nicholls said the Addis ration (and the Johnson exclusion area) was an implied term that could not ‘co-exist satisfactorily’ with the unfair dismissal legislation and which ‘produces … strange results’¹⁴

What the majority said next is interesting for the law of contract in Australia. While the majority acknowledged the importance of the UK developments above, it stressed that the UK body of case law could not be ‘transplanted to Australia’ as the Australian legal context was too different. The High Court cited its 2014 decision in Commonwealth Bank of Australia v Barker (‘Barker’) to affirm that no ‘generalised implication’ should be drawn from the recognised, existing duty of trust and confidence in employment contracts in Australia because (1) the statutory circumstances between the UK and Australia differed; and (2) if the judicature were to introduce such an implied term into Australian contracts, this would ‘intrude a common law policy choice of broad and uncertain scope into an area of frequent, detailed and often contentious legislative activity.’¹⁵

In short, the burgeoning regulation of contractual obligations via employment law, presumably through the Fair Work Act 2009 (Cth) and other instruments, created a legal context wherein judges should appropriately tread lightly and with reluctance. Accordingly, the majority found that

the scope of contractual duties in an employment contract in Australia is unaffected by questions concerning the relationship between a generalised implied term developed by the common law and particular statutory provisions… [On the contrary,] the scope of the contractual duty falls to be considered by reference to the usual considerations of the nature of the liability that, in light of the parties’ agreement, the parties might fairly be regarded as having been willing to accept.

Those usual considerations were to be found in Baltic Shipping, where Mason CJ and McHugh J recognised psychiatric injury as a ‘class of physical or personal injury for which damages were recoverable,’ there being ‘no suggestion’ that there was a ‘class of contract’ that excluded or precluded the recovery of physical or personal injuries.¹⁶ The majority emphasises that no justification or principle had been advanced by Vision that would have persuaded the High Court to impose on every express term concerning the manner of dismissal in every employment contract in Australia an assumption that psychiatric injuries arising from a breach of that term should lay beyond the liability of the parties because it could somehow be said that psychiatric injuries were a form of liability beyond that which the parties were willing to accept. All in all, the majority found that Addis did not bite for psychiatric injuries, and that there was no contention from the respondent to compel the majority to ‘make new law’ (by imposing a limit on employer liability where they breached a contractual term relating to the termination of employees resulting in psychiatric injuries). It appears Vision would have had to have made the case for making such a law to have succeeded.

4.3. Was the psychiatric damage too remote?

The last issue to be determined was whether psychiatric damage was too remote, which, as the majority stressed, is a separate question from questions relating to the scope of a contractual duty. This was important for the majority to clarify, because both counsel for Vision and the Court of Appeal had, it said, conflated or relied on scope erroneously.

4.3.1. What is the correct remoteness test in breach of contract?

The majority confirmed that the correct test for whether damages were too remote in a breach of contract claim was the two-limbed test first enunciated by Alderson B in Hadley v Baxendale (‘Hadley‘).¹⁷ As Daniel P O’Gorman has noted, the Court of Exchequer in Hadley adopted a rule from the French Civil Code that was well known to English lawyers at the time: namely, that ‘[t]he debtor is only liable for the damages foreseen, or which might have been foreseen, at the time of the execution of the contract, when it is not owing to his fraud that the agreement has been violated.’¹⁸ However, Alderson B separated the rule out into two limbs.

The first limb of the test involves determining whether, on the facts of the case and by reference to the date of the contract, the breach arose ‘according to the usual course of things’ while the second limb involves asking whether it can be reasonable supposed that the damage had ‘been in contemplation of both parties’ and would occur as a ‘probable result’ of the breach of contract. Referring to another expression of the test the High Court had previously used, the majority said that the damage must have been reasonably contemplated as ‘on the cards’ when the contract was made.¹⁹

In her Honour’s separate concurring judgment, Jagot J said that the first limb of the Hadley test was not relevant in this case, as there was no contention that Elisha’s psychiatric illness would arise according to the usual course of things. Instead, the key question was what was contemplated by the parties at the time at which the contract was entered. As Jagot J noted, the specific degree to which the parties must have contemplated the kind of injuries that came home at the time of the contract formation, so as to satisfy the second limb, ‘has been expressed in a variety of terms.’ However, as Jagot J noted, ‘one description that has not found favour in Australia or (ultimately) in the United Kingdom is that the kind of damage caused merely be “on the cards.”‘²⁰

The majority similarly said that, when determining whether the damage could reasonably have been contemplated by the parties involves asking whether it was ‘on the cards’ at the time. On this matter, the majority identified two issues that needed to be determined about ‘what the defendant knows or must be taken to know.’ First, the court has to consider the level of generality at which the damage, and the process by which it occurred, should be described. Second, the court has to consider how demanding the test is when it asks whether the defendant had in contemplation that the damage was ‘on the cards.’

For the first issue, the majority said that the precise damage, and the precise manner of the occurrence of damage, need not have been contemplated. (As it happens, this principle reminds me somewhat of two unrelated but resonant principles from the law of negligence: first, that the precise sequence of events leading to a specific harm need not be foreseen for the harm to be understood as reasonably foreseeable: Hughes v Lord Advocate²¹; and second, that the causal mechanism for any injury needs to be characterised by the plaintiff at an appropriate level of generality: Tapp v Australian Bushmen’s Campdraft.²²) If the damage claimed is more particular, however, the majority found that it became more likely to be concluded that the damage was too remote.²³ This, it appears, is a spectrum-based test: the more general the damage, the more likely to be within the bounds of contemplation; but the more particular, the more likely it will be found to be outside those bounds.

For the second issue, the majority noted that Lord Reid had used the expression ‘on the cards’ (although not in its first utterance)²⁴ and then noted the ‘wide gulf’ between a ‘not unlikely’ or ‘quite likely’ event and something that is ‘a serious possibility’ or ‘real danger,’ or — more to the point — something that is ‘on the cards.’ Amusingly, the majority then repeated Lord Reid’s rather literal analysis of the metaphoric expression, wherein ‘the possibility of a nine of diamonds being the top card in a pack’ was said to have been a ‘serious possibility,’ notwithstanding that most people would not consider this to be ‘likely.’ Then, amidst this discussion of probability and card decks, the majority note that the expression has been criticised as imprecise and acknowledged there is force in the criticism. In this regard, the majority cite a High Court decision from the ’70s — Wenham v Ella (1972)²⁵ — but they also cite John Cartwright’s article from the 1990s, wherein the author is said to have preferred the expression ‘realistic level of foreseeability’ to ‘on the cards.’²⁶

The majority then considers the connection between the law of remoteness in breach of contract as having assimilated with the tests applicable in ‘reasonable foreseeability’ in the law of torts (reflecting the resonances I noted I perceived above). But the majority then go on to make clear the difference: ‘the present law treats the requirement of a serious possibility,’ it said, ‘as something that is more stringent than merely a reasonably foreseeable possibility that is not far-fetched or fanciful.’²⁷ All of this is by way of setting up the analytic framework for the remoteness analysis.

4.3.2. How could the remoteness test be applied to psychiatric injury?

The majority found that determining remoteness requires the court to consider, in turn:

• the circumstances of the breach;
• the general type of damage, together with
• the general manner of its occurrence; and
• whether the general type and manner of the damage was within the reasonable contemplation of the parties as a ‘serious possibility.’

In Elisha, the majority considered these factors as follows.

4.3.2.1. Circumstances

When the majority analysed the circumstances of the breach in this case (as considered by the courts below), it found those circumstances were ‘serious’ for several reasons. First, the disciplinary process was a ‘sham’ because the hotel manager’s version of events was preferred to Elisha’s prior to the meeting; second, the real reason for Elisha’s dismissal was not actually the hotel incident but a ‘pattern of aggression’ that had been alleged against Elisha by his manager. This reason had been shared by his manager to other Vision staff, both in senior leadership and HR, but ‘lacked any foundation [and] were never raised with Mr Elisha.’²⁸ Although the majority does not suggest it is necessary that the circumstance be ‘serious’ for an injury to be ‘on the cards’ or a ‘serious possibility,’ it may be that a more serious set of circumstances (as a matter of logic or common sense) will end up giving rise to stricter considerations about what the parties can be reasonably taken to have contemplated.

4.3.2.2. The general type of damage

In Elisha, the type of damage was psychiatric injury. But what does that mean? The majority found that psychiatric injury was ‘an illness which is a different type of damage from mere mental distress.’ Adopting a line from the reasons of Lord Justice Rix (who was dissenting) in Essa v Laing Ltd²⁹, the majority said that ‘injury to feelings’ was a ‘common-day experience’ distinct from a medical illness. A psychiatric illness, the majority continued — now citing the negligent mental harm case Tame v New South Wales ³⁰ — is a diagnosable medical condition. Finally, an assessment about the severity of a psychiatric illness, the majority said, would be ‘a statement about the extent of the injury’ and ‘not its type.’³¹ In other words, it seems that an extremely mild psychiatric illness is still a psychiatric illness; an extremely severe case of injured feelings would still be a case of injured feelings. As with negligently inflicted mental harm, the diacritical factor would appear to be medical evidence.

4.3.2.3. The general manner of its occurrence

The majority confirmed that the ‘precise manner’ in which the breach caused the injury need not have been contemplated by the parties. However, it did note that, in the circumstances of Vision’s dismissal of Elisha, that

an important element in the causal sequence by which Mr Elisha’s psychiatric injury occurred was that without Vision Australia’s breach, Mr Elisha would not have been dismissed for alleged misconduct. This causal element was entirely predictable in light of the nature of Vision Australia’s breach.

Although the majority do not refer by name to a formal causa sine qua non or ‘but for’ test of causation from the law of torts (negligence), the above line of reasoning indicates that considerations about the connection (or ‘causal sequence’) between the breach and the occurrence of the injury will assume greater importance if they follow the ‘but for’ or ‘factual’ pattern of causation.³² And this ‘important’ element will, it seems, assume greater significance still in the employment setting: ‘That causal element is significant,’ the majority stated, because (quoting from Lord Millet in Johnson v Unisys Ltd) ‘[m]any people build their lives round their jobs and plan their future in the expectation that they will continue. For many workers dismissal is a disaster.’³³

The majority then went on to outline the importance of employment as one of the most important conglomerations of interests in a person’s life. Quoting from Lord Hoffman (although not citing his Lordship specifically) in Johnson v Unisys Ltd (‘Unisys‘), the majority said it was a ‘social reality’ that a person’s employment ‘is usually one of the most important things in his or her life.’ The majority quote some of the following from Lord Hoffman, those sections being italicised. However, it is worth quoting a little more to shed light on the context in which the importance of employment is not merely a timeless phenomenon but one that has increasingly been recognised through legislation:

At common law the contract of employment was regarded by the courts as a contract like any other. The parties were free to negotiate whatever terms they liked and no terms would be implied unless they satisfied the strict test of necessity applied to a commercial contract. Freedom of contract meant that the stronger party, usually the employer, was free to impose his terms upon the weaker. But over the last 30 years or so, the nature of the contract of employment has been transformed. It has been recognised that a person’s employment is usually one of the most important things in his or her life. It gives not only a livelihood but an occupation, an identity and a sense of self-esteem. The law has changed to recognise this social reality. Most of the changes have been made by Parliament.³⁴

As noted above, as a background to the majority’s analysis of remoteness in Australia is its recognition that the statutory contexts in the UK and Australia are different; nevertheless, it appears that the recognition of employment as central to a person’s identity in both jurisdictions has been valorised not at common law but by statutory law. Although the majority do not refer to the Fair Work Act 2009 (Cth) to make the case that the termination of one’s employment affects central interests, the majority’s reasons call on conclusions of that character deriving from legislative developments in the UK. In what follows, the majority state the issue clearly: ‘An unfair process of termination for alleged misconduct could affect all three of those interests; ie, a person’s livelihood, identity, and self-esteem.’³⁵

As the analysis continues, however, the majority find that Vision’s DP 2015 also recognised the ‘social reality’ that the loss of employment could affect important interests. The DP 2015 required ‘due process’; provided for a support person to attend the disciplinary meeting; and Vision, the majority found, agreeing with the primary judge, had also established separate ‘processes of support, including counselling for employees, to anticipate and address risks of psychiatric injury, including in relation to the process of dismissal.’

4.3.2.4. Whether the general type and manner of the damage was within the reasonable contemplation of the parties as a serious possibility

The majority finally considered the nature of the impact of the breach, which was a psychiatric injury, and whether the psychiatric injury, and its cause, was reasonably contemplated by the parties as a serious possibility at the time the contract was entered into. They said that

The psychological impact of the breach that could reasonably be supposed to have been in the parties’ contemplation included not only the grave effect of Mr Elisha’s wrongful dismissal for alleged misconduct, but also the “unfathomable nature” of what occurred.³⁶

The majority adopted this expression — ‘unfathomable nature’ — from O’Meara J, the primary Judge. His Honour had used the expression to describe the unpredictable and surprising fact that his employment was terminated in the manner it was. His Honour had said as follows:

It was very clear that the unfathomable nature of what had occurred contributed very significantly to the disturbance of [Mr Elisha’s] mind and, consequently, the chronic psychiatric illness of which various medical witnesses gave evidence … [T]he psychiatric illness and disability of [Mr Elisha] likely built as time passed and as he continued to ruminate over what had occurred.³⁷

Thus, the fact that the termination was a surprise or shock (‘unfathomable’) appears to have been important in considerations of the ‘manner in which it occurred.’ Although the majority do not break down the characteristics of the ‘occurrence’ of the harm, it appears that where the harm is engendered by events that may come as a shock or surprise, this will enhance the likelihood that the harm will be considered a ‘serious possibility.’

The majority went on the note that, in terms of remoteness, the ‘precise psychiatric injury … need not have been contemplated at the time’ of the original contract.³⁸ It was therefore not required that Vision contemplated Elisha’s psychiatric injury at the time it offered employment to Elisha in 2006. However, the majority said it was

reasonable to expect that Mr Elisha would have been so distressed by the manner in which Vision Australia breached the 2006 Contract and by the consequences of the breach for him, including his dismissal for alleged misconduct from the employment that he had held for nearly a decade, that there was a serious possibility that Mr Elisha would suffer a serious psychiatric injury.³⁹

As I have suggested several times already in this blog post, there are curious similarities between the law of contract and the law of negligence with respect to the tests for remoteness of damage. However, it is also important to note the majority’s care not to invoke the tests of negligence here. While the word ‘reasonableness’ is used, the expression ‘reasonable foreseeability’ is not. The clear demarcation between the mental state in negligence and the mental state in breach of contract is as follows: for the negligence action, a not insignificant risk of harm must have been reasonably foreseeable before the breach; by contrast, for the breach of contract action to succeed, there must have been a ‘serious possibility’ of the injury in contemplation at the time of the execution of the contract.⁴⁰

The majority then explained why it departed from the reasoning of the Court of Appeal. It found that the Appeal Judges erred in two ways. Firstly, the majority rejected the conclusion reached by the Court of Appeal that the evidence of an expert, one Associate Professor Doherty, supported a finding that Elisha’s psychiatric injury was too remote because it was ‘extraordinary and not in keeping with the reasonable and … understandable effect of a sense of being terminated wrongly.’⁴¹ The majority found the expert’s evidence relied on a false premise — namely, that Elisha was ‘wrongfully terminated’ — and did not account for the breach of procedure (and breach of contract) by Vision.⁴² But it also noted that the primary judge had only used this expert evidence with reference to Elisha’s negligence claim, where the test for remoteness was different (as noted earlier). As the majority said, for negligence, the ‘reasonable foreseeability of psychiatric injury was to be assessed at the date of breach.’ By contrast, in a breach of contract claim, pursuant to the second limb of Hadley, the relevant test — whether there is a serious possibility of harm — is to be determined at the time of the contract being made or executed. That is when the parties must have had in reasonable contemplation the serious possibility of the kind of harm that ultimately came home.

The majority then quoted a crucial paragraph from the reasons of the primary Judge, O’Meara J, who had clearly concluded Elisha had been made the subject of a grossly unfair process (my words). As the Judge had said,

To conduct a sham process in which a worker is, in effect, convicted by reference to insubstantial secret slurs, and in a manner such that the worker perceives that something below the surface is ‘wrong’ but is therefore unable to be combatted, does seem to me to present an acute position of powerlessness in the worker that could well be at reasonably foreseeable risk of precipitating more than the usual degree of distress and even psychiatric injury.⁴³

This quotation presaged what the majority would then identify as the second error of the Court of Appeal: namely, that ‘its focus upon the breach was abstracted from the circumstances of the breach as found by the primary judge.’ Those circumstances were captured in the quotation above. While the Court of Appeal had correctly identified that the breach had occurred and had also correctly identified that some form of psychological impact could arise as a result of the breach (as a ‘possibility), it had ‘abstracted’ or drawn apart its breach analysis from the ‘circumstances of the breach’ analysis, thus neglecting the latter. The majority acknowledged that, in principle, there may be assumed to be ‘no serious possibility of psychiatric injury resulting from a breach of contract’ where a party ‘merely’ fails to ‘put some allegations’ to an employee.⁴⁴ However, the majority found that this description of the events was too skeletal to adequately particularise the events that occurred here: it ‘omits the relevant detail of the breach that actually occurred and the manner by which it caused psychiatric injury.’⁴⁵

4.4. Steward J’s dissent in relation to the remoteness of ‘psychiatric injury’

Interestingly, Steward J dissented in relation to the majority’s assessment of the remoteness of the psychiatric injury, finding that Elisha’s injury was too remote in this case, and that the Court of Appeal was correct to find in favour of Vision in this regard.

Steward J puts into more pellucid expression something like what I was discussing above when I suggested the majority’s reasons have an affinity with what might be called the Goldilocks rules of appropriateness when it comes to determining the generality with which to characterise harm. My observation was that the majority’s reasons seems to elucidate something of a ‘spectrum-based test’ where ‘the more general the damage, the more likely to be within the bounds of contemplation; but the more particular, the more likely it will be found to be outside those bounds.’ I also compared the majority’s reasons here with the rule the High Court laid down in Tapp v Australian Bushmen’s Campdraft stipulating that, in a claim for negligence, the causal mechanism described with respect to any injury needs to be characterised by the plaintiff at an appropriate level of generality. While these are my observations only — and observations given strictly in the form of comparative legal theory rather than a restatement of the law — it is fair to say that Steward J’s dissent only reinforces the value in comparing the jurisprudence of causation that has been developed separately in negligence and contract. Accordingly, I will quote from Steward J’s dissent liberally.

Steward J starts with the established⁴⁶ proposition that remoteness is an area of law where the rules are not rigid but ‘prima facie rules,’ and that they may be modified or displaced where awarding reasonable compensation and avoiding undue liability require such a course.⁴⁷ As a second line of reasoning, Steward J then focuses on the what I have called the Goldilocks rule, above. His Honour endorses the majority’s observation that ‘the classification of the subject matter of damages, including the level of generality with which the damage is described, is critical to the issue of remoteness.’⁴⁸

However, Steward J then expands on this observation, providing something of a grounding for the importance of classification: his Honour says that parties to a contract are ‘more likely to have in contemplation broadly characterised categories of damage than categories of damage that are more precisely confined.’⁴⁹ This logic, as I have already suggested, echoes the logic of the rule that requires plaintiffs to characterise the mechanism of causation in negligence law at an ‘appropriate’ level of generality: not too precise and not too broad. The same underlying problem demands the Goldilocks rule in both negligence and contract law. In causation for negligence, the retrospective question of whether a risk was reasonably foreseeable may depend on the degree of precision or generality with which that risk is characterised. In the same way, in remoteness for breach of contract, the retrospective question of whether the parties could reasonably be supposed to have had in contemplation when contracting that a type of injury was a ‘serious possibility’ may well depend on the degree of precision or generality with which that injury if classified.

To be sure, the rules are different in these different domains of law. However, the ‘forensic’ advantage and disadvantage that may be conferred by the ‘characterisation’ or ‘classification’ of, respectively, causal mechanisms in negligence or injuries in remoteness for breach of contract, is, I would suggest, the key reason why this Goldilocks principle must be applied. In each case, it is a control mechanism.

Steward J then quotes McHugh JA’s statement regarding damages in Alexander v Cambridge Credit Corporation Ltd (‘Cambridge CC‘):

The most difficult question in determining the relevant kind of damage concerns the level of classification of the damage which the parties must have contemplated. Clearly the level must not be so high that the parties are required to contemplate the very loss in question or the precise manner of its occurrence. Nor must it be so low that any loss or damage, no matter how unusual in nature or occurrence, would fall within the classification.⁵⁰

Following this principle, Steward J then states that a ‘middle course must be taken,’ although his Honour accepts that subsequent analyses can diverge: ‘where that leads in a given case can give rise to different conclusions, reasonably held.’ In short, Steward J disagrees with the majority’s classification of Elisha’s damage as psychiatric injury’ because, in his Honour’s view, this is ‘too broad, in the same way that the classification of “physical injury” would be too broad.’ His Honour then goes on to elaborate the complexities involved:

The sheer number of different kinds of psychiatric injury that might be suffered — for example, from mild depression to acute schizophrenia — justifies its rejection as an appropriate classification of damage, both generally and in the circumstances here, for the test of remoteness.

By way of supporting his position, Steward J cites several authorities, most important among them Rowe v McCartney (‘Rowe‘).⁵¹ As it happens, Rowe was a negligence case, although it was cited, Steward J notes, by McHugh JA in Cambridge CC as a case that posed a ‘similar problem’ to that raised by remoteness problems in breach of contract.⁵² (Steward J also cites Unisys and Goldman Sachs JBWere Services Pty Ltd v Nikolich as supporting his conclusion.⁵³)

In Rowe, the friend of a car owner crashed her car. The owner recovered damages from the driver for her physical injuries but was unable to recover damages for loss for depression (a ‘depressive neurosis’) expressed as ‘feelings of guilt in letting the friend drive the car.’ Steward J quotes Samuels JA (with whom Moffitt P agreed; Glass JA dissenting), who formed the view that this category of injury was too wide. Samuels JA stated that ‘in this case, it is necessary and legitimate to penetrate the categories [of mental illness] more closely.’ Upon doing so, Samuels JA found that the ‘harm suffered was, in my opinion, of an entirely different kind from that to which the defendant ought reasonably to have had regard as a likely consequence of his negligence.’ Steward J followed this negligence decision to find that

the type of harm which needed to be in the contemplation of the parties in 2006, when the employment contract was entered into, [was] ‘serious psychiatric injury’ as distinct from merely ‘psychiatric injury.’ Given that in 2006 — and, as it happens, thereafter — the respondent had no knowledge of any kind concerning the mental health of the appellant, and could not reasonably have acquired such knowledge, it cannot be said that the occurrence of a serious psychiatric injury, arising from the manner of terminating the appellant’s employment, could have been considered in 2006 to be a ‘serious possibility’ or ‘a real danger’ … . The appellant’s mental health was a ‘special
circumstance,’ about which the respondent must have known before or at the time of entry into of the contract of employment, in order for the damage suffered by the appellant not to be too remote.⁵⁴

On this last point about the special circumstance, Steward J cites, but does not analyse, two cases. The first is Victoria Laundry (Windsor) Ltd v Newman Industries Ltd,⁵⁵ an English breach of contract and remoteness case, where Asquith LJ in the Court of Appeal found a ‘super profit’ — anticipated to derive from a third party government contract expected to be entered into from the delivery of a boiler, and lost due to the non-delivery of the boiler (and thus the contractual breach of the supplier) — to be too remote. The second was Kollman v Watts.⁵⁶ That case involved a purchaser of a business who did not know when contracting that the seller needed money urgently to buy a house and might have to borrow money at a higher rate of interest if he was late in making payment; however, he did know of this circumstance when the purchaser breached the contract.⁵⁷ The Supreme Court of Victoria found that the relevant time at which a party must have in contemplation the relevant damage is at the time of entry into the contract.

Ultimately, Steward J found that the ‘severe psychiatric disability’ that Elisha had suffered was too remote from the breach of contract and Vision did not necessarily have this reasonably in contemplation as a serious possibility at the time of entering into the contract:

… it is one thing to contemplate that an employee might become ‘upset, stressed and disturbed’ as a result of an employer’s breach of an employment contract. It is quite another to contemplate that they would suffer a ‘severe’ psychiatric disability. In that respect, the categorisation of the type of damage which must be in the contemplation of the parties when the contract is entered into may be distinguished from the extent and seriousness of any damage within that category. The latter need not have been reasonably contemplated by the parties.⁵⁸

I must say that, to my ears, the last sentence is confusing, as it suggests that Steward J is laying down a rule as to what the law requires — and, stranger still, a rule that seems contrary to what the Justice has argued up to this point. However, when his Honour states that Vision ‘need not have’ had in contemplation the extent and seriousness of Elisha’s damage, I take Steward J to mean that, on the facts, Vision cannot necessarily be supposed to have had in reasonable contemplation the extent and seriousness of Elisha’s damage. That is, Elisha’s damage ‘need not have been [factually] reasonably contemplated by [Vision].’ In other words, we ‘need not’ conclude that such damage was contemplated. Thus, Steward J would have dismissed the appeal on the breach of contract claim.

4.5. Jagot J’s separate judgment

Given the dissent of Steward J on the remoteness point, it is worth briefly considering how Jagot J’s separate reasons made sense of remoteness. But before delving into that analysis, I will just note Jagot J’s conclusion as to the applicability of Addis. Jagot J’s reasons resembled those the of the majority (and of Steward J). Ultimately, her Honour found, that ‘it is not the law in Australia that the decision in Addis precludes an employee from recovering damages for ‘injured feelings.’⁵⁹ But Jagot J’s analysis is slightly more detailed in its discussion of the classification of mental harms in Australia between Addis (1909) and the current day. That history has been tied up with nervous shock — a classification of psychiatric damage considered in the High Court of Australia in a few tort cases after the decision of the House of Lords decision in Addis, such as Bunyan v Jordan⁶⁰ and later Chester v Waverley Council.⁶¹ While neither of those were considered here, they form the backdrop for the cases Jagot J does cite: namely, Baltic Shipping⁶² and Mount Isa Mines Ltd v Pusey (‘Mount Isa Mines‘).⁶³

4.5.1. Jagot J’s analysis of the compensability of psychiatric injury

Her Honour observed Mason CJ’s articulation of the exceptions to the general rule that ‘damages for breach of contract’ are ‘compensatory,’ are ‘confined to the award of that sum of money which will put the injured party in the financial position the party would have been in had the breach of contract not taken place,’ and that ‘anxiety and injured feelings do not … form part of the plaintiff’s compensable loss which flows from a breach of contract.’ The exceptions were that ‘a plaintiff can recover damages for pain and suffering, including mental suffering and anxiety, where the defendant’s breach of contract causes physical injury to the plaintiff.’ As Jagot J then notes, ‘in the related footnote,’ Mason CJ specifies ‘that the “class of physical injury for which damages are available includes nervous shock,”‘ relying on Mount Isa Mines. Jagot J then surveys the reasons in Mount Isa Mines, noting that Barwick CJ ruled out ‘schizophrenia’ as a psychiatric injury that could be compensated where it followed the shock of witnessing the aftermath of an accident; and that Windeyer J found that although ‘sorrow does not sound in damages,’ it was possible such injuries as ‘nervous shock’ and ‘severe emotional distress’ to precipitate ‘a lasting disorder of mind or body’ for which damages could be recoverable.

As it happens, I have written briefly on Windeyer J’s decision in Mount Isa Mines before, specifically his Honour’s oft-cited quotation that the law is ever ‘marching with medicine but in the rear and limping a little.’⁶⁴ My point in discussing that quote was incidentally to draw attention to the fact that Mount Isa Mines was a case that concerns psychiatric injury (rather than medicine in general). As I wrote,

It is fair to observe that Windeyer J’s quotation about law and medicine is frequently taken out of context. It is regularly used, for instance, to suggest the common law is merely responsive to medical innovation, trailing behind technology because its pre-established principles are incapable of encompassing … future innovations. Yet in the context of his Honour’s reasons, Windeyer J’s observation underlines the capability of law, specifically tort law (and particularly the action on the case) to embrace new medical knowledge in the right circumstances.

It is obvious Windeyer J’s comment laments the Privy Council’s rejection of the compensability of “psychogenic” injuries (as potentially “imaginary”) in Victorian Railways Commissioners v Coultas (Coultas). However, Windeyer J is careful to note that Dixon J, in Bunyan v Jordan, found the injury of nervous shock, rejected by the Privy Council, as ‘sufficient for the purpose of any action on the case in which damage is the gist of the action.’ The Dixon J passage is ‘noteworthy,’ Windeyer J said, because ‘it emphasizes [sic] that nervous shock cases are not a new tort.’ Thus, although the law may sometimes march ‘in the rear’ of medicine, this is not due to some essential lacuna of legal principle. Rather, the inability of the law to recognise medical novelty might arise for other reasons: not only, as Martha Chamallas and Linda Kerber suggest, because courts expressed a structural bias against certain classes of plaintiffs but perhaps also owing to the law’s reliance on biased evidence, as Ralph Harrington suggests. In this sense, the law might not so much trail behind medicine as sometimes wait for evidence corresponding to existing legal principle to materialise. In other words, if the law is ‘in the rear,’ it may be so because the law depends on medical knowledge to permit it to make findings of fact – knowledge that may or may not always be available.⁶⁵

My point in the above passage was to suggest that in cases where psychiatric injury (or medical injury more broadly) is claimed, the classification of the injury by the law will often depend on the medical expertise presented in evidence. The law is not necessarily slow; rather, the production of law-changing medical evidence at the appellate (precedent-setting) level can be slow. After all, most litigation in Australia is settled: between 67 and 94 per cent.⁶⁶ Indeed, As we have seen in Steward J’s dissent, the specificity of some medical diagnosis will suggest rarity; Steward J found Elisha’s injuries should be regarded as ‘severe’ and thus as a ‘special circumstance’ not necessarily contemplated by Vision at the formation of the contract. The law, for his Honour, should become more restrictive and should not recognise these injuries. This could be a case of the law catching up to the medicine of severe psychiatric injury. Indeed, Steward J described the injury as a ‘disability’ (incidentally, under article 1 of the Convention on the Rights of Persons with Disabilities, ‘psychosocial disabilities’ include ongoing mental conditions⁶⁷).

Given that a string of cases recognising ‘nervous shock’ and ‘injured feelings’ has developed in Australia, Jagot J’s reasons seem to seek to to reconcile the nature of the diagnosis adduced by Elisha in this matter — major depressive disorder (MDD), adjustment disorder and depressed mood — with this historical classification of nervous shock. In doing so, Jagot J writes as follows:

… it is not the law in Australia that the decision in Addis precludes an employee from recovering damages for ‘injured feelings’ if, by ‘injured feelings,’ what is meant is an ‘injury’ of the type formerly described as ‘nervous shock’ and its sequelae of ongoing psychiatric illness.

Thus, Jagot J recognises that an ‘ongoing psychiatric illness’ that results as a sequela of ‘injured feelings’ can be recoverable in damages. As Jagot J notes, this is not an issue in Australian law. Instead, ‘[t]he questions in Australia remain those of breach, causation of harm, and remoteness of damage in the contractual context.’⁶⁸

4.5.2. Jagot J’s remoteness analysis

Jagot J’s analysis of remoteness is consistent with the majority’s reasons in relying on the test from Hadley, and primarily the second limb. However, the key to Jagot J’s reasoning is that it draws attention to the ‘relationship between the concepts of the kind of breach that occurred and the kind of resulting injury, harm or loss to the employee being in the reasonable contemplation of the parties as a serious possibility at the time of entry into the contract,’ which her Honour says it ‘critical.’⁶⁹ In order to illustrate the relationship, Jagot J says that, where such a minor and innocuous breach of the Vision DP 2015 occurred (eg, where three members attended a disciplinary meeting rather than the two required by the contract), then it could not ‘reasonably have [been] contemplated as a serious possibility’ at the time of entering the contract that this would ’cause harm in the form of psychiatric illness’ (relying on the UK case Bristol City Council v Deadman⁷⁰). However, as Jagot J then stipulates, such an innocuous and minor breach was not in issue in this appeal. Rather, the breach here was more serious.

After detailing why the breach was more serious, her Honour then reframes the question:

… [T]he relevant question in the present case is not whether, at the time they entered the contract, Vision and Mr Elisha should reasonably have contemplated as a serious possibility that Vision acting in breach of any contractual term of its disciplinary procedures in any manner could result in Mr Elisha suffering any psychiatric illness. The relevant question is whether, at the time they entered the contract, Vision and Mr Elisha should reasonably have contemplated as a serious possibility that Vision acting in serious breach of its disciplinary procedures in a way that worked a serious injustice to Mr Elisha could result in Mr Elisha developing a serious psychiatric illness.⁷¹

As I read it, Jagot J’s reframing serves to emphasise the importance of particularising the circumstances of the breach; the general type of damage, together with the general manner of its occurrence; and, finally, pursuant to the second limb of Hadley, whether the general type and manner of the damage was within the reasonable contemplation of the parties as a ‘serious possibility.’

Her Honour then goes on to consider psychiatric illnesses more explicitly:

There are many different forms of psychiatric illnesses, with different diagnostic criteria. The focus of the law of remoteness on the contemplation of the ‘type’ of injury rather than the specific injury caused by the breach reflects a common-sense approach to the limited scope of human imagination and the unlimited scope of human experience. In the context of psychiatric illness caused by a breach of contract, it is one thing to accept that there is no difference in type between a serious psychiatric illness and a refractory serious psychiatric illness. It may be another, however, to accept no difference in type between, for example, contemplation of any diagnosable psychiatric illness, no matter how mild in nature (eg, a mild and transient anxiety disorder), and of a serious psychiatric illness (eg, major depressive disorder, schizophrenia). That parties to a contract should reasonably have contemplated the former as a serious possibility consequent on a kind of contractual breach does not, for the purposes of the test of remoteness, necessarily bring the latter within the parties’ contemplation. The observation of Wilcox J in Nikolich v Goldman Sachs J B Were Services Pty Ltd, that it ‘may be unusual for disturbance of mind to lead to a psychological condition as severe [or resistant to treatment] as that suffered … However, that is a statement about the extent of the injury, not its type,’ should not be understood as proposing that reasonable contemplation of a serious possibility of any psychiatric illness resulting from a contractual breach necessarily encompasses reasonable contemplation of every resulting psychiatric illness.⁷²

Where Steward J seems to rely on the negligence case of Rowe to find that ‘the type of harm which needed to be in the contemplation of the parties in 2006, when the employment contract was entered into, [was] ‘serious psychiatric injury’ as distinct from merely ‘psychiatric injury,’ Jagot J distinguishes Rowe based on the cause of the psychiatric illness in that case. As her Honour notes, the ’cause of [the] depressive illness’ in Rowe ‘was not the accident or the physical injuries that the plaintiff had suffered, but feelings of guilt for allowing the other person to drive the car with the result that he suffered a life-changing injury.’ In a sense, the guilt of the injured party in Rowe, which was, seemingly for Jagot J, a consequence of their own psychogenic neuroses, rather than the accident itself (ie, not a sequela), was different to the situation here, where Elisha’s depressive state was a sequela, without any intervening cause, of the impact of the termination of his employment. Indeed, this was so notwithstanding Elisha’s pre-existing psychiatric condition. By way of distinguishing Rowe and Elisha’s circumstances very clearly, Jagot J reasoned that something more would be needed for Elisha’s injuries to match those of the plaintiff in Rowe:

Translated to the circumstances of Mr Elisha’s psychiatric illness, it would be as if his illness had been caused not by the manner and fact of his unlawful summary dismissal but by anxiety and distress at, say, his parents’ reaction to his unlawful summary dismissal.⁷³

Ultimately, much like the majority, Jagot J finds that the Court of Appeal erred in two ways: first, by accepting Associate Professor Doherty’s evidence, which ‘was not based on a proper characterisation of the nature of the contractual breach’; and, second, in characterising the breach at ‘an impermissibly high level of generality.’⁷⁴

5. The negligence claim

The majority said that it was ‘strictly unnecessary’ to consider Elisha’s other ground of appeal concerning the tort of negligence; however, it acknowledged that the negligence claim ‘raises very different issues’ and proceeded to canvass them briefly.

It is trite to say that negligence involves proving: first, that there was a duty of care; second, that the duty was breached; and, third, that the breach caused the harm. As to the first element, the majority noted that the duty of care alleged by Elisha was a duty held by employers to provide ‘a safe system of investigation and decision-making with respect to discipline and termination of employment.’ Immediately with respect to the establishment of this duty, the majority acknowledged an important difference between duties of care as they can be established in negligence (which are established at common law) and as they can be established in a breach of contract action. Drawing on the well-established ratio of Sullivan v Moody,⁷⁵ and the much more recent case of Mallonland,⁷⁶ the majority noted that a duty of care in negligence ‘can be shaped by, and must be coherent with, relevant legislation.’ In other words, where there is a conflict between statute and an alleged duty of care, the latter is unlikely to be recognised.

The majority then went on the consider that Spigelman CJ (with whom Mason P and Giles JA agreed) had found, in New South Wales v Paige (‘Paige‘) in the New South Wales Court of Appeal, that an employer was not under a general common law duty of care to provide a safe system of work that encompassed the provision of ‘a safe system of investigation and decision making’ in respect of employees who are considered to be in breach of their terms of employment. The key feature of that conclusion was that imposing this duty on the employer would have been incoherent with employment law legislation.

Counsel for Elisha contended that the reasons of Paige were distinguishable because they fastened on incoherence between the scope of the duty of care and a statutory regime applicable to teachers (eg, under the Teaching Services Act 1980 (NSW)).

While accepting that the ‘statutory regulation of the modern-day employment contract has been treated as a significant matter curtailing the common law’s extension into the field,’ the appellant submitted that a ‘more precise analysis is required’ in this case.⁷⁷ The problem, however, was that no detailed arguments appear to have been advanced by Elisha with respect to the scope of the alleged duty and employment law (eg, the Fair Work Act 2009 (Cth)). As the majority said,

… no detailed consideration was given to the authorities concerning the scope and availability of compensation for psychiatric injury under the Fair Work Act, including in relation to the operation of provisions concerning adverse action, or the history and purpose of relevant provisions of the Fair Work Act insofar as they relate to the proposed scope of the duty of care to provide a safe system of work. Further, in oral submissions, issues were also raised by this Court concerning the coherence of the proposed duty with State workplace health and safety regimes. There was also little consideration given to whether the alleged duty of care was coherent with the employer’s contractual power to dismiss the employee.

Given these deficiencies in argument, the majority ended its consideration of the negligence claim — a consideration that it had been unnecessary to commence in any event. Jagot J, in her Honour’s separate judgment, similarly concluded it was necessary to consider the tort of negligence.⁷⁸

5.1. Steward J’ consideration of the negligence appeal ground

Since Steward J considered the appeal ground concerning breach of contract to have failed, his Honour went on the consider the negligence appeal ground in more detail than the majority and Jagot J. However, Steward J’s reasons were based on the same legal principle as those adverted to by the majority: namely, that there must be coherence as between a statutory regime and an imputed duty of care. Quoting from Sullivan v Moody, his Honour wrote that

[a] duty of care may not exist where it would otherwise ‘cut across other legal principles [so] as to impair their proper application’ or where it ‘would subvert … other principles of law, and statutory provisions, which strike a balance of rights and obligations, duties and freedoms.’⁷⁹

Of course, the specific duty of care advanced in a given matter may or may not be coherent with a given legislative regime. In Elisha, the proposed duty of care concerned ‘the incidents of his contract of employment’ and specifically ‘a duty of care to provide a safe system of work that extends to investigation and decision-making with respect to discipline and the termination of employment.’⁸⁰

As with the majority, Steward J found that Paige was ‘dispositive of this issue’ whether or not such a duty could exist.⁸¹ However, unlike the majority, which said that there were deficiencies in the plaintiff’s argument because ‘no detailed consideration was given to the authorities concerning the scope and availability of compensation for psychiatric injury under the Fair Work Act,’ Steward J found that the ‘proposed duty of care, including its coherence with existing law, was sufficiently argued before this Court such that it is appropriate to address it here.’⁸²

Steward J first distinguished the duty of care contended for here from the recognised duty in which an employer must provide a ‘safe system of work.’ That ‘system,’ his Honour noted, ‘is concerned with the performance by an employee of their work’ and the ‘decision of this Court in Kozarov v Victoria is a recent example of an application of that duty of care.’⁸³

Second, Steward J then considered Spiegelman CJ’s consideration in Paige of the four factors identified by the House of Lords in Unisys that meant there was ultimately no cause of action either in contract or in tort for loss flowing from psychiatric injury caused by the manner in which an employee was dismissed. Those factors consisted of the creation of a statutory adjudication process and the crafting of three statutory limitations, as follows:

• The creation of specialist tribunals as part of the UK’s legal system to hear and determine unfair dismissal cases
• The limitation of the class of applicants who could bring such an action (an ‘attempt to balance fairness to employees against the general economic interests of the community’);
• The limitation on the size of the awards that may be made; and
• The time limitations imposed on applications and concomitant limitations on the ‘workability of the duty or obligation contended for, in that causation would be almost impossible to establish with any certainty and would result in generalised awards, infringing the rule that there are no damages for dismissal per se’⁸⁴

Steward J acknowledged that the last point (concerning time limitations and causation) was not longer applicable in Australia and referred to the majority’s reasons as a source of that observation. The majority had recognised that four members of the High Court in Baltic Shipping had held that damages for psychiatric injury were available for breach of contract without any suggestion of an exception for employment contracts; and it appears that it is this development that Steward J indicates has overtaken the fourth ‘incoherency factor’ in Unisys.⁸⁵

As Steward J then observes, in Paige, Spigelman CJ sought to apply the factorial process undertaken in Unisys to the relevant NSW legislation so as to determine whether the proposed duty (to provide a safe system of work extending to investigation and decision-making involving disciplinary processes and dismissal) was coherent with it. In this regard, Spigelman CJ considered the Industrial Relations Act 1996 (NSW) and the Workplace Relations Act 1996 (Cth). After quoting from Spigelman CJ’s consideration of those legislative schemes, Steward J notes Spigelman CJ’s conclusion: namely, that the statutory regulation of unfair dismissal was ‘carefully calibrated’ to balance the conflicting interests of employers and employees and that, therefore, litigation about the termination of a contract of employment should be ‘left to the law of contract and statute, and no additional duty of care should be imposed over this extensive legal regime.’⁸⁶

As the majority had pointed out, however, the problem with applying Spigelman CJ’s reasoning in Paige to the present case was that Paige dealt with different legislation predating the Fair Work Act. But whereas the majority held that the appellant had advanced inadequate argument to assist the Court to make determinative conclusions as to whether the Fair Work Act might by incoherent (or not) with the proposed duty of care, Steward J was satisfied to apply Paige. As his Honour said,

Nothing has occurred since the decision in Paige that would compel any different conclusion today, save that, relevantly, the [Workplace Relations Act 1996 (Cth)] has been replaced by the Fair Work Act 2009 (Cth)… But nothing in the [Fair Work Act 2009 (Cth)] contradicts the reasoning of Spigelman CJ.⁸⁷

Steward J went on to identify four factors in the Fair Work Act corresponding to those identified in Paige regarding the Workplace Relations Act 1996 (Cth) and Unisys regarding the Employment Rights Act 1996 (UK). These concerned

• the unfair dismissal provisions;⁸⁸
• the statutory limitation imposed on those who can sue;⁸⁹
• the statutory limitations (cap) on compensation;⁹⁰ and
• the statutory time limit imposed on unfair dismissal actions.⁹¹

Notably, as Steward J states, s 392(4) of the Fair Work Act provides that any order to pay compensation to a person must not include a ‘component by way of compensation for shock, distress or humiliation, or other analogous hurt, caused to the person by the manner of the person’s dismissal.’ Furthermore, Steward J draws attention to the Explanatory Memorandum accompanying the Fair Work Bill 2008 (Cth) to explain this section, noting that it states the following:

Under subclause 392(4), any compensation ordered by [the Commission] must not include a component by way of compensation for shock, distress or humiliation caused by the manner of the person’s dismissal. This reflects the common law position that shock, distress or humiliation resulting from the dismissal is not compensable (Addis v Gramophone Co Ltd [1909] AC 488 and Baltic Shipping Co v Dillon (1993) 176 CLR 344).

Obviously, this is an important element of Steward J’s dissent, because it addresses an apparent incoherence that the majority does not: namely, that the Fair Work Act was drafted on the assumption that the common law would not and could not compensate for shock, distress or humiliation resulting from dismissal. The Explanatory Memorandum does not identify whether or not these matters are compensable in the circumstances of tort or breach of contract; nor does it clarify whether these terms — shock, distress or humiliation — would encompass psychiatric illness. Nevertheless, as Steward J’s proceeding observations suggest, there is now a tension in the law, because it appears that this section of the Fair Work Act was drafted on an erroneous assumption about the common law: namely, that it was not possible to compensate for psychiatric injury. However, Steward J does not take this tension as an occasion to find that coherence is possible (eg, because the statutory provision can be construed narrowly so as to exclude clinically diagnosed psychiatric injuries) but rather as an occasion on which to reaffirm there is incoherence between the asserted common law duty and the statutory scheme:

That Parliament has enacted complex legislation on the basis, at least in part, of an apparent understanding of a prior decision of the House of Lords (namely Addis v Gramophone Co Ltd) — which understanding is now considered by this Court to be incorrect — makes it all the more compelling that the duty of care to provide of work should not be extended in the way sought by the appellant.⁹²

Indeed, as Steward J then notes, the appellant had already availed himself of the scheme in this case, having instituted an action for unfair dismissal in 2015 — an action that was settled shortly after it was commenced for the maximum amount payable under the Fair Work Act (being an amount equal to 26 weeks of pay). Accordingly, Steward J ultimately finds that there cannot be coherence between the alleged duty to provide a safe system of work extending to investigation and decision-making involving disciplinary processes and dismissal and the Fair Work Act, especially where the full entitlements of the Fair Work Act had already been obtained and, as discussed above, a provision of that Act was drafted around a misunderstanding of the strictures of the common law. It seems that just because that understanding of the common law was wrong, that is not a reason to expand the common law now.

6. Conclusion

Elisha v Vision Australia Limited is an interesting case that may have significant implications for future proceedings against employers that (a) commit serious breaches of contract concerning the dismissal of employees; and (b) where those serious breaches cause the dismissed employee to sustain a psychiatric injury. Ultimately, the holding of the majority indicates that such contractual breaches will sound in damages where the employer, at the time at which they execute the contract with their employee, should reasonably have contemplated as a serious possibility that acting in serious breach of its procedures in a way the works a serious injustice to the employee could result in the employee developing a serious psychiatric illness. What is notable about the decision is that it carves out a space for the courts to award compensation for psychiatric injuries in employment termination settings where such a space did not exist before; that was because such a space was hitherto understood as precluded by the holding in Addis.

Although the majority and Jagot J do not entertain the merits of the negligence action, Steward J’s dissent raises important issues relating to incoherency between a proposed duty to provide a safe system of work extending to investigation and decision-making involving disciplinary processes and dismissal and the Fair Work Act. One issue is that it appears that s 392(4) of the Fair Work Act now appears to have been drafted under a misapprehension of the scope of the common law insofar as it understood that Australian courts could not award damages for psychiatric harms in dismissal contexts, whereas now, according to the majority, it can do so in certain breach of contract claims where those injuries are recognised psychiatric injuries (distinguishing them from hurt feelings). Whether future decisions of the High Court will follow Steward J’s dissent in terms of any negligence claims remains to be seen.

1. Elisha v Vision Australia Limited [2024] HCA 50 [1] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ) (‘Elisha HCA‘). ↩︎
2. Ibid [106] (Jagot J) ↩︎
3. Toll (FGCT) Pty Ltd v Alphapharm Pty Ltd (2004) 219 CLR 165, 179 [40] (Gleeson CJ, Gummow, Hayne, Callinan and Heydon JJ). ↩︎
4. Elisha HCA [42] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
5. Ibid [47] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
6. Ibid. ↩︎
7. Ibid [49] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Young v Chief Executive Officer (Housing) (2023) 278 CLR 208, 235 [69] (Gordon and Edelman JJ) and quoting Baltic Shipping Co v Dillon (1993) 176 CLR 344, 365 (Mason CJ) (‘Baltic Shipping‘). ↩︎
8. Elisha HCA [ [49] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Fidler v Sun Life Assurance Co of Canada [2006] 2 SCR 3, 16 [36], 20 [46] and referring to Young v Chief Executive Officer (Housing) (2023) 278 CLR 208, 235 [70] (Gordon and Edelman JJ). ↩︎
9. Baltic Shipping Co, 362 (Mason CJ), quoting from Guenter Heinz Treitel, The Law of Contract, 8th ed (1991), 878. ↩︎
10. Addis v Gramophone Company Ltd [1909] AC 488 (‘Addis‘). ↩︎
11. Ibid, 490–91 (Loreburn LC, with Lord James, Lord Atkinson, Lord Gorell and Lord Shaw concurring). ↩︎
12. Elisha HCA [53] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Addis, 493 Lord James), 496 (Lord Atkinson). ↩︎
13. Elisha HCA [55](Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Johnson v Unisys Ltd [2003] 1 AC 518, 541 [44] (Lord Hoffman) (‘Unisys’). ↩︎
14. Elisha HCA [55] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Eastwood v Magnox Electric plc [2005] 1 AC 503, 524 [12], 528 [27], 529 [32]–[33] (Lord Nicholls). ↩︎
15. Elisha HCA [56] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
16. Ibid [59] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Baltic Shipping, 362 fn 95, 383, 387, 405. ↩︎
17. Elisha HCA [62] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Hadley v Baxendale (1854) 9 Ex 341, 354; 156 ER 145, 151 (Alderson B). ↩︎
18. See Daniel P O’Gorman, ‘When Lightning Strikes: Hadley v. Baxendale’s Probability Standard Applied to Long-Shot Contracts’ (2016) 47(3) Loyola University of Chicago Law Journal 859, 867 https://lawecommons.luc.edu/luclj/vol47/iss3/6; also see Wayne Barnes, ‘Hadley v Baxendale and Other Common Law Borrowings From The Civil Law (2005) 11(2) Texas A&M Law Review 627, 632;
    Patrick S Atiyah, The Rise and Fall of Freedom of Contract (Oxford, 1979) 399; Robert J Pothier, A Treatise on Obligations, Considered in a Moral and Legal View, tr William David Evans (Robert H Small, 1806) 91. ↩︎
19. Elisha HCA [61]–[62] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing The Commonwealth v Amann Aviation Pty Ltd (1991) 174 CLR 64, 92 (Mason CJ and Dawson J.) ↩︎
20. Elisha HCA [126] (Jagot J). ↩︎
21. Hughes v Lord Advocate (1963) AC 837, 855 (Lord Guest). ↩︎
22. See Tapp v Australian Bushmen’s Campdraft & Rodeo Association Ltd (2022) 273 CLR 454, 490 [106]– 494 [119] (Gordon, Edelman and Gleeson JJ). ↩︎
23. On this point the majority cited H Parsons (Livestock) Ltd v Uttley Ingham & Co Ltd [1978] QB 791, 801, 804, 812–813; the majority also cites Andrew Burrows, Remedies for Torts, Breach of Contract, and Equitable Wrongs, 4th ed (Oxford University Press, 2019), 94–95. ↩︎
24. Elisha HCA [64](Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Koufos v C Czarnikow Ltd [1969] 1 AC 350, 390 (Lord Reid); the majority cites the earlier usage of the expression in Victoria Laundry (Windsor) Ltd v Newman Industries Ltd [1949] 2 KB 528, 540. ↩︎
25. Wenham v Ella (1972) 127 CLR 454, 471 (Gibbs J). ↩︎
26. John Cartwright, ‘Remoteness of Damage in Contract and Tort: A Reconsideration’ (1996) 55 Cambridge Law Journal 488, 494 fn 23. ↩︎
27. Elisha HCA [64] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), citing Alexander v Cambridge Credit Corporation Ltd (1987) 9 NSWLR 310, 365 (McHugh JA) and referring to Astley v Austrust Ltd (1999) 197 CLR 1, 27–28 [60] (Gleeson CJ, McHugh, Gummow and Hayne JJ). ↩︎
28. Elisha HCA [65](Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
29. Ibid [65](Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ); Essa v Laing Ltd [2004] ICR 746, 780 [117] (Rix LJ, dissenting). ↩︎
30. Tame v New South Wales (2002) 211 CLR 317, 382–383 [194] (Gummow and Kirby JJ). ↩︎
31. Here the High Court majority cited Goldman Sachs JBWere Services Pty Ltd v Nikolich [2007] FCAFC 120 at [74] (Black CJ); Nikolich v Goldman Sachs J B Were Services Pty Ltd [2006] FCA 784 [330] (Wilcox J); it also referred to Mount Isa Mines Ltd v Pusey (1970) 125 CLR 383, 402, 413–414 (Walsh J). ↩︎
32. The classic case regarding the Australian approach to ‘but for’ causation is March v Stramare (E & MH)
    Pty Ltd (1991) 171 CLR 506 wherein, at 522, Deane J said ‘The “but for” (or “causa sine qua non”) test may well be a useful aid in determining whether something is properly to be seen as an effective cause of something else … [and] will commonly exclude causation for the purposes of the law of negligence if the answer to the question it poses is that the accident which caused the injuries would have [still] occurred in the same way and with the same consequences in any event.’ ↩︎
33. Elisha HCA [67] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), quoting Johnson v Unisys Ltd [2003] 1 AC 518, 547 [72] (Lord Millet). ↩︎
34. Elisha HCA [67] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), quoting Johnson v Unisys Ltd [2003] 1 AC 518, 547 [35] (Lord Hoffman) ↩︎
35. Elisha HCA [67] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
36. Ibid. ↩︎
37. Ibid ↩︎
38. Ibid. ↩︎
39. Ibid. ↩︎
40. See, eg, Civil Liability Act 2002 (NSW) s 5B. ↩︎
41. Elisha HCA [70] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ), quoting Vision Australia Ltd v Elisha (2023) 328 IR 299, 339 [185] (O’Meara J). ↩︎
42. Elisha HCA [71] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
43. Ibid. ↩︎
44. Ibid [71] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
45. Ibid. ↩︎
46. Ibid [82] (Steward J); Steward J quotes Walsh J in Wenham v Ella (1972) 127 CLR 454, 466 (Walsh J), and cites Monarch Steamship Co Ltd v Karlshamns Oljefabriker (A/B) [1949] AC 196, 223 (Lord Wright). ↩︎
47. Elisha HCA [82] (Steward J). ↩︎
48. Ibid [83] (Steward J). ↩︎
49. Ibid. ↩︎
50. Alexander v Cambridge Credit Corporation Ltd (1987) 9 NSWLR 310, 366 (McHugh JA) (‘Cambridge CC‘). ↩︎
51. [1976] 2 NSWLR 72. ↩︎
52. Cambridge CC, 366 (McHugh JA) ↩︎
53. See Johnson v Unisys Ltd [2003] 1 AC 518; Nikolich v Goldman Sachs J B Were Services Pty Ltd [2006] FCA 784, [330] (Black CJ, with whom Marshall J agreed). ↩︎
54. Elisha HCA [90] (Steward J). ↩︎
55. [1949] 2 KB 528. ↩︎
56. [1963] VR 396. ↩︎
57. For a brief note on this case, see Adam Kramer’s chapter on remoteness in The Law of Contract Damages, 3rd ed (London: Bloomsbury, 2022) 295. ↩︎
58. Elisha HCA [90] (Steward J). ↩︎
59. Ibid [125] (Jagot J). ↩︎
60. Bunyan v Jordan (1936-1937) 57 CLR 1. ↩︎
61. Chester v Council for the Municipality of Waverley (1939) 62 CLR 1. ↩︎
62. Baltic Shipping Co v Dillon (1993) 176 CLR 344 ↩︎
63. Mount Isa Mines Ltd v Pusey (1970) 125 CLR 383. ↩︎
64. See Christopher Rudge, ‘Challenging Pandemic Law: From Vaccine Mandates to Judicial Review of Vaccine Approvals’ (2024) 31(2) Journal of Law and Medicine 230. ↩︎
65. Ibid (footnotes omitted). ↩︎
66. See Yun-chien Chang and Daniel Klerman, ‘Settlement Around the World: Settlement Rates in the Largest Economies’ (2022) 14(1) Journal of Legal Analysis 80 https://academic.oup.com/jla/article/14/1/80/7114596. ↩︎
67. Convention on the Rights of Persons with Disabilities, opened for signature 30 March 2007, 2515 UNTS 3 (entered into force 3 May 2008) art 17 (‘CRPD’). ↩︎
68. Elisha HCA [125] (Jagot J). ↩︎
69. Ibid [128] (Jagot J) (my emphasis). ↩︎
70. Bristol City Council v Deadman [2007] IRLR 888, 894 [46] (Moore-Bick LJ). ↩︎
71. Elisha HCA [130] (Jagot J). ↩︎
72. Ibid [133] (Jagot J). ↩︎
73. Ibid [135] (Jagot J). ↩︎
74. Ibid [136], [137] (Jagot J) ↩︎
75. See Sullivan v Moody (2001) 207 CLR 562, 579–580 [50]. ↩︎
76. See Mallonland Pty Ltd v Advanta Seeds Pty Ltd (2024) 98 ALJR 956, 967 [37]; 418 ALR 639, 649 (. ↩︎
77. See Adam Elisha (Appellant), ‘Appellant’s Submissions’, Submission in Elisha v Vision Australia Limited, M22/2024, 22 April 2024 [50]–[56]. ↩︎
78. Elisha HCA [138] (Jagot J). ↩︎
79. Ibid [94] (Steward J), quoting Sullivan v Moody (2001) 207 CLR 562 at 576 [42] (Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ). ↩︎
80. Elisha HCA [92] (Steward J). ↩︎
81. Ibid. ↩︎
82. Ibid. ↩︎
83. Ibid [92] (Steward J). ↩︎
84. Ibid [97] (Steward J); New South Wales v Paige (2002) 60 NSWLR 371 at 397 [142] — [145] (Spigelman CJ), citing Johnson v Unisys Ltd [2003] 1 AC 518, 543 [54]. ↩︎
85. Elisha HCA [51] (Gageler CJ, Gordon, Edelman, Gleeson and Beech-Jones JJ). ↩︎
86. Ibid [99] (Steward J), quoting Paige, 400 [154] — [155] (Spigelman CJ). ↩︎
87. Elisha HCA [100] (Steward J). ↩︎
88. An application for a remedy for unfair dismissal can be made to the Commission and the Commission can order reinstatement or a payment of compensation: Fair Work Act 2009 (Cth) s 394(1); div 4 pt 3-2 ch 3 (‘Fair Work Act‘). ↩︎
89. Section 380 of the Fair Work Act provides that a ‘national system employee’ may bring an action for unfair dismissal against a ‘national system employer’; however, as Steward J states, the class of employees who may sue is limited: see Elisha HCA [100] (Steward J) . ↩︎
90. Sections 392(5) and (6) of the Fair Work Act impose a cap on the amount of compensation that can be paid to a person for unfair dismissal. As Steward J states, s 392(4) provides that any order to pay compensation to a person must not include a ‘component by way of compensation for shock, distress or humiliation, or other analogous hurt, caused to the person by the manner of the person’s dismissal’: see Elisha HCA [100] (Steward J). ↩︎
91. Section 394(2) of the Fair Work Act stipulates that unfair dismissal actions must be commenced within 21 days after the time when the dismissal took effect or within such further period as the Commission allows. ↩︎
92. Elisha HCA [102] (Steward J). ↩︎