Dr V Powell v University of Portsmouth & Anor

This appeal raises questions as to the approach to be taken by the Employment Tribunal (“ET”) when determining claims made under sections 15, 20 and 21 of the Equality Act 2010 (“EqA”), and as to the test to be applied when considering, in the context of a claim of constructive unfair dismissal, whether there has been a breach of the implied obligation of trust and confidence.

I refer to the parties as the claimant and as the first and second respondents, as below. This is my judgment on the claimant’s appeal against the reserved decision of the Employment Tribunal sitting at Southampton (Employment Judge Gray, sitting with members Ms Lloyd-Jennings and Mr Spry-Shute, over some eight days between 26 April and 6 May 2022), sent to the parties on 8 June 2022. The ET was concerned with two separate claims made by the claimant: the first on 3 May 2019, which made complaints of disability discrimination against the first respondent at a time when the claimant continued to be employed; the second, presented on 3 January 2020, after the claimant’s employment came to an end, by which the claimant made claims of unfair constructive dismissal, disability discrimination, and victimisation. After some six case management hearings, the full merits hearing took place over eight days, with the ET hearing from the claimant and from seven witnesses for the respondents. By its reserved judgment, the ET dismissed each of the claimant’s claims, explaining its decision in written reasons of 64 pages, 334 paragraphs.

The claimant’s appeal was initially considered on the papers by His Honour Judge James Tayler, who took the view it revealed no arguable question of law. After a hearing under rule 3(10) Employment Appeal Tribunal Rules 1993 before His Honour Judge Auerbach, at which the claimant was represented by counsel acting under ELAAS, the appeal was permitted to proceed on amended grounds, as follows (I summarise):

At the outset of the full hearing of the appeal, the claimant made an application to adduce a document that had not been disclosed or relied on before the ET or at any prior stage in the proceedings. Having regard to the relevant provisions at paragraph 8.12 of the EAT Practice Direction 2023 and to the test laid down in Ladd v Marshall [1954] 1 WLR 1489 CA, I refused that application for reasons given orally at the hearing.

The claimant started his employment with the first respondent on 1 October 2006, initially part-time but subsequently working on a full-time basis, first as a Senior Lecturer and then, from 1 August 2017, as a Principal Lecturer, working within the School of Creative Technologies. The second respondent is Dean of that School.

The academic year at the first respondent is comprised of semesters, with the first and second 12-week semesters involving more face-to-face scheduled teaching. There was an issue between the parties as to the proportion of the claimant’s role that related to teaching: the claimant contended his role was split 50/50 between teaching and research, with a greater leaning to the latter; the respondents’ case was that teaching was his principal role. In support of his case, the claimant relied on the fact that his job title was “Principal Lecturer – Research Lead”; the evidence of the respondents was, however, that, as it was focused on teaching and industry practice, research was only a very small part of the overall School and the time allocated to fulfil this aspect of the claimant’s role was minimal (80 hours a year). On this dispute, the ET accepted the respondent’s case, finding that teaching was the claimant’s principal role, certainly in the first semester, which was the time focus of the disability discrimination claims.

It was common ground between the parties that the claimant was a disabled person for the purposes of the EqA: from August 2017, he was disabled by reason of stress and depression; from February 2018, he was also disabled due to a cardiac impairment, described as syncope/blackouts/bradycardia and cardiac prodromal symptoms, albeit the claimant did not receive a diagnosis in this respect until December 2018.

The claimant had taken time off work due to stress in late 2017 and early 2018, and, on 31 January 2018, his GP had signed him off for a month as unfit for any work on grounds of mood disorder. It was not in dispute that these periods of absence had been very disruptive in terms of the claimant’s teaching commitments, and, by email of 30 January 2018, the School Manager had flagged up the need to give thought to how the affected units would be delivered the following year.

It was in February 2018 that the claimant’s cardiac impairment initially manifested itself, when he suffered a collapse requiring emergency treatment. By an occupational health (“OH”) report dated 22 February 2018, it was recorded that the claimant:

By 10 May 2018, the OH advice continued to be that the claimant remained unfit for normal duties and hours; whilst the frequency of the blackouts had decreased, they remained unpredictable and very disruptive to his ability to function. It was proposed, however, that a phased return to work should be considered in August 2018 if the claimant’s health had improved (as was expected).

The suggestion of a phased return to work was the subject of further OH advice in a report of 5 July 2018, at which stage it was recorded that the claimant was continuing to suffer blackouts, although these were less frequent and intense. It was noted that the cause of the claimant’s symptoms had not been identified and no treatments had been initiated:

In considering a return to work, the report of 5 July 2018 opined as follows:

The report went on to consider workplace adjustments for such a phased return, advising:

The report of 5 July 2018 was not seen by the first respondent until 3 August 2028. The interim Head of School, Mr Kevin Curtis, had, however, met with the claimant on 5 July, at which point the claimant expressed the view that it would be unlikely that he would be able to return to face-to-face teaching before Christmas. A phased return to some of the claimant’s duties was, however, agreed, on the basis that he would start working four hours a day.

The claimant was on annual leave from 9 July to 1 August 2018, but met again with Mr Curtis on 2 August 2018 (albeit Mr Curtis had still not seen the 5 July OH report), when the claimant explained that his condition had deteriorated; he was still having blackouts, was considering seeking a permanent reduction in hours, and did not feel he would be safe in front of a class. As Mr Curtis recorded in his follow-up email to the claimant the same day, although the claimant had managed some PhD supervision, he had not been in contact with any of the course leaders of the 2019 curriculum or carried out other aspects of his role. It was, however, agreed that the claimant would continue to try to work on a four-hours a day phased basis, which (as was common ground) he did between 2-3 and 6-9 August 2018.

The claimant was again on annual leave from 15 August to 7 September 2018. On 5 September 2018, Mr Curtis emailed occupational health (copied into the claimant), raising his concerns about the claimant’s ability to return to work, and ending:

As the ET recorded Mr Curtis’ evidence:

The ET noted that it was not disputed that the claimant’s absence from teaching was problematic for the first respondent and it accepted that it was reasonable for Mr Curtis to take the view that the claimant was required to carry out some teaching as this was a key ingredient of his role; as the ET recorded his evidence:

Mr Curtis also gave evidence about the possible adjustments that had been referenced. He rejected the suggestion that someone might have been engaged to accompany the claimant, saying this was not feasible, in particular (even if someone could have been found) as the individual in question would have had to wait for several hours in between different seminars and lectures, given the (inevitably) disjointed nature of a lecturer’s timetable. Accepting that a preferable option would have been to give the claimant a break from teaching, to let others cover this work whilst the claimant focused on non-teaching duties, Mr Curtis’ evidence was that he considered that it would be better for the claimant to have a complete break; as he had stated to occupational health at the time (again as noted by the ET):

Although the July OH report had suggested that a risk assessment be carried out to try to mitigate possible causes of stress in the workplace (that being seen as a possible cause of blackouts), the ET noted that this was on the assumption that the claimant would return to his normal hours by the end of August, which had not happened. The ET again accepted Mr Curtis’ evidence, that such an assessment would have been very difficult absent the claimant’s attendance at work and that he had reasonably understood that the claimant was still not fit enough for this at that time.

A further OH report was produced on 6 September 2018 - albeit that was not released to the respondents at the time – in which it was recorded that the claimant had reported continuing to have regular blackouts, which remained “very unpredictable”, and that he had “only been functioning in a limited sense”. During the OH assessment for the purposes of that report, Mr Curtis’ email of 5 September 2018 had been read to the claimant, who was not recorded as expressing any disagreement with what had been said. In any event, the 6 September report advised that:

As for possible adjustments, it was observed:

As the OH report of 6 September 2018 was not released to the respondents, there then followed a period until the next review, scheduled for late October 2018, during which the claimant continued to work from home “in a limited fashion”, with no teaching.

A further OH report was produced on 25 October 2018, which recorded that the claimant had still received no formal diagnosis and was continuing:

The report advised that the claimant still had:

On 25 October 2018, the claimant was advised that he had 31 days of full sick-pay left from 10 June 2018 and that, once that was exhausted, he would then move to half pay. On 7 November 2018, the claimant confirmed to the first respondent that his GP would not sign him off work as he (the GP) considered this to be an employment decision for OH, not a medical decision. On 8 November 2018, Mr Curtis received an email from OH, advising that, following the assessment of the claimant on 25 October, it was concluded that:

In the early hours of 3 December 2018, the claimant suffered a severe prolonged blackout and was called into the cardiac clinic for review that afternoon. He suffered two further cardiac standstill events that evening. Subsequently, the claimant contacted Mr Curtis (and OH) to advise that he would be having a pacemaker fitted.

On 13 December 2018, an updated OH report was produced, recording that the claimant had now received a diagnosis, described as “intermittent cardiac standstill”, and had had a pacemaker fitted, which had led to an immediate improvement in his symptoms such that he was then fit to return to work on a phased basis, building his hours up gradually, with a view to returning to his full-time hours by mid-January 2019 without any further adjustments.

Subsequently, a further OH report was produced, on 18 March 2019, which expressed the opinion that the claimant was not a disabled person for the purposes of the EqA. That report was forwarded to the second respondent on 8 May 2019, with the claimant’s annotations and further information (which he contended supported a contrary conclusion).

In May 2019, the second respondent was involved in addressing a grievance that the claimant had pursued, sending out his decision report on 29 May 2019, when the claimant was again signed off work due to ill health. On 8 July 2019, the claimant lodged an appeal against that decision and there then followed correspondence between the claimant and the second respondent, relied on by the claimant as acts of victimisation (it was his case that the second respondent refused to meet him because he had undertaken protected acts); the ET rejected that case and it is not a live issue on this appeal.

As part of his grievance appeal, the claimant objected to an observation made by the second respondent in his report, where he had stated:

In giving evidence to the ET, the second respondent disavowed any such belief, albeit he accepted his report was poorly worded and clarified he had not been using the term “reasonable adjustments” in the sense meant under the EqA.

On 3 October 2019, the claimant received a package of documents from the first respondent relating to his grievance appeal. It was the claimant’s case that these contained further examples of victimisation and discriminatory attitudes by the respondents, including (as recorded by the ET):

As the ET noted, the claimant’s objection was that he had expected the respondents to accept his position on matters; by not doing so, he considered the respondents’ conduct was intended to fundamentally breach his contract. The ET disagreed, finding that the respondents were expressing what they believed the position to be at the time; the appeal hearing would provide the opportunity for the difference of views to be aired. Before that could happen, on 8 October 2019, the claimant resigned, giving two months’ notice.

The ET found that the first respondent had had constructive knowledge that the claimant was disabled by 3 August 2018, and that it was similarly aware of the disadvantage he suffered of not being able to cover face to face teaching.

There was no dispute before the ET that the claimant’s disability gave rise to an inability for him to teach in a classroom, and it was similarly accepted that, for the relevant period, he had been required to continue to teach in a classroom for a set number of hours, and that the first respondent had determined that he was unable to do so. The first respondent conceded that this requirement amounted to unfavourable treatment arising from something – the claimant’s inability to teach in a classroom at that time – arising from his disability. As the ET accepted, however, given the OH advice and the deterioration of the claimant’s condition during 2018, the first respondent had understandably taken the view that the claimant’s debilitating blackouts represented a risk to him, the students, and the university more generally. It was, further, the first respondent’s case that the treatment of the claimant had been a proportionate means of achieving a legitimate aim.

It was not in dispute that the provision of consistent high-quality teaching, in accordance with the prospectus and the expectations of students, had been a legitimate aim for the first respondent. The ET further accepted that teaching was the claimant’s principal role, certainly in the first semester, and that central to his role was the requirement that he prepare for, and deliver, 10 hours face-to-face teaching each week in the period October to March. More particularly, the ET accepted the evidence of Mr Curtis, to the effect that, in September 2018, the School of Creative Technologies was ensuring it was ready to begin teaching, while the claimant was still unable to complete the tasks set for two previous attempts at a phased return. The ET was satisfied that Mr Curtis was balancing a concern for the claimant’s wellbeing (and the risk of his overcommitting himself) with the need to provide a consistent, high-quality experience for the students, avoiding further disruption. At the same time, the ET found that Mr Curtis had to ensure that other staff – already working beyond their job roles to cover for the claimant – were not impacted further.

As for the alternatives posited as part of the claimant’s case, the ET again accepted Mr Curtis’ evidence, finding:

The ET further considered that the suggestion of a support worker was not just impracticable from the first respondent’s perspective, it also would not have had the effect of ameliorating the effect of the claimant’s disadvantage as:

As for the claimant’s argument that undertaking a risk assessment in relation to his blackouts might have been a reasonable adjustment, the ET: (1) noted that, as a matter of law, the duty on an employer is to act, not to consult, and, as such, a risk assessment did not amount to a reasonable adjustment; but, in any event, (2) agreed with Mr Curtis’ evidence, that this was very difficult to undertake absent the claimant’s attendance at work and that such an assessment would have a time element and, if the claimant was not fit, then whether it was worth doing at that stage was a relevant consideration.

Carrying out the required balancing exercise, the ET concluded as follows:

Having found that the first respondent’s treatment of the claimant was justified for the purposes of section 15 EqA, the ET then turned to the claimant’s reasonable adjustments claim in relation to the same conduct and period. In this regard, it was not in dispute that the requirement to provide classroom-based teaching (that is, face-to-face with students, classroom and lecture style) amounted to a provision, criterion or practice (“PCP”). It was also accepted that not being able to teach unaccompanied due to blackouts and/or a risk of blackouts was a substantial disadvantage. The ET recorded that the claimant had posited three steps that he contended could have been taken to avoid the disadvantage: (1) no large class teaching or lectures but allowing the claimant to undertake final year supervision and postgraduate teaching; (2) providing someone to accompany the claimant when teaching; (3) undertaking a risk assessment in relation to the blackouts and another risk assessment in relation to stress.

The ET noted that the claimant had accepted that, given the OH advice, as from 25 October 2018 there were no reasonable adjustments appropriate as he had been determined to be unfit for all work until 7 December 2018; the relevant period was thus from 5 September 2018 to 24 October 2018. Referring back to its earlier findings, the ET was satisfied that, for this period, teaching had been the claimant’s primary role.

Considering then the content of the email of 5 September 2018, the ET found this set out what Mr Curtis had known at the time, which was, in the main, based on what the claimant had told him:

More than that, the ET accepted Mr Curtis’ assessment as set out in his 5 September email, taken together with the OH report of 6 September:

In these circumstances, the ET concluded that the adjustments suggested by the claimant would not have assisted and, therefore, it would not have been reasonable to expect the first respondent to put these in place. Further finding that, as a matter of law, the duty on an employer is to act, not to consult, the ET held that conducting a risk assessment was a reasonable adjustment. In any event, the ET concluded that the suggested adjustments would not have been reasonable for the reasons it had already provided in considering the issue of justification under section 15 EqA. Accepting that the OH report of 5 July 2018 had anticipated a more positive trajectory for the claimant, the ET found that his health had in fact deteriorated and, on the same day, he had told Mr Curtis he would be unable to teach before Christmas. The ET was clear: it was not a reasonable adjustment for the claimant to remain at work without teaching for this extended period; in the circumstances, it did not find the first respondent had failed in its duty to make reasonable adjustments.

In pursuing his claim of constructive unfair dismissal, it was the claimant’s case that the respondents had acted in fundamental breach of contract in relation to the implied term of trust and confidence. In this regard, the claimant relied on the matters detailed in his claims of disability discrimination and on the contents of the documents disclosed to him on or around 3 October 2019 (relating to his grievance appeal).

In terms of the approach it was required to adopt, the ET directed itself as follows:

In cross-examination, however, the claimant’s case had been clarified as follows:

Considering the last act relied on by the claimant – the content of the documents he received on or around 3 October 2019 - the ET did not accept this amounted to a repudiatory breach of contract:

The ET then considered whether the content of the documents relied on by the claimant was, nevertheless, part of a course of conduct comprising several acts and omissions which, viewed cumulatively, amounted to a repudiatory breach of trust and confidence. Again it answered that question in the negative:

The ET therefore rejected the claimant’s contention that he had been constructively dismissed.

By section 15(1) EqA it is provided that:

In respect of the appeal relating to the claimant’s claim under section 15, the focus is on the question of justification – whether the first respondent was able to show its treatment of the claimant was a proportionate means of achieving a legitimate aim.

When determining whether the impugned treatment is thus justified, an ET should keep in mind the principles set out by Elias J (as he then was) in MacCulloch v ICI [2008] ICR 1334 EAT (subsequently approved by the Court of Appeal in Lockwood v DWP [2014] ICR 1257), namely that (I summarise): (1) the burden of proof is on the respondent; (2) the treatment must correspond to a real need, be appropriate with a view to achieving the objectives pursued, and reasonably necessary to that end; (3) this involves the application of the proportionality principle, which requires an objective balance to be struck between the discriminatory effect of the measure and the needs of the undertaking: the more serious the disparate adverse impact, the more cogent must be the justification for it; (4) it is for the ET to weigh the reasonable needs of the undertaking against the discriminatory effect of the employer’s measure and to assess whether the former outweigh the latter: there is no “range of reasonable response” test in this context.

As Lady Hale observed at paragraph 22 of Homer v Chief Constable of West Yorkshire Police [2012] UKSC 15; [2012] ICR 704, for a measure to be “proportionate”, it:

It is for the ET to reach its own judgment as to whether a measure is “proportionate” for these purposes, albeit that it is to do so:

As for the evidence that will need to be adduced to demonstrate the impact relied on by an employer, at paragraph 45 O’ Brien V Bolton St Catherine’s Academy [2017] ICR 737 Underhill LJ observed:

The important nature of the ET’s role in determining questions of objective justification was considered by Pill LJ in Hardys & Hanson plc v Lax [2005] EWCA Civ 846 ; [2005] ICR 1565, who provided the following observations as to the respective functions of the first-instance tribunal and the appellate tribunal:

By section 20 EqA, it is provided (relevantly):

Section 21 then provides:

As Lord Toulson observed at paragraph 83 FirstGroup Ltd v Paulley [2017] UKSC 4; [2017] IRLR 258, the concept of “reasonable adjustments” for these purposes is “intensely practical”. More specifically, the duty on an employer is to act, not to consult (albeit good practice will generally require consultation); see Tarbuck v Sainsbury Supermarkets Ltd [2006] IRLR 664 EAT.

Noting that the test of the “reasonableness” of any step is an objective one, dependent upon the circumstances of the case, paragraph 6.28 of the Equality and Human Rights Commission Code of Practice on Employment provides examples of potentially relevant factors, as follows:

The objective assessment that is required in order to determine “reasonableness” requires that the ET’s focus must be on the practical result of the measures that might have been taken, not on the process of reasoning by which a possible adjustment was considered (see Royal Bank of Scotland v Ashton [2011] ICR 632 EAT; an approach endorsed by the Court of Appeal in Owen v Amec Foster Wheeler Energy Ltd and anor [2019] ICR 1593). In particular, the essential rationale of section 20 is for adjustments to be made that are effective in keeping a disabled person in employment. Thus a question for the ET is how the step(s) proposed would have been effective to enable the disabled person to work, albeit it will be sufficient if there was a prospect of the disadvantage being alleviated even if the adjustment in question would not have been completely effective; as Elias LJ observed in Griffiths v Secretary of State for Work and Pensions [2017] ICR 160 CA:

The tests to be applied under section 15 and under section 20 EqA are different, although there may be a close relationship between the considerations that arise in any particular case (see, by way of example, the discussion at paragraphs 50-55 Monmouthshire County Council v Harris UKEAT/0332/14).

As for the claimant’s complaint of constructive unfair dismissal, it was his case that the first respondent had acted in breach of the implied obligation not, without reasonable and proper cause, to act in a manner calculated or likely to destroy or seriously damage the relationship of confidence and trust between employer and employee (Malik v BCCI [1998] AC 20 HL; Baldwin v Brighton & Hove CC [2007] ICR 680). Whether such a breach has occurred will always be highly context-specific; the question is ultimately whether:

If found, however, a breach of the implied term of trust and confidence will necessarily go to the root of the contract and amount to a repudiatory breach; see Morrow v Safeway Stores plc [2002] IRLR 9 EAT.

The first ground of appeal relates to the claim under section 15 EqA, the claimant contending that the ET erred in its approach as to whether the first respondent’s stipulation - that he was not permitted to attend work unless he taught a set number of hours - was justified for the purposes of section 15(1)(b). In particular, he says that the ET failed: (1) to consider what proportion of his teaching he was unable to perform in the period prior to 25 October 2018 (distinguishing between classroom or lecture teaching, and smaller teaching contexts, such as supervisions) - necessary for the ET to weigh the needs of the first respondent against the discriminatory effect; (2) to subject the case on proportionality to “critical evaluation” (per Pill LJ in Hardy and Hansons, and see Gray v University of Portsmouth EA-2019-000891) – the question for the first respondent was why it was more proportionate to stop the claimant carrying out any duties, rather than just classroom teaching. The claimant submits that, properly analysed, the first respondent’s case on necessity and proportionality was centred on avoiding increased workload for others, but: (1) there was no finding that colleagues were not prepared to continue to provide partial cover until Christmas 2018 (the only evidence of disruption was an older email relating to periods of unexpected absence); (2) the ET made no finding as to what cover arrangements were in place; (3) the ET’s conclusions were not consistent with the first respondent’s case; and (4) the ET failed to subject the reasons for not providing a support worker to critical scrutiny (large lectures would routinely have PhD students present, who could look out for blackout warning signs). It is said the ET thus failed to make necessary findings or provide adequate reasons, and, accordingly, was not in a position to properly weigh the reasonable needs of the first respondent as against the unfavourable treatment.

The second ground of appeal concerns the reasonable adjustments claim under sections 20 and 21 EqA, whereby the claimant says the ET erred in failing to consider whether there was a real prospect the proposed adjustments could remove or ameliorate the substantial disadvantage. In this regard, the claimant says if he is successful on his first ground of challenge, this would necessarily vitiate the ET’s conclusions in respect of the reasonableness of the adjustments. More specifically, he argues that the ET’s failure to make any findings as to what proportion of teaching he was still able to do precluded it from considering the reasonableness of the proposed adjustments; further, although not a reasonable adjustment of itself, the first respondent should not be permitted to hide behind a failure to carry out such an assessment.

By the third ground of appeal, the claimant contends the ET erred in its approach to his claim of constructive unfair dismissal, considering only whether the first respondent’s conduct was calculated or intended to destroy or seriously damage the relationship of mutual trust and confidence, and failing to ask whether, regardless of the first respondent’s specific intent, its conduct was likely to destroy or seriously damage that relationship. Although there was an overlap between the claims of constructive unfair dismissal and victimisation, the ET ought to have been clear as to the different tests to be applied: the latter required a specific intent but the question of breach of contract was to be answered by applying an objective test as to the consequence of the first respondent’s conduct.

For the respondents it is emphasised that the scope of the enquiry for the first and second grounds of appeal is limited to the period from the start of the academic year in 2018 to 25 October 2018 (it being accepted that the claimant was unfit for any duties from then until December 2018).

Addressing the first ground of appeal, the respondents contend that the ET made clear findings of fact about the true nature of the claimant’s role, in particular as to the need for him to teach students face-to-face in the first semester. In the light of what OH had advised, and what the claimant had himself said about the ongoing effects of his (then undiagnosed) medical condition, the ET found he was unable to carry out that teaching, which was his primary role, and permissibly accepted the evidence relating to the disruption arising from the claimant’s earlier absences, and his inability to complete the limited tasks set during the initial phased return to work. It had also accepted the first respondent’s evidence as to the feasibility of the suggested adjustments, as it was entitled to do (as to which, see the submissions on the second ground). The ET had carried out its own analysis of the working practices and operational considerations of the first respondent, set against the discriminatory effect of the decision (which was not in dispute), and had applied appropriate critical scrutiny, meeting the guidance laid down in Hardy and Hansons).

As for the second ground of appeal, the ET had permissibly accepted the first respondent’s evidence as to the practicability of the provision of a support worker and as to how this would not provide an effective means of ameliorating the disadvantage the claimant suffered as a result of his (then still undiagnosed and untreated) blackouts. It had also appropriately considered the reasonableness of the adjustments identified in the OH report of 5 July 2018 in the light of the information available to the first respondent as at the date of Mr Curtis’ email of 5 September 2018 (when the claimant’s condition had deteriorated).

Turning to the third ground of appeal, and the claimant’s constructive dismissal claim, the respondents observe that there is no suggestion that the ET erred in its self-direction as to the law. As for any reference to whether the first respondent’s conduct was “calculated”, this was because this was how the claimant had put his case before the ET (as made clear in his answers in cross-examination). In any event, the ET’s approach was consistent with the dicta in Tullett Prebon).

In considering the first and second grounds of appeal, it is apparent that there is a degree of overlap in the points made. Although the ET was asking itself different questions when considering the claims under section 15 and under sections 20 and 21 EqA, it was the claimant’s case that the first respondent’s failure to comply with its duty to make reasonable adjustments was relevant to the question whether its treatment of him amounted to a proportionate means of achieving a legitimate aim for section 15 purposes. Moreover, in pursuing this appeal, the claimant says the ET failed to make findings relevant to both claims, such that its conclusions on each were rendered unsafe. In these circumstances, it is necessary to first consider the claimant’s arguments relating to his reasonable adjustments claim, before turning to the appeal against the decision under section 15.

The claimant’s claim under sections 20 and 21 EqA identified three adjustments that it was said would have been reasonable for the first respondent to have taken to avoid the disadvantage the claimant suffered as a result of the requirement to provide classroom based teaching: (1) no large class teaching or lectures, but allowing him to undertake final year supervision and postgraduate teaching; (2) providing someone to accompany the claimant when teaching; (3) undertaking a risk assessment in relation to the blackouts and another risk assessment in relation to stress. In respect of the first proposed adjustment, the claimant says the ET failed to properly engage with the distinction between the different types of teaching that formed part of his role; focusing only on the classroom or lecture teaching, that he was unable to do, and losing sight of the smaller-scale teaching that he could do. As for the support worker suggestion, the claimant contends that the ET simply accepted the evidence of Mr Curtis, without scrutiny, and without considering the fact that that evidence was given absent any risk assessment. It is in this regard that the claimant relies on the third matter, the undertaking of a risk assessment: accepting that an assessment is not an adjustment, the claimant says that, absent such an assessment, the first respondent ought not to have been allowed to assert that the provision of a support worker was not a reasonable adjustment (it is his case that a risk assessment would have shown this could have been an effective adjustment, as the support worker – who might have been a PhD student - could have identified the warning signs preceding his blackouts).

In considering the reasonable adjustments claim, it is first necessary to recall the period to which this related. A phased return to work had been agreed with the claimant on 5 July 2018; although the OH advice on possible adjustments had not been received at that stage, there was no requirement for the claimant to carry out teaching in a classroom or lecture setting at that time and tasks could be set which allowed for a significantly reduced working day. The claimant’s complaints relate to the subsequent view expressed by Mr Curtis, by his email of 5 September 2018, that it would not be a reasonable adjustment for him to “continue to work without teaching” up to Christmas. Although the claimant accepted Mr Curtis had reasonably taken the view that he would need to be signed off work after 25 October 2018, it was the refusal between 5 September and 25 October 2018 (a period of roughly six weeks) to permit him to return to work, with the proposed adjustments in place, that was the subject of the claimant’s complaint.

As the ET permissibly found, while the OH report of 5 July 2018 could be seen to provide support for the claimant’s case on reasonable adjustments (albeit that advice was predicated on an expected improvement in the claimant’s condition), the view expressed in Mr Curtis’ email of 5 September 2018 was informed by the update he had been given by the claimant, which presented a far more pessimistic picture. Although there was a dispute on the evidence in this regard, the ET was entitled to accept (as it did) the testimony of Mr Curtis (supported by his contemporaneous emails) that: (1) on 2 August 2018, the claimant had told him that his condition had deteriorated, he did not feel he would be safe in front of a class, and had not been able to carry out the reduced workload that had been agreed; and (2) by 5 September 2018, the claimant had had more opportunity to work to the reduced hours agreed but had still been unable to complete the tasks assigned as part of the agreed phased return. In the circumstances, the ET permissibly accepted the first respondent’s evidence that there was a real question whether the claimant was fit to return to any duties and that, in any teaching context, the debilitating blackouts he was continuing to experience represented a risk to him, the students and to the university more generally.

The claimant says that the ET ought not to have accepted the first respondent’s assertion as to the risk posed by his condition, given that it had not obtained an assessment as to the real extent of that risk, and how it might have been managed by (for example) providing for a support worker (possibly in the form of a PhD student) to accompany the claimant, who might be trained to look out for warning signs of a possible blackout. The ET, however, considered this argument (put as part of the reasonable adjustments claim) but accepted the evidence of Mr Curtis that this was not a practical step at that time; the ET further accepted Mr Curtis’ assessment as to the practicalities of engaging a support worker to accompany the claimant (whether to lectures or class-based teaching, or to industry visits).

In testing the ET’s scrutiny of the first respondent’s case in this regard, I bear in mind that the assessment of reasonableness for the purposes of sections 20 and 21 EqA is objective: the focus must be on the practical result of the measure in question, not on the process of reasoning adopted by the employer (Ashton; Owen v Amec). At the relevant time, however, the position remained that the claimant was reporting suffering blackouts such that he did not feel safe in front of a class, and still had no diagnosis and (as a result) no treatment plan. Although the claimant had come into work as part of a phased return during July and August, the ET accepted that Mr Curtis was genuinely, and reasonably, concerned as to how this might have negatively impacted his recovery, in particular as he had not been able to fully carry out the reduced duties agreed. Given the position thus facing the first respondent at the start of the academic year in 2018, the ET was entitled to find that a workplace risk assessment at that time was not a practical measure, not least as the claimant was unable to properly return to the workplace and there was no medical diagnosis of his condition that could inform that assessment. More than that, it was open to the ET to find that the suggestion of a support worker being provided to accompany the claimant was not a reasonable step: even if such an individual could have been procured (a questionable proposition given the intermittent nature of the work), there was no basis for thinking they would have been able to prevent the claimant suffering a blackout, with the disruption and upset that would inevitably cause for his students (whether in a classroom or smaller teaching setting). The claimant asserts that a PhD student could have been trained to look out for the warning signs of a possible blackout, but that is simply not consistent with any objective assessment of the contemporaneous evidence; indeed, although not released to the respondents at the time, the occupational health report of 6 September 2018 recorded that the claimant was continuing to suffer regular blackouts, which remained “very unpredictable”, and that he had “only been functioning in a limited sense”.

As for the suggestion that a reasonable adjustment might have been to have permitted the claimant to return to work but with no large class teaching or lectures, instead focusing only on his research and final year supervisions and postgraduate teaching, I note that the ET did not simply rely upon Mr Curtis’ evidence in this respect (although it did accept his assessment, as set out in his email of 5 September 2018), but also had regard to the OH report of 6 September 2018. Carrying out the requisite objective assessment, the ET concluded that there was no basis for thinking that the claimant would in fact have been able to “reasonably and consistently do other things than face to face [teaching]” (ET, paragraph 314). The claimant criticises this conclusion, objecting that the ET failed to undertake a proper calculation of his classroom or lecture teaching hours as compared to the time he would spend undertaking teaching duties on a smaller scale (PhD supervisions and the like). That, however, misses the point. The ET was concerned with the question whether it would have been a reasonable step for the respondent to have permitted the claimant to return to work but with the removal of some (as it found, the majority) of his duties. Given what the ET accepted were the first respondent’s reasonable concerns as to the negative impact on the claimant’s recovery of the earlier attempt at a phased return, taken together with the unpredictable blackouts he was continuing to suffer on a regular basis (see the OH report of 6 September 2018), his statement that he would not feel safe in front of a class (as the claimant had explained to Mr Curtis on 2 August 2018), and the fact that he was only functioning in a limited sense (as Mr Curtis recorded on 2 August and 5 September 2018, and as confirmed by the OH report of 6 September 2018), the ET plainly reached a conclusion that was open to it on the evidence.

The claimant’s case was (and is) predicated on the argument that his condition could only reasonably have been seen to have impacted on his ability to carry out classroom or lecture style teaching; that, however, was (and is) not supported by the evidence. As the respondents observed in argument, the possibility of the claimant suffering a blackout whilst at work would not only pose an obvious danger to him, but would also risk disruption and upset to others present, whether in a lecture or PhD supervision setting or in any other work-related context. Furthermore, the claimant’s argument simply fails to engage with what the ET permissibly found to be the first respondent’s reasonable concern about the potential negative impact on his recovery if the claimant were to return to any duties at a time when he was still experiencing blackouts caused by a still undiagnosed condition.

Turning then to the appeal against the decision on the section 15 claim, in arguing that the ET failed to carry out the required proportionality assessment, the claimant again emphasises the distinction between class and lecture based teaching and the more individual-based duties that he contends he might still have undertaken. It is his case that the ET’s failure to have regard to this context meant that it did not carry out the required critical evaluation of the first respondent’s case on justification.

In assessing whether the first respondent had demonstrated that the requirement to carry out face-to-face, class-based teaching was justified, the ET accepted both that teaching was the claimant’s principal role (at least for the semester in question) and that the provision of consistent high-quality teaching was a legitimate aim. Going into the new academic year, therefore, the first respondent needed to be confident that it could meet that aim, in accordance with the prospectus and expectations of the students. As the ET recorded, the claimant’s absences in the previous academic year had demonstrated the very real difficulties that could arise if his ability to carry out his teaching load was not assured; it permissibly accepted that this was a legitimate and genuine concern for the first respondent at the start of the 2018/2019 academic year.

In its findings as to the significance of class and lecture style teaching for the claimant’s role, the ET further accepted Mr Curtis’ evidence that the School was primarily teaching-based; the claimant may have undertaken other, research-related duties (including PhD supervisions etc) but the ET was entitled to find that the kind of teaching that was being required of the claimant was a key part of his role. That finding was not dependent upon a precise calculation of the hours spent by the claimant in a classroom or lecture hall (as compared to other teaching duties): in terms of the first respondent’s legitimate aim, it was sufficient that the larger scale teaching in issue was a significant (if not the only) part of the claimant’s duties at the time. Moreover, and in any event, the real issue raised by the claimant’s case related not to the precise assessment of the hours he spent engaged in the various different forms of teaching but to his complaint that the first respondent was effectively imposing an all or nothing condition to his return to work, with the consequence that the claimant’s inability to undertake classroom or lecture based teaching meant he was prevented from carrying out other aspects of his role, such as research supervisions or mentoring.

Thus having regard to the impact of the measure on the claimant, the assessment of proportionality in this context was for the ET; it was required to undertake (and to demonstrate that it had done so) a critical evaluation of the measure adopted by the first respondent to achieve the legitimate aims it had identified (Hardys & Hanson). The role of the appellate tribunal is to scrutinise the decision thus reached by the ET; in so doing, however, I am bound to have regard to the entirety of the reasoning provided, which, in this instance, includes that provided in relation to the reasonable adjustments claim.

For the reasons I have already set out, in its evaluation of the decision that the claimant should not be permitted to return to work when he was unable to carry out the principal functions of his role, the ET permissibly rejected the various steps suggested by the claimant as reasonable adjustments that would have provided a potentially less discriminatory alternative. These expressly included the claimant’s contention that he should have been permitted to undertake the more limited teaching requirements associated with the research aspects of his role. Considering the condition imposed upon the claimant’s return to work – that he should be able to carry out face-to-face, class-based teaching – the ET permissibly took into account the fact that the claimant had only recently completed a six-month sabbatical to develop his own research, and that there was insufficient need for him to undertaken further research or mentoring duties. It further accepted the first respondent’s concerns regarding the claimant’s health, his own reported concerns about the unpredictable nature of his blackouts (the cause of which remained undiagnosed), and his apparent inability even to undertake the more limited duties (not requiring face-to-face class-based teaching) agreed as part of the earlier phased return. On the evidence, it was thus open to the ET to accept the first respondent’s case that it was unable to offer the claimant alternative duties that would warrant permitting him to return to work, with all the risks that that entailed, whilst not undertaking the class-based teaching required of his post.

Turning then to the final ground of appeal, which relates to the decision on the constructive dismissal claim, it is the claimant’s contention that the ET only considered the question whether there had been a breach of the implied term through the prism of the first respondent’s intent, failing to ask whether – whatever it might have subjectively intended – its conduct had (objectively) been likely to destroy or seriously damage the relationship of mutual trust and confidence. This argument is based on the ET’s explanation of its reasoning, at paragraph 332 d. of its decision, where it is stated that it did not find that the first respondent’s conduct “suggests an intention to abandon and altogether refuse to perform the contract”.

In considering this ground of challenge, I note that, in its self-direction on the law, the ET expressly reminded itself of the relevant guidance provided by the case-law, most notably as set out in Malik v BCCI, Baldwin v Brighton, and Tullett Prebon. In the circumstances, I would not readily infer that it then failed to apply those principles when considering its decision on the facts it had found in this case (see paragraphs 57 and 58 DPP Law v Greenberg [2021] EWCA Civ 672). Furthermore, when considering the ET’s explanation of its reasoning, it is right that I do so bearing in mind the way the case was put: an ET does not err in law in determining the case as presented to it, rather than some alternative way in which it might have been put.

In this regard, I note that it was the claimant’s case that the final straw, giving rise to the fundamental breach of the implied term of trust and confidence was the delivery to him on 3 October 2019 of documents setting out the management side response to his grievance appeal. In cross-examination, the claimant had made clear that he was putting his case on the basis that the first respondent’s actions in this respect were “calculated and deliberate”. That was a necessary clarification of the claimant’s case, given that it would otherwise be hard to see how the mere presentation of the management side to a grievance appeal could, without more, be likely to destroy or seriously damage the relationship of mutual trust and confidence: as the whole point of the appeal process would be to allow for both “sides” to be presented, it is unlikely that any objective assessment of the content of the documents presented to the claimant on 3 October 2019 could lead to the conclusion that there had thereby been a breach of the implied term.

Thus having regard to the case it was required to determine, it is apparent that the ET did not err in considering whether the first respondent’s conduct had suggested “an intention to abandon and altogether refuse to perform the contract”. Moreover, in approaching its task, it is clear that the ET applied the requisite objective test, considering the question it had to answer “from the perspective of a reasonable person”. The conclusion it reached was plainly one that was open to the ET on the facts of this case. No error of law is disclosed in the ET’s reasoning and there is no proper basis for the third ground of appeal.

For all the reasons thus provided, this appeal is duly dismissed.