D’Silva v Manchester Metropolitan University (Race Discrimination) [2013] UKEAT 0437_11_0412

EAT
D’Silva v Manchester Metropolitan University (Race Discrimination)
[2013] UKEAT 0437_11_0412 · 2013-12-04
[1]We are here concerned with two joined appeals. In giving our Judgment, we refer to the parties as the Claimant and the Respondent, as below.[2]The first in time in that in 0437/11, which is an appeal against the Judgment of the Employment Tribunal, sitting at Manchester under the chairmanship of Employment Judge Sneath (“the Sneath ET”), which sat on 5-9, 12, 14-16 July 2010 and 2, 6-8 and 20 September 2010 and deliberated on 8, 10, 11 and 12 November in chambers. We think that makes it 14 days of hearing plus four days of deliberations.[3]The second appeal is in 0533/11, which is an appeal against a Judgment of an Employment Tribunal under the chairmanship of Employment Judge Robertson (“the Robertson ET”), sitting in Manchester on 27 April 2011 and again on 11 May 2011 for its deliberations and sent with Reasons to the parties on 17 April 2011.[4]In both Tribunals, the Claimant was represented by his lay representative, Dr Deman. At the Sneath ET, the Respondent was represented for the first week by Mr Gilroy QC and thereafter by Ms Connolly, who also represented the Respondent in the Robertson Employment Tribunal and has represented the Respondent again before this EAT. The Claimant, on the other hand, was represented before us by Mr de Mello of counsel, who had not appeared below.[5]Before we turn to the substantive appeals, we record that an email was received at the EAT earlier this week from the Claimant, raising his concern about the possible issue of a perception of bias as to the constitution of this Court. This apparently related to the fact that, when in private practice, I had previously been in chambers to which Mr Gilroy also had a connection. It was unclear as to what, if any, basis existed for concern in these circumstances but Mr de Mello was invited to make any application he wished to make on behalf of the Claimant at the beginning of this hearing. Mr de Mello stated that no application was being made for any member of the Court to be recused. That being so, no further attention was given to that matter.[6]For good measure, it was also made clear that one of our number, Ms Branney, sat as a member of the previous division of this Court (the Hon Mrs Justice Slade presiding), which had considered this matter and agreed that certain points should go through to a full hearing. That matter was drawn to the attention of the parties, prior to this hearing. No point was taken. The factual and procedural background[7]We then turn to the factual and procedural background relevant to these appeals. At all relevant times, the Claimant was a Senior Lecturer at the Respondent university. He had originally graduated from Lancaster University and gained his PhD at Birmingham. He had a successful early career in research. He became a Fellow of the Royal Society of Chemistry and joined the Respondent in 1993 as a Senior Lecturer in its Department of Chemistry.[8]At the time of the Sneath Employment Tribunal, the Claimant had made five prior claims of race discrimination (we use that term to cover various forms of such discrimination) against the Respondent in the Employment Tribunal. He had made the first such claim at the end of August 2002. That was compromised on 22 October 2003, when the Claimant agreed to withdraw his claim and the parties agreed that there should be no recriminations by either against the other by reason of the proceedings and no application for costs.[9]On 8 December 2003 the Claimant presented a second claim of race discrimination, harassment and victimisation against the Respondent. He complained about the Respondent’s decision that he had not met the criteria for promotion to Professorship in the 2002/2003 promotions round. He submitted a further, third claim on 31 March 2004 and his fourth claim on 17 May 2004, the last claim relating to the Respondent’s decision not to shortlist him for the post of Principal Lecturer. His fifth claim was presented on 23 August 2004, which included a complaint about the Respondent’s decision not to appoint him to either the post of Reader or Professor in the 2003/2004 promotions round. We will refer back to that promotions round later on in this Judgment.[10]The Claimant’s second to fifth claims - referred to as the consolidated proceedings - were heard originally in the absence of the Claimant by an Employment Tribunal chaired by Ms O’Hara (“the O’Hara ET”) between 18 April and 14 May 2005. The initial decision of the O’Hara ET was to dismiss a number of the Claimant’s claims in the consolidated proceedings, but to find there was race discrimination in respect of six matters. Those findings were then appealed by the Respondent to the EAT, which overturned five of the findings of discrimination and remitted the sixth such finding back to the O’Hara ET for re-hearing. That re-hearing took place over five days in December 2008 and March 2009 and resulted in the Tribunal dismissing the outstanding matter of complaint.[11]Meanwhile, the Claimant had presented his sixth and seventh Employment Tribunal claims against the Respondent on, respectively, 15 September 2005 and 14 September 2006. These included complaints of race discrimination and/or victimisation in respect of the Respondent’s failure to appoint the Claimant as a Reader or a Professor in the promotion exercises of 2004/2005, the subject of the sixth claim, and 2005/2006, the subject of the seventh claim. There were other matters also raised in these claims and we are conscious that the issues we are now considering were not the only matters before the Employment Tribunal.[12]It was thus the sixth and seventh claims that were heard by the Sneath ET in 2010. Those claims were dismissed in their entirety by a Judgment sent to the parties on 30 December 2010.[13]The Claimant appealed against the Sneath ET Judgment and was ultimately permitted to proceed to a full hearing on limited grounds after a preliminary hearing before the EAT (Slade J presiding). That was on 27 March 2012.[14]It was pursuant to that limited permission that the first of the two appeals we are charged with hearing (that in 0437/11) comes before us today. We will return to the grounds on which the appeal was permitted to proceed in due course.[15]The Claimant applied for permission to appeal to the Court of Appeal against that Judgment of the EAT but that was refused.[16]In the meantime, we understand the position to be that, on 4 October 2010, the Claimant presented his eighth Employment Tribunal claim against the Respondent. That claim was struck out by the Robertson ET (the Judgment being sent to the parties on 31 May 2011) as having no reasonable prospect of success.[17]The Claimant appealed against the dismissal of his eighth Tribunal claim and, on the basis that it raised substantially the same issues that had already been permitted to proceed to a full hearing in EAT/0437/11, that was also permitted to proceed to a full hearing, joined with the first appeal in 0533/11.[18]It was common ground before us that the second appeal was dependent upon the success of the first. The Respondent reserved its position as to what further submissions it might make if the appeal was successful. If the first appeal failed, however, it was agreed that the second appeal must also fail. We indicated that we would allow the parties to make further representations on that issue after we had given our Judgment, but that was our understanding of how the appeals were being put. Certainly, Mr de Mello, appearing for the Claimant before us, made his position clear in this regard; relying only on his arguments in the first appeal.[19]To update the history further, we understand that the Claimant has since been dismissed by the Respondent, which has led to two further claims, the ninth and tenth Tribunal claims brought by the Claimant against the Respondent. We understand that these are currently stayed pending the outcome of the present appeals.[20]Returning then to the issues before us in the current appeals, these were identified by the Slade EAT on 27 March 2012. The first such issue was whether the Sneath ET erred in its approach to determining the claims in respect of failing to appoint the Claimant to Readership or Professorship in 2004/05 and 2005/06. Specifically: “whether the Sneath ET erred in its approach to whether there were facts from which an inference of race discrimination could be drawn in respect of the non-appointment of the Claimant to a Professorship or Readership in 2004/05 or 2005/06 promotion rounds” in that the Claimant had progressed to the second stage of (the Readership) promotion selection round in 2005/06 and had done so in previous years but not beyond the first in 2004/05. There is no explanation why the Appellant was thought to not satisfy the necessary criteria to progress to the first stage in 04/05 when he had done so in previous years and did in 05/06.[21]We pause there to make plain that, although the order did not specify it, it is apparent - and common ground between the parties - that the Claimant only progressed to the Readership second round in 05/06. It was not suggested that there was any evidence before the Tribunal of the Claimant having succeeded in getting through to the second stage prior to the ‘03/’04 promotions round.[22]The second issue identified was whether the Tribunal had failed to pay any, or any proper, regard to:(1) the fact that, in the ‘05/’06 round the referees selected by the Respondent at the second stage for the Claimant’s application were not, or not predominantly, from the Claimant’s area of expertise, namely Materials Chemistry;(2) that the referees were chosen from higher grade institutions than the Respondent;(3) that seven referees were contacted in respect of the Claimant, rather than the three for the successful candidate; and(4) that the referees contacted by the Respondent for Dr Rybakov (that is the Claimant’s comparator) were nominated by Dr Rybakov himself.[23]It is to be noted that, although the claim before the Sneath ET had included complaints of victimisation and much of the evidence and argument were obviously directed towards that part of the claim, the appeal before us was only permitted to proceed in relation to the grounds relating to race discrimination. It is, however, apparent that the issue of victimisation had raised broader points before the Tribunal, not least because it found that the Claimant had raised previous complaints of discrimination in bad faith (see paragraph 249). The Employment Tribunal’s decision[24]Given the way the appeal has been put before us, it is necessary to descend into the Tribunal findings of fact in some detail.[25]As the Sneath ET found (paragraph 56), the process of application and appointment to the post of Professor or Reader within the Respondent involves two stages. First, a judgment by the Professoriate Committee, based upon the contents of application forms together with views expressed by referees nominated by the applicants. If the Professoriate Committee judges that an application has merit at stage 1, it then proceeds to stage 2. At that second stage, it is judged again on the application form, together with the view expressed by all the referees, including those identified by the Respondent.[26]The Claimant’s applications for promotion to Professor or Reader did not progress beyond stage 1 in the 2004/2005 round.[27]The Respondent’s criteria for promotion are set out in the documentation contained within our bundle, but expressly found to be as set out by the Employment Tribunal as follows (referring to paragraph 57 in respect of Professors, and paragraph 58 in respect of Readers):
“57. The Respondent’s criteria for the appointment of professors were:- ‘Candidates will be expected to demonstrate a high level of academic distinction, as evidenced in – (i) Research, consultancy, scholarship or creative achievement which has contributed to the furtherance of knowledge and its application (ii) Professional standing; (iii) Academic leadership, including contributions to subject and course development and the development of student learning (iv) Candidates must demonstrate outstanding achievement in relation to at least one of the clients, normally (1) and a high level of achievement in the others.” 58. The respondent’s criteria for the appointment of readers were:- ‘Candidates for reader posts must demonstrate a high level of achievement in relation to the first and at least two more of the following criteria: (1) A capacity for, and achievement in, personal research or scholarship as evidenced by publications in research journals of repute or by other evidence including external appraisal; (2) A proven ability to lead and develop research within the department and the university including an involvement in supporting and assisting the work of colleagues undertaking research; (3) An ability to supervise research students, including successfully completed supervisions to PhD level; (4) A demonstrated capacity to attract external resources in support of research at the university and in the promotion of external academic collaboration. Additionally candidates should be able to demonstrate their involvement in ensuring that teaching programmes benefit from their research activity.”
[28]In considering the claim in respect of the 2004/05 promotion round, the Sneath ET looked first at the Claimant’s application for Professorship. It addressed the identification of the comparator and then made findings in respect of that comparator (Dr Rybakov). The Tribunal then went on to make findings in respect of the referees nominated by the Claimant, the responses from Professor Leach (Head of the Department of Chemistry and Materials in which the Claimant worked), and to the reference from Professor Neal (the Dean of the Respondent’s Faculty of Science and Engineering).[29]The Professoriate Committee met on 24 February 2005. As the Tribunal found (paragraph 75):
“The Professoriate Committee met on 24 February 2005 to consider the claimant’s and seven other applications for professorship. None of the eight applications was successful. The committee’s minute in respect of the claimant said: ‘Criteria not met. It was noted that Dr D’Silva’s application contained factual inaccuracies in its analysis of others’ promotions and that it lacked a critical and objective evaluation of his own research record against the scheme’s criteria. His application showed an inadequate current level of research output and lacked objective evidence of academic leadership. Dr D’Silva commented in his application that the faculty renovation building programme had disadvantaged his research but it was noted that the overall level of research submissions in the faculty has in fact risen during the works programme.”
[30]The approach adopted is then set out by the Tribunal (paragraph 78):
“The assessment of any candidate’s application owes much to the clarity with which the application establishes the candidate’s case against the stated criteria. The statements from the Head of Department and Dean of Faculty are valuable in providing a frame of reference for the applicant’s achievements. Guidance to candidates encourages them to select their own referees in such a way that those nominated will be able to offer an authoritative and objective assessment of the candidate’s work. As Professor Plumb indicated, the committee’s role was to make an overall academic judgment on each application. There were no fixed quotas of jobs available. Each application was judged on its own merits against the published criteria.”
[31]The Sneath ET found that the discussion of the Claimant’s application was limited to its merits (see paragraph 83); no account was taken of the Claimant’s previous complaints of discrimination. The Tribunal’s findings as to the committee’s view of the Claimant’s application are set out at paragraphs 84 to 88:
“84. In that regard the committee had to deal with the fact that the claimant in his application for professorship had addressed the criteria for readership. What he did was to address those criteria first and then return to the second criterion for the professorship appointment. Professor Plumb shared the view of one of the claimant’s referees that the supporting statement was poorly composed and had, as he put it, the feel of an unstructured engagement with the process. In passing, we observed the same approach in the claimant’s witness statement and the sometimes breezy way in which he read it to us. We endorse counsel’s submission that the witness statement in evidence was jumbled and disjointed. As originally drawn, he tried to re-litigate many of the matters that had been the subject of the previous unsuccessful claims, thus supporting counsel’s submission that the claimant was unable to accept any decision, whether employment or judicial, that did not go in his favour. 85. Despite shortcomings in presentation, the committee did its best to assess the overall worthiness of the content against each of the appropriate criteria. Nevertheless, they noted the final paragraph of the claimant’s submission:- ‘Dr D’Silva is outstanding in relation to two of the three criterion for professorship outlined in MMU’s document for the appointment of professors and is outstanding in research in relation to all appointed professors in the faculty of science and engineering for the last ten years’. As Professor Plumb put it, that opinion was inevitably at variance with that of the committee which had assessed all of its applicants using the scheme criteria down the years rather than those that the claimant had chosen to employ such as impact factors. 86. As for the references, the committee considered that they provided a mixed picture. Professor Neal’s statement did not meet the criteria. Professor Leach challenged the factual accuracy of some of the claimant’s assertions particularly about the availability of laboratory facilities during the refurbishment. The committee also noted that the claimant had himself supplied information to one of his referees, the authenticity and factual accuracy was unknown to the committee members. 87. Generally, the committee concluded that, whilst some of the research related activities could be confirmed as at a high level, it could not regard the claimant’s profile as outstanding in that area. There was ample evidence of a capable researcher who over a period of time had produced a number of good quality research outputs. But that had dwindled in recent years so that there was only one eligible publication related to the year 2004 and one to the previous year. 88. In respect of the second criterion, the committee’s view was that the claimant’s achievement approached a high grade rather than an outstanding one. As for the third criterion, it appeared to the committee to be no more than the respondent might expect of any of its academic staff. Specifically the committee decided that the claimant had in his submission demonstrated an inadequate current level of research output and insufficient objective evidence of academic leadership.”
[32]The Tribunal then turned to the Readership application for 2004/2005 (paragraphs 89-97). It considered first the comparator, Dr John Binnie, and then the Claimant’s application (see paragraphs 92-97):
“92. Of the Claimant’s application for readership, the committee minute noted:- ‘(Also applied for professorship) It was noted that Dr D’Silva’s 2003/2004 application had had been submitted to externals in order to demonstrate the committee’s objectivity in dealing with it and that it had not been supported. The current application failed to demonstrate that Dr D’Silva met the required criteria in terms of his current publications record, academic leadership and recent research funding. The application did not show any progress over previous applications.’ The minute there refers to an exercise initiated by Professor Plumb arising out of the claimant’s challenge to the committee’s objectivity when considering his 2003/2004 application for readership. Professor Plumb approached three external independent academics who confirmed the committee’s view that the claimant’s overall performance had not met the criteria for appointment. This external assessment gave the committee confidence to consider objectively the claimant’s 2004/2005 application and enabled it to use the previous application as a starting point against which to measure the claimant’s progress. 93. In doing so, the committee members noted a publication record whose rate had slowed in the latter five-year period. Equally the volume of conference publications was lower in that period. Whilst the committee noted the Dean’s opinion that the claimant had met the first criterion, it formed the view that the claimant’s achievements against it could not be regarded as high. 94. In relation to the second criterion, the committee formed the view that there was little evidence of the claimant’s role in leading and developing research either within the department or the faculty or in supporting colleagues undertaking research. Again the committee did not think that the claimant had met the high level. 95. In respect of the third criterion, the committee felt unable to give it a high grade because, on the claimant’s own admission, he was not currently supervising research students. As for the fourth criterion, whilst the claimant had in the past attracted good funding, his record had tailed off in recent years. The last significant award had been in 2000. 96. Finally, in addition to the four criteria, the candidate was required to demonstrate...this involvement in ensuring that teaching programmes benefited from research activity. In the committee’s view, his application presented evidence of no more than that which the university might expect of any of its academic staff. Accordingly the committee decided that that criterion had not been met either. 97. The committee also looked at the references of the nominated referees. They have already been described above in connection with the professorship application. Notwithstanding the largely positive responses of the claimant’s nominated external referees, the committee considered that the claimant had failed to show sufficient progress in his research so that the committee members could have confidence that his research achievements would be sustained.”
[33]As to the 2005/2006 promotions round, the Tribunal dealt with this at paragraphs 160-196. In respect of his application for Professorship, the Claimant again did not get past the first round. The Tribunal dealt with the references considered by the committee at paragraphs 165-166. The Professoriate Committee met on 19 January 2006 and considered the Claimant’s application, as the Tribunal records (paragraph 167):
“As we have indicated, the claimant did not get past stage 1 and the minutes of the Professoriate Committee meeting on 19 January 2006 recorded:- ‘Current, as distinct from past, performance does not meet criteria in that an outstanding level of achievement was not currently evident in respect of any of them.’”
[34]The Tribunal expressly accepted the committee’s analysis of the Claimant’s Professorship application (paragraph 173):
“Accordingly, we were able to accept the committee’s analysis of the professorship application described by Professor Plumb that it provided no outstanding evidence in respect of any of the criteria. Specifically, the committee noted that the claimant’s research output had continued at a lower rate than in previous years, there being only one additional single journal publication to those cited in his application form the previous year. The claimant had cited in his current application form several publications that were not eligible for inclusion because they were in preparation or subject to revision. The last significant grant attracted by the claimant had been made in 2000. Finally, the claimant’s statement failed to demonstrate any contribution to teaching or academic leadership that exceeded what the university would expect of any of its academic staff.”
It then referred back to its earlier findings regarding Dr Rybakov as a comparator.[35]As for the Claimant’s 2005/2006 Readership application, the Tribunal’s findings of fact are at paragraphs 175-177. In relation to the referees in respect of that application, the Tribunal found (paragraphs 177-180):
“Finally, the referee from Newcastle University felt able strongly to recommend the promotion of Dr Edensor to reader, although he expressed reservations about his record in attracting research income. 178. In due course Professor Plumb secured four references for the claimant’s readership application. Professor Williams of the University of Liverpool said that he paid particular attention to evidence of international recognition, especially through publications and invited conference papers and could not find the evidence compelling. He considered that a reader should still have an upward profile in terms of those outputs and did not see that from the claimant’s CV. Following further observations he concluded that he could not support the claimant’s proposal for promotion to a readership. 179. Dr Ahmed of the University of Kingston on the other hand was fully supportive whereas the referee from the University of Manchester thought that, although the claimant had built up some research momentum in the 1990s, it had since subsided and the claimant needed to retrench and rebuild his research group. 180. Finally, Professor Sterling of Swansea University concluded by saying that the claimant’s case for promotion merited consideration and in light of his persistence and high teaching load he should be commended for his effort. Professor Sterling, however, had difficulty in being convinced of strong research leadership and international level of scholarship that would make the case for readership more compelling.”
[36]The Claimant’s arguments before the Employment Tribunal in respect of the referees an and the Respondent’s responses are set out by the Tribunal at paragraphs 181-189:
“181. Faced with that evidence, the claimant asserted in his witness statement that Dr Binnie’s three external referees were not impartial and had been chosen via social contacts; what he described as the grapevine effect (sic). The claimant went on to say that in his Race Relations Act questionnaire he asked as to the procedures of how external referees were picked. He said that the answers he obtained were evasive in that the processes used were the same in all cases. He went on to argue that they were not because in the cases of the successful candidates, the committee was happy to accept less critical references that did not address appointment criteria. Furthermore, insofar as Professor Plumb or Professor Allen was involved in the process, their choices were tainted because of their earlier involvement in the claimant’s applications. 182. In respect of his own referees, the claimant said of Professor Williams that he was from a 5* research department, implying that higher standards were being applied than were appropriate. The claimant also objected that although Professor Williams claimed to do similar work to that of the claimant, his CV showed that he was a metallurgist and engineer, which somehow showed that collusion was involved in the promotion process, meaning that Professor Plumb set out to victimise him. He contrasted his UK referees with those international ones for other candidates and asserted that promotions in academia were based on grapevine recommendations and social networks. He also asserted that there were academics keen to support victimisation of anyone who raised the issue of race discrimination in promotions. 183. In his 2006 questionnaire the claimant asked what the reviewers’ credentials were that made him or them suitable. The respondent answered that stage 2 referees were selected having regard to their discipline, subject expertise and their own and their department’s standing in research. That was the same answer to the claimant’s question about the criteria used to select the reviewers. He went on to ask whether the selection was made randomly or from a pool or were they handpicked. The respondent answered that they were selected on the basis of the criteria set out above. The claimant asked whether, if the referees were handpicked, who made the decision and the respondent replied that the deputy vice chancellor, Professor Plumb, made the decision, advised usually by deans, heads of department and the director of research development. 184. Professor Plumb addressed this in evidence. His practice was to approach senior research staffs (usually professors and/or heads of departments) in the general subject discipline of the candidate’s work, who were themselves working in acknowledged leading areas of research. He used the Higher Education Funding Council for England’s National Research Assessment Exercise (RAE) rating for each UK university department to identify the leading centres in a particular subject area. The ratings ranged from 5* to 2 in descending order that included 3(a) and 3(b). 185. Secondly he identified the members of the National RAE panel who had generated the rankings but this was a limited pool, since he did not wish to overburden them. Thirdly he identified those university departments from which the assessment panel members came because other senior staff in those departments represented a source of potential referees. He also sought advice about potential referees from the dean of the candidate’s faculty. Finally he sought the views of the university’s director of research development. 186. The sciences inevitably achieve the highest rankings whereas less well developed (in research terms) subject areas have a much lower number of higher ranked research departments. This suggests that consistent RAE ratings are not a useful tool for the Professoriate Committee to use when considering individual candidates from a range of disciplines. 187. Professor Plumb also preferred to approach new referees at each round (a) to avoid what he described as a fatigue factor and (b) to avoid approaching those who had perhaps earlier given a critical or negative reference to the same candidate. 188. The claimant suggested that Professor Plumb had deliberately approached referees in 5* rated research departments, only in his case. In fact none of those requested were from 5* universities. There were from 5 rated, one from a 4 rated and one from a 3(a) rated department. 189. The claimant also made an issue of the number of referees (seven) who were approached in his case. To counter this, Professor Plumb gave the statistics for the rounds from 2001/2002 to 2005/2006 that showed he had previously approached up to seven referees, although seven appears to have been the highest number in the 2005/2006 round.”
[37]The Tribunal then summarizes the position regarding the Claimant’s references and sets out the Respondent’s minute of the decision taken in respect of his application (paragraph 192):
“192 The record showed:- ‘There were four references of which one supported the promotion. The other three references drew attention to Dr D’Silva’s loss of research momentum in recent years. This was typified by one reference which compared Dr D’Silva’s productivity over two phases, 1987-1996 and 1997-2006. This showed a marked decline in journal and conference publications, invited lectures and income generation. Dr D’Silva had cited his involvement in litigation against the university as a principal cause of this decline. Irrespective of the merits and outcome of the litigation, the committee accepted as a matter of fact that this might well have dissipated the amount of time and concentration that Dr D’Silva was able to devote to research in the relevant period. Members were aware that he was absent on health grounds for some of this time. They were less sympathetic to his proposition that the large scale refurbishment of the John Dalton laboratories was an additional disruptive factor since the research output of the faculty of science and engineering had continued to rise during this period, despite the less than ideal conditions. Dr D’Silva’s past records showed evidence of solid progress. Viewing his research output as a whole, it was at least arguable that, at an earlier stage of his career, he had shown the potential for eventual progression to a readership. Unfortunately that progress had not been sustained and, on his recent record, there was little evidence of recovery and hence a limited basis for promotion. Professor Taylor advised that on the evidence before the committee this ought to be an open and shut rejection (Professor Taylor was the external member). Notwithstanding this, the committee had some sympathy for Dr D’Silva in view of his partially realised potential. Unlike the RAE, the scheme criteria did not make allowance for loss of momentum on health grounds and the committee had not made a practice of allowing for what might broadly be termed ‘mitigating factors’. Over the years, the committee had consistently applied the concept of trajectory to inform its decision-making. On this measure Dr D’Silva did not currently make the grade. Although well established both on its own practice and across the university sector, members were [conscious] that trajectory (meaning the continuing or accelerating pace of research output) was not explicitly defined in the scheme criteria, although the salients of output over the last five years was included in the guidance to applicants. Whilst it was an important determinant of excellence and future promise for young researchers, it might not be a sensitive enough indicator of excellence when applied to achievement over a researcher’s career as a whole. The question was whether it unfairly disadvantaged the researcher who produced the threshold quantity and quality of outputs, but over many years, and, perhaps, sporadically. This was a complex issue requiring detailed consideration since it exposed the balance between academic judgment and the Equality Opportunity issues such as age, career breaks and interruption through ill-health. On balance giving full consideration to the wide range of contributory issues and noting that the profile of Dr D’Silva’s research work had changed little over the last several years, the committee was not persuaded that his application met its criteria for promotion. Whilst reaching that conclusion in relation to readership promotion, the committee was sensitive to the wider aspects of employment circumstances in which the candidate operated. Accordingly it requested the chairman to make a more detailed assessment of Dr D’Silva’s career and future potential for promotion outwith the ambit of the committee having consulted with his Dean and head of department.’”
[38]The Tribunal then summarised its finding in respect of the decision taken (paragraph 193):
“Specifically, in relation to each of the four criteria, Professor Plumb reflected the committee’s view that the claimant had not demonstrated achievements to merit a ‘high’ grading in respect of any of them. In respect of the first, research output, three of the claimant’s referees had drawn attention to his failure to maintain momentum; there had been a recent decline in research output and his achievements were seen as on a downward profile. In respect of the second criterion, the committee found little evidence of the claimant’s role in leading and developing research within the department or faculty or in his support for colleagues undertaking research. In respect of the third, the committee noted that the claimant had not yet again undertaken the role as a supervisor of PhD research students, although he had in the previous year taken on the supervision of a Master’s research student. Finally his record in attracting grants to support his research work, whilst good in the earlier years, had tailed off significantly in recent years.”
[39]Having considered the detail, the ET then stood back, giving a more general overview of its findings of fact in respect of both promotion rounds and the Claimant’s applications (paragraphs 194-196):
“194. Generally, in relation to the process of promotions, it was what Professor Plumb described as meritocratic. As we have seen, there were no quotas for promotion to either post. The process was not competitive. In the event no-one was successful in the 2004/2005 professorship round and only one of the seven applicants got to stage 2. In that round the university considered nineteen applications for reader, eleven of which were rejected at stage 1 and ultimately only five were successful including one applicant from the claimant’s faculty, Dr Jon Binnie. 195. In the following round, eleven applied for professorships. Four were rejected at stage 1 and the remaining seven progressed successfully through stage 2 to appointment including the remaining applicant from the claimant’s faculty, Dr Vladimir Rybakov. 196. For readership that year there were eighteen applications of which eleven were rejected at stage 1. Of the five that progressed to stage 2, two were successful including one from the claimant’s faculty, Dr Tim Edensor. The claimant was one of two unsuccessful candidates from that faculty.”
The law[40]We have had regard to the relevant legislative provisions contained in the Race Relations Act 1976 .

(a) :

“(1) A person discriminates against another in any circumstances relevant for the purposes of any provision of this Act if -- (a) on racial grounds he treats that other less favourably than he treats or would treat other persons;”
[41]We have also had regard to section 4(2)(b) and (c), which makes it clear that: “It is unlawful for a person, in the case of a person employed by him at an establishment in Great Britain, to discriminate against that employee- ... (b) in the way he affords him access to opportunities for promotion, transfer or training, or to any other benefits, facilities or services, or by refusing or deliberately omitting to afford him access to them; or (c) by dismissing him, or subjecting him to any other detriment.[42]We have also noted section 3(4) and had specific regard to section 54A, dealing with the burden of proof under the Race Relations Act 1976 , as follows:
“(1) This section applies where a complaint is presented under section 54 and the complaint is that the respondent – (a) has committed an act of discrimination, on grounds of race or ethnic or national origins, which is unlawful by virtue of any provision referred to ... (2) Where, on the hearing of the complaint, the complainant proves facts from which the tribunal could, apart from this section, conclude in the absence of an adequate explanation that the respondent – (a) has committed such an act of discrimination or harassment against the complainant... the tribunal shall uphold the complaint unless the respondent proves that he did not commit or, as the case may be, is not to be treated as having committed, that act.”
[43]The Sneath ET recorded the case-law to which was referred on the question of the burden of proof and summarized its understanding of the approach it should adopted, at paragraph 201 of its Judgment. No criticism of that self-direction is made by either party.[44]The Tribunal summarised its analysis and approach at paragraph 206 of its Judgment in general terms:
“Before addressing the specific issues, we understand that we should guard against dealing with them in isolation and risk losing ‘the big picture’. At the end of our specific findings, we step back from each individual allegation and look at the totality of the case. In doing so, we conclude that, in the various employment decisions and actions or omissions under review, he was treated less favourably either on racial grounds or because he had done protected acts. We have focussed there on the ‘reason why’ question but, given the need for the same or not materially different circumstances between the claimant’s case and that of his comparators, actual or hypothetical, it is difficult to see that he has been less favourably treated in respect of any of those decisions and actions or omissions. In respect of the promotion complaints, for example, the claimant has failed to prove that his applications were as good as those of his comparators. Having made that general statement, however, it is necessary to go back into the detail of the allegations in order to explain it.”
[45]Mr de Mello for the Claimant criticizes that passage, in particular the reference to the statement “the claimant has failed to prove that his applications were as good as those of his comparators”, the criticism being made that it would not be for the Claimant to prove that; he would not have to go so high, that would be a misunderstanding of the burden of proof.[46]If that was where the Sneath ET had stopped, we would have had some sympathy with that submission. It was not. The ET then descended into the detail of the matters before it and we have to have regard to that detailed assessment in our consideration of this Judgment.[47]As for the case-law, the skeleton arguments before us referred to a number of cases, particularly on the question of the burden of proof, albeit that we were only referred to the case of Bahl in oral submissions. We make clear, however, that we in any event reminded ourselves of the approach that has been laid down, particularly in the Court of Appeal, to the question of the application of the burden of proof, see Igen Ltd v Wong [2005] IRLR 258 CA and Madarassy v Nomura International plc [2007] IRLR 246 , approved by the Supreme Court in Hewage v Grampian Health Board [2012] IRLR 870 . We also had regard to the earlier guideline cases of King v Great Britain China Centre [1991] IRLR 513 CA and Bahl v Law Society & Others [2004] ILR 799. We had regard to the additional authorities referred to in the Respondent’s skeleton argument, Martin v Devonshires Solicitors UKEAT/0086/10 and Maksymiuk v Bar Roma Partnership UKEATS/0017/12. We also reminded ourselves of the approach to be adopted in the EAT to cases raising allegations of perversity and the high test laid down in well-known cases such as Yeboah v Crofton [2002] IRLR 634 , CA. The Claimant’s case[48]We turn then to the arguments before us. Mr de Mello submitted that the Claimant’s case had been put on the basis that the Respondent had misapplied the selection criteria for the post of Professorship or Readership in his case, such misapplication being unfavourable to him but favourable to others, who were white, such as the Claimant’s comparators. The Sneath ET failed to apply the correct approach in determining whether this difference in treatment was race based or not and failed to draw the correct inferences of discrimination. This was a case where it had been critical for the Tribunal to apply the correct approach in drawing inferences of race discrimination but it failed to do so. Although the Tribunal had made reference to King and Igen , in reality it had failed to apply the approach set down in those cases and equally failed to provide adequate reasons for its decision.[49]Before turning to the specific matters raised by this appeal, Mr de Mello further observed that the Tribunal had noted in respect of the promotions complaints that the Appellant had failed to prove that the applications were as good as his comparators (paragraph 206).[50]Mr de Mello then went into the more specific arguments raised by the appeals. In particular, he submitted that the Tribunal had failed to make the findings of fact that should have been made or draw the inferences that should have been drawn as to how the Respondent’s promotions committee approached and applied the criteria for promotion in the Claimant’s case. Specific examples were the committee’s inclusion of a requirement of currency of research ability and of supervision of research students and attraction of funding, when the written criteria only required evidence of capacity in these respects rather than currency.[51]Similarly, with regard to the trajectory that the committee apparently felt was important, the Tribunal should properly have found that this was the imposition of an additional criterion that served to shift the burden of proof and required explanation. The Respondent’s case[52]For the Respondent, it was contended that the Tribunal’s findings were clear. Ultimately the Tribunal had found that the Claimant was not appointed to the role or roles in question because his application did not satisfy the relevant criteria. It had rejected the contention that the promotions committee had made the relevant decisions either because of the Claimant’s Indian ethnic origin/race or because he had previously brought claims against the Respondent.[53]Particularly important for the Respondent, was the finding that Professor Plumb was an honest and reliable witness (paragraph 81) and the Tribunal’s acceptance of the Respondent’s evidence in all important respects.[54]Ms Connolly further submitted that, in reaching its conclusions, the Tribunal correctly directed itself as to the law including the burden of proof. On that point, the primary submission for the Respondent was that, where the Tribunal is able to make very clear findings of fact, the operation of the burden of proof is less relevant. It was the ET’s very clear findings of fact that made the present appeal hopeless. In making that submission, the Respondent relied on the observation of the former President of this court, Underhill J (as he then was) in Martin v Devonshires , in particular at paragraph 39, an observation that had been noted with approval by the Supreme Court in the Hewage case. Discussion and conclusions[55]In considering the arguments before us, we bore in mind that the EAT is not the Tribunal of fact. We have not had the benefit of seeing witnesses giving evidence and of their evidence being tested over a period of some 14 days. Our role is limited to the correction of errors of law. That can include consideration as to whether findings are perverse, but it is not for us to make the primary findings of fact or draw inferences from those facts. We make that point at the outset because at times we felt that Mr de Mello was seeking to draw us into making findings of fact in this case.[56]Taking Mr de Mello’s primary submission before us – that is, that the Employment Tribunal was effectively perverse not to draw an inference of discrimination or, perhaps more properly, not to find that the burden of proof had shifted in terms of the 2004/2005 application for promotion. The main point of the Claimant’s submission was that the Employment Tribunal should have found, as a primary fact, that the promotion criteria were incorrectly and unfavourably applied to the Claimant and that the Tribunal erred in not so finding and in failing to draw an inference from that.[57]We see the way that the case was put by Mr de Mello and the fact that the wording of the criteria does not specifically and expressly lay down a requirement of currency or talk in terms of a trajectory. We can see that this might have been a good argument before an Employment Tribunal, charged with having to determine these matters on the facts. It is possible that a different Tribunal would have been persuaded by the force of the argument and it is unclear whether the point was put as clearly and as forcefully to the Sneath ET as it has been to us. In any event, however, we agree with Ms Connolly: the Employment Tribunal rejected that submission and specifically found that the promotions committee had been entitled to have regard to the trajectory of the Claimant’s achievements (paragraph 59).[58]It is also relevant on this point that the Tribunal found further that the application form made clear the interest in a trajectory of research and that this was something to which the Claimant and his referees were alive to (paragraphs 60, 78 and 92).[59]That was all part of the factual matrix that the Tribunal had to consider. It was, in our judgment, entitled to reach the conclusions it did both as regards the case of the Claimant and also that in respect of his comparators.[60]Turning to the 2004/2005 promotion round - looking back to 2003/2004 and looking forward to 2005/2006 - the argument on appeal was that the Tribunal failed to apply the burden of proof correctly and require an explanation from the Respondent for why the Claimant had not got through to the second round in 2004/2005 when he had in 2003/2004 and then did in 2005/2006. That can only apply to the Readership application because that is the only one where he had got through to the second round, 2005/2006.[61]The difficulty with that submission is that the Tribunal did find an explanation to have been given (see paragraph 92):
“Of the claimant’s application for readership, the committee minute noted:- ‘(Also applied for professorship) It was noted that Dr D’Silva’s 2003/2004 application had been submitted to externals in order to demonstrate the claimant’s objectivity in dealing with it and that it had not been supported. The current application failed to demonstrate that Dr D’Silva met the required criteria in terms of his current publications record, academic leadership and recent research funding. The application did not show any progress over previous applications.’ The minute there refers to an exercise initiated by Professor Plumb arising out of the claimant’s challenge to the committee’s objectivity when considering his 2003/2004 application for readership. Professor Plumb approached three external independent academics who confirmed the committee’s view that the claimant’s overall performance had not met the criteria for appointment. This external assessment gave the committee confidence to consider objectively the claimant’s 2004/2005 application and enabled it to use the previous application as a starting point against which to measure the claimant’s progress.”
[62]On the Tribunal’s findings of fact, what had happened in 2003/2004 is that the committee itself had concluded that the application should not go through to a second round but wanted to test the objectivity of that conclusion by going out to external referees. The external referees confirmed the committee’s view. It was not that the Claimant had got through to the second stage in that year. He had simply been given special - and more favourable - treatment because of the committee’s desire to demonstrate objectivity and to test its own conclusions.[63]As for 2005/2006, again the reason why the Claimant progressed that year whereas he had not in the preceding year is explained in the Tribunal’s reasons (paragraph 175):
“Turning to the claimant’s application for readership, his comparators are Dr Binnie who applied in the previous year successfully and Dr Edensor who applied successfully in the 2005/2006 round. The Professoriate Committee decided to allow the claimant to go to stage 2 in respect of his readership application because he had now tied his application more directly and clearly to the criteria, had a further paper published in a refereed journal, had recently taken on a research master’s student and had obtained a further grant and grant review work.”
In other words, on the Tribunal’s findings of fact, the reason the Claimant did better in 2005/2006 is because he put in a better application. That was the explanation for the difference between this year and 2004/2005.[64]In any event, as Ms Connolly has submitted to us and as we accept, the Tribunal did make findings as to why the Claimant was not successful in 2004/2005.[65]We then turn to the second issue raised, i.e. that the referees selected by the Respondent at the second stage were not, or were not predominantly, from the Claimant’s area of expertise, namely Materials Chemistry.[66]One of the difficulties for the Sneath ET on this point was that the issue was apparently only raised during the course of the hearing so that the evidence that the Respondent was able to point to was not as full as it could have been (see paragraphs 214-216). In any event, however, the Employment Tribunal’s rejected this point on the facts. Its finding was (paragraph 216) that there was no less favourable treatment of the Claimant: at least one of his comparators was in the same position, with his referees also being drawn from a range of particular specialisms around his area of expertise.[67]The Employment Tribunal also accepted the logic of the Respondent’s case, i.e. that the evidence did not demonstrate that someone who worked in a slightly different field would necessarily be less favourable to the person about whom they are commenting. As Ms Connolly demonstrated to us, that was precisely the case with the Claimant: those who worked most closely in his field of expertise were in some respects less favourable to his application than those who had perhaps worked at some slight distance.[68]In our view, it was for the Employment Tribunal, which had before it all the evidence, to make the relevant findings of fact in respect of these matters in the context of that evidence. That is what it did and it is not something with which we can interfere.[69]The next issue before us related to the point that, in 2005/2006, the referees selected by the Respondent were chosen from higher grade academic institutions than the Respondent itself.[70]The problem for the Claimant here is that, contrary to what might have appeared to be the case at the preliminary hearing, the Sneath ET plainly rejected his case in this regard and made primary findings of fact contrary to the arguments being presented on his behalf (see, e.g. paragraph 188).[71]For completeness, we note that an additional point was raised relating to a late reference, but that was a case management issue, specifically rejected as a potential ground of appeal at the EAT preliminary hearing.[72]The penultimate issue identified was that seven referees were contacted in respect of the Claimant rather than three for the successful candidate. The complaint is the Tribunal failed to address this apparent difference between the Respondent’s treatment of the Claimant and that of its comparator.[73]That argument does not, however, withstand scrutiny. It is plain that the Tribunal did address the point and made findings contrary to the case being presented by the Claimant.[74]The context is made clear by the Tribunal’s findings at paragraph 189 in relation to past practice and other cases where a similarly large number of referees were approached. It went on (paragraph 222) to consider the comparator’s case and why this was different to that of the Claimant, thus explaining why different numbers of referees were approached. Moreover, (paragraph 254) the Tribunal specifically accepted the Respondent’s evidence on this point. Thus, the Tribunal has carried out precisely the exercise that it was obliged to do on this issue and made its primary findings of fact. Those were contrary to the Claimant’s case.[75]Lastly, again in relation to the ‘05/’06 round, the complaint was that the Tribunal should have looked for an explanation in respect of the fact that the referees approached in the second stage in Dr Rybakov’s case were nominated by him.[76]Ms Connolly submits that this is not an entirely fair way of representing the actual findings relating to Dr Rybakov’s referee. The point was that he had identified a referee at stage 1 who had not been approached at that stage but was then approached at stage 2. Thus it was only in an indirect sense that the referee had been identified by the candidate.[77]On this point, we accept the Respondent’s submission. Even on the Claimant’s case, it is difficult to see where the point could go and it is fair to note that Mr de Mello did not place much reliance on it. Even if the Claimant’s argument was accepted on this point, it would not be sufficient facts from which an Employment Tribunal could begin to draw an inference or find that the burden of proof had shifted. Conclusion[78]For all these reasons, paying due respect to the quality of the submissions that have been made before us, we consider that the appeal should be dismissed. Given the way in which the arguments were presented to us, we believe that this will mean that both appeals should be dismissed. We have, however, allowed for the parties to make additional representations on this question if appropriate.