The High Court held that the UK government had failed to make a reasonable adjustment by not providing British Sign Language (BSL) interpreters for two televised Coronavirus data briefings. However for ongoing Coronavirus briefings – given the use of detailed data slides – the court held that “in-screen” BSL interpretation was enough to satisfy the reasonable adjustment duty. In the circumstances the government was not required to provide an “on-platform” interpreter.
This case was a judicial review claim on the provision of British Sign Language (BSL) interpreters for televised Coronavirus briefings in England, so that they could be understood by Deaf people. The claimant argued that the UK government had failed to make reasonable adjustments by:
- providing no BSL interpreter for two “data briefings” (given by government scientists rather than ministers) on 21 September 2020 and 12 October 2020, and
- ongoingly using “in-screen” BSL interpreters for Coronavirus briefings in England, rather than “on-platform” BSL interpreters like in Scottish and Welsh Government live briefings.
An “in-screen” BSL interpreter is superimposed on the screen using a live “feed” from a studio, whereas an “on-platform” BSL interpreter is in the room with those giving the briefing.
Held by the High Court:
- The UK government had breached the reasonable adjustment duty for the two data briefings where no BSL interpreter was provided.
- However providing an “in-screen” rather than an “on-platform” interpreter was not an ongoing breach of the reasonable adjustment duty, particularly because an “on-platform” interpreter would make the detailed data slides less clearly visible.
- The government was not in breach of the Public Sector Equality Duty (PSED).
Summary of court judgment
Subtitles for the briefings were provided. However the court explained that BSL is a language in its own right, quite separate from English. For English many d/Deaf readers have an average reading age of 8 to 11 years; some d/Deaf people are reading at levels below 8 years of age (para 11).
The most relevant strand of the reasonable adjustment duty in this case was the requirement to provide an auxiliary aid or service, namely BSL interpretation.
This reasonable adjustment duty was “triggered” if disabled persons generally would, but for the provision of an auxiliary aid or service, be put at a substantial (ie “more than minor or trivial”) disadvantage in comparison with persons who are not disabled: below Comparative disadvantage. The relevant group of disabled people in the present case was Deaf BSL users (below). In applying this trigger test, any reasonable adjustments actually made were stripped out, so the provision of subtitles and in-screen interpreters was ignored. The court held that Deaf BSL users were put at a more than minor or trivial “disadvantage” (below) compared with non-disabled people, and indeed that this would be so even if the adjustments actually made were not ignored.
The trigger test for the reasonable adjustment duty was therefore met. The claimant had suggested possible reasonable adjustments, so the burden of proof shifted to the government to show that these were not steps it was reasonable for the government to have to take:
- Failure to provide any BSL interpretation for two data briefings: The government had failed to show it would not be reasonable to have to provide a BSL interpreter: below Reasonableness: the two data briefings with no BSL. Therefore the government had failed to make a reasonable adjustment. The claimant had a claim under the Equality Act as she had suffered a detriment (below). The court said damages (below) should be determined by the county court if not agreed between the parties.
- Ongoingly providing in-screen rather than on-platform BSL interpretation: The government had showed it was not reasonable for it to have to provide on-platform interpretation. In principle, an on-platform BSL interpreter gave access closer to that enjoyed by the rest of the public, and in-screen provision was a “lesser” step. The court set out four advantages (below) of having an on-platform interpreter. However the court decided on-platform interpretation was not legally required in the present circumstances, particularly because of difficulties of having an on-platform interpreter while also showing detailed data slides in a way viewers could see them properly. The data slides were currently shown in-screen. Below Reasonableness: in-screen rather than on-platform.
The court rejected a claim that the government was in breach of the Public Sector Equality Duty (below).
Service to the public
It was common ground that the government was providing a service to the public as regards the briefings. So technically the case was dealt with in that way rather than as one relating to public functions. (para 22)
The most relevant type of reasonable adjustment related to failure to provide an auxiliary aid or service. The court said this duty is triggered where “disabled persons generally” (EqA Sch para 2(2)) “would, but for the provision of an auxiliary aid, be put at a substantial disadvantage in relation to a relevant matter in comparison with persons who are not disabled” (EqA s.20(5)). There is a breach of the Equality Act if the breach in relation to the claimant of the reasonable adjustments duty involves subjecting that person to a detriment (below) (para 18). Points on deciding whether there was a comparative disadvantage:
- By using the phrase “disabled people generally” in the test of comparative substantial disadvantage, Parliament ensured that the test is not individualised but class-based. As the Services Code of Practice (para 7.24) says, the duty is anticipatory in the sense that it requires consideration of, and action in relation to, barriers that impede people with one or more kinds of disability prior to an individual disabled person seeking to use the service.
- In this case, the relevant class of disabled people (to whom “disabled people generally” refers) is Deaf BSL users: below Relevant class of disabled people.
- When deciding whether Deaf BSL users were at a substantial disadvantage. the court preferred the view that they should be compared with people who had no disability at all: below “Persons who are not disabled”.
- The test of comparative disadvantage must be applied after “stripping out” any steps already taken by way of potential reasonable adjustments, such as subtitles and (where provided) in-screen BSL interpretation: below Stripping out adjustments made.
- The court held (para 27-28) that in this case, without BSL interpreters, Deaf BSL users were indeed at a substantial (ie “more than minor or trivial”) disadvantage in comparison with persons who are not disabled. Indeed the nature and extent of the disadvantage were serious. See further below “Disadvantage”.
Comparative disadvantage: Relevant class of disabled people
At para 24 the court held that the relevant class of disabled people was Deaf BSL users – “people who are Deaf and use BSL”. It preferred this to the government’s suggestion of “people who are hearing-impaired”. However the court could not see that the answers in this case turned on that choice. The court gave its reasons for choosing “Deaf BSL users” as follows:
- Where reference is made in the Services Code of Practice to “people with different kinds of disability” (§7.24), examples given include: “people with dementia”; “people with… mental health conditions”; “people with … mobility impairments”; but also “visually impaired people who use guide dogs”; and “visually impaired people who use white canes” (§§7.24 and 7.25). The court said that if “visually impaired people who use guide dogs”, or “visually impaired people who use white canes”, can be the relevant class, it could not see what excludes “hearing impaired people who use BSL”.
- EqA s.6(3)(b) says persons who share the protected characteristic of disability are to be seen as those who “have the same disability”. Finnigan had seen the relevant group as “deaf persons”, and in MM the court focused on “mental health patients” In VC the court focused on “mentally ill detainees” (§153).
- However in the court’s judgment, the most reliable and authoritative guide was the idea of “people disabled in the same way”, derived by the Court of Appeal in VC at §153 from the Supreme Court in Paulley. The Supreme Court had seen the relevant group in Paulley as “wheelchair users”, not “people who are mobility-impaired”. Like “visually impaired people who use guide dogs” or “visually impaired people who use white canes”, this showed that the relevant group may be a sub-group. It fitted alongside Roads which took as the relevant group “those whose disability makes them dependent on a wheelchair”, ie “wheelchair users”.
- The court rejected a government suggestion of focusing on a sub-sub-group of BSL users “who would tune into the Briefings”. Contrary to the government’s argument, Roads did not view the relevant group as “wheelchair users using [the Thetford] train station”. In Finnigan the court did not take the relevant group as “deaf persons whose properties may be searched by the police”. In Allen the focus was not on “wheelchair users wishing to use services at the main Sheffield branch of the bank”, but on wheelchair users.
Comparative disadvantage: “Persons who are not disabled”
The reasonable adjustment duty is triggered where [the relevant class of] disabled persons generally – here “Deaf BSL users” (above) – “would, but for the provision of an auxiliary aid, be put at a substantial disadvantage … in comparison with persons who are not disabled”.
In MM, the Court of the Appeal spoke of a comparison between mental health patients and “those not so disabled” (§59), ie not disabled “in the same way”.
However the preference of the court in the present case was to compare the relevant group with people who are not disabled at all. This reflected the statutory language (s.20(5)). It fitted with the Services Code of Practice which says (at §7.13): “The disadvantage created by the lack of a reasonable adjustment is measured by comparison with what the position would be if the disabled person in question did not have a disability” (the word is “a” not “the”). Also, said the court, this conclusion “avoids the risk of introducing invidious comparisons with those who may have other disabilities, disadvantages and needs (for which different reasonable adjustments may also be necessitated)”.
In any event, the court was quite satisfied that the outcome could not, in the circumstances of the present case, turn on which interpretation is chosen.
Comparative advantage: Stripping out adjustments made
The test of comparative disadvantage must be applied after “stripping out” any steps already taken by the government by way of potential reasonable adjustments – for example subtitles in this case. This was the approach taken by the Court of Appeal in Finnigan. (para 26)
Similarly, as regards whether BSL interpretation should be on-platform rather than in-screen, comparative disadvantage was to be judged assuming no BSL interpretation at all. Whether the BSL interpreter should have been on-platform rather than in-screen arose at the later stage of deciding reasonableness (below). (para 45)
Comparative disadvantage: “Disadvantage”
The court held that Deaf BSL users were at a more than minor or trivial “disadvantage” compared with non-disabled people, whether or not the availability of subtitles or (for ongoing briefings) the in-screen interpreter was taking into account.
Meaning of “disadvantage”
The court first considered the meaning of “disadvantage” (para 27):
- The court had regard to the Services Code of Practice (§5.10):
- This suggests, as relevant to “disadvantage”, the concept of “something that a reasonable person would complain about”. So “an unjustified sense of grievance would not qualify”.
- However the disadvantage “does not have to be quantifiable and the service user does not have to experience actual loss (economic or otherwise)”.
- It is “enough that the person can reasonably say that they would have preferred to be treated differently”.
- The court said one reference point, albeit in the context of public functions involving detriments, is Parliament’s description of “an unreasonably adverse experience” (EqA Schedule 2 para 2(5)(b)). The court had in mind the relevance of evaluating not only whether the trigger test of comparative substantial disadvantage is met (as an on/off switch), but also the nature and extent of any substantial disadvantage (on a scale). That was important when it comes to addressing the question of the content of the reasonable adjustments duty, and the question of reasonable steps in all the circumstances.
Disadvantage as regards the data briefings with no BSL interpreter
The court held that here Deaf BSL users were at a substantial – meaning “more than minor or trivial” – disadvantage compared with non-disabled people. Indeed the disadvantage was one whose nature and extent were serious. The disadvantage had to be judged from the base position ignoring any reasonable adjustments actually made, so assuming that no subtitles were available: above Stripping out adjustments made. However the court said it would have reached the same conclusion even if it took account of the subtitles. In reaching its decision that there was a more than minor or trivial disadvantage, the court took into account (para 28):
- The information related to a matter of the greatest public interest and a vital concern, the pandemic
- The data briefings were important, focusing on objective data and led by government scientists.
- The circumstances were unprecedented and challenging for the public, who needed access to information to manage their conduct and expectations for the future. Messages or alarm or reassurance, about being “in this together” and acting responsibly, about “following the science”, required inclusion and accessibility.
- This was information which was being provided in a fast-moving context, needing clarity, with frequent resort to science and statistics.
- Given the position regarding BSL and Deaf BSL users (including low literacy levels), without BSL interpretation there was a clear barrier for a vulnerable and marginalised group, undermining accessibility of information. The lack of BSL was a failure of inclusion, suggestive of not being thought about. The immediate experience was of important urgent messaging being delivered to the public, known to be being provided, but with an inability to access or understand it.
- “The substantial, foreseeable and palpable effect would be an exclusion, and a justified sense of grievance, about which a reasonable person would certainly have good reason to complain, and about which affected people would reasonably say that they would have expected and urged – let alone preferred – to have been treated differently. All this, moreover, for a significant and substantial number of people.”
Disadvantage as regards lack of on-platform interpretation
As regards ongoing provision of in-screen rather than on-platform BSL interpretation, again whether Deaf BSL users were at a substantial disadvantage had to be judged from the base position without any reasonable adjustments: above Stripping out adjustments made. So the availability of an in-screen BSL interpreter was ignored, and for the same reasons as above Deaf BSL users were at a more than minor or trivial disadvantage. Indeed again, for the same reasons, the disadvantage was one whose nature and extent were serious. (para 45(3))
The court said Deaf BSL users would be at a more than minor or trival disadvantage even if the in-screen interpretation were not disregarded. However in that case the court would not describe the nature and extent of the disadvantage as “serious”. But even if the test had been whether in-screen as opposed to on-platform interpretation put Deaf BSA users at a more than minor or trivial disadvantage compared with non-disabled people, the test was met. (para 45(4))
Reasonableness: the two data briefings with no BSL
The claimant had identified BSL interpretation as a possible adjustment for these two data briefings, and the burden of proof had shifted to the government to show that this would not have been a reasonable adjustment (para 33).
- The court was very clear that subtitles were not an answer for BSL users. Subtitles in this context were “fast-moving text in relation to technical information in a language which is not the first language of BSL users and assumes a level of literacy (above) in that further language which very many of them simply will not have”. The idea that “subtitles are an answer” amounted to a stereotypical opinion formed without proper knowledge of people with this disability, constituting “prejudice” according to EHRC Technical Guidance. (para 35(4))
- In this case there was no conflict between considerations pulling in different directions, needing to be reconciled. “There is no evidence that anything would have been sacrificed, or detracted from, in securing BSL interpretation for all Briefings.” BSL interpretation would be effective in overcoming the substantial disadvantage that Deaf BSL users face in accessing the relevant services, it was practicable for the government for provide it, and there was no financial cost impediment. (para 35(6)
- One argument by the government was that it was entitled to rely on the BBC to make suitable provision for interpreters – and that it was the BBC, unknown to the government, who decided BSL interpretation was not needed for these two data briefings led by scientists rather than ministers (para 34). However the court said that as a service provider the government was itself responsible for ensuring that reasonable adjustments were made. Further the duty was a proactive, anticipatory one. A serious underlying problem was that arrangements for interpreting briefings into BSL were allowed to remain the subject of an undocumented informal arrangement between the government and the BBC. The government did not have visibility and clarity as to what the BBC’s BSL interpretation would cover, nor as to any “warning” if a gap arose as happened on these two data briefings. “Undocumented informality produced an unknown, unknowable, uncontrollable and unalerted gap.” (para 35(7))
Reasonableness: in-screen rather than on-platform
The claimant argued that in-screen BSL interpretation was not enough for ongoing briefings, that it should be on-platform. Since the claimant had identified on-platform BSL interpretation as a possible adjustment, the court said the burden of proof had shifted to the government, to show that this would not be a reasonable adjustment (para 45(5), 55(3)).
The court rejected the government’s argument that, even put at its highest, on-platform and in-screen provision of BSL interpreters were “reasonable alternative methods” falling within the government’s latitude. The court said the case for on-platform BSL interpreters was a powerful one, and was directly linked to the legislative policy of approximating access as close as reasonably possible to that enjoyed by the rest of the public. It had four distinct strands: (para 46-48, 53, 55(5))
- Replication: On-platform interpretation is automatically included when clips are shown on TV or in social media. The government argued that (unless the broadcaster provided BSL anyway) the clip would still appear within coverage not available in BSL. However the court said replication retains real force, particularly in conjunction with inclusion and promotion below.
- Elimination: Having an on-platform interpreter, standing behind the speaker and in camera shot, guarantees that nothing can be missed through miscommunication or a technical issue or for any other cause or reason. The evidence included examples of breakdowns with in-screen interpretation. The government argued that there had only been a very small number of technical issues over the past 14 months, and for example an on-platform BSL interpreter might be late, unwell, or unavailable. However the court considered that elimination is still a concrete advantage.
- Inclusion: With an on-platform interpreter, standing behind the speaker and in camera shot, the experience for Deaf BSL users is that they are being included, alongside everybody else, through the same primary routes and the same choice of routes as are available to the public generally. This, clearly, was the “closest approximation” to the experience of the public generally.
- Promotion: An on-platform interpreter gave a clear message and experience – for everybody – that Deaf BSL users are being included. As the EHRC compellingly put it, an on-platform interpreter constituted a demonstration of the commitment to equality for all. The government’s PSED Assessment acknowledged that one of the advantages for BSL users of on-platform interpretation is that the BSL interpreter is visible automatically to all viewers which raises awareness of BSL views and needs. It could foster the inclusion of d/Deaf people (and disabled people generally) in wider society, particularly during the pandemic.
The court said that in principle, on-platform provision of a BSL interpreter approximates accessibility for Deaf BSL users more closely to that enjoyed by the rest of the public. Subject to questions of reasonable practicability and of what is reasonably possible, on-platform provision better promotes the legislative policy, and in-screen provision is a “lesser” step. (para 55(1))
In the light of the case for on-platform provision (the four points above), the government could not discharge the burden of proof of showing that in-screen BSL interpretation was a sufficient reasonable adjustment unless it could point to some disadvantage of on-platform provision (para 55(7)).
- The government had previously argued that the briefings were held in a cramped room where proper social distancing was not possible with an interpreter present. However this justification had faded since the briefings moved into a larger purpose-built studio in March 2021. (para 50, 55(8))
- Disadvantages identified in the government’s PSED Assessment as regards what is best for Deaf BSL users themselves – in particular the prominence and visibility of the BSL interpreter in-screen compared with on-platform – were unpersuasive. They were not linked with what Deaf BSL users or groups representing them said. There was no evidence that “appropriate lighting”, and a fixed position, were unachievable on-platform, still less in the new Downing Street press room. (para 55(9))
- Taking the points made so far, the court would not have accepted that the government had discharged the burden of showing it had complied with the reasonable adjustment duty. (para 56)
- However the government had identified a substantial problem concerning general accessibility and data-slides. Data slides were often used in briefings. They were presented “in-screen” for clarity. The government argued that filming the slides on a screen with an interpreter stood next to them would make the slides less clear. If the data slides were “in-screen”, Deaf BSL users would need an in-screen BSL interpreter when the slides were being shown. The court was shown a screenshot of a Scottish briefing where a slide saying “Coronavirus: Stay Safe: Protect Others: Save Lives” was visible in the background without being blocked by the interpreter or the First Minister. However the UK government argued “convincingly” that the visibility of detailed data slides used in English briefings would be compromised by adopting that approach. (para 56(11)-(13))
- There was also real force – though not as much as for data slides – in the similar point made about journalists asking questions being shown in full-screen. The government had said that requiring two people always to be in shot would significantly limit the angles and scope of any footage. (para 56(13))
- These points needed to be considered “in the round”, together with other points including as to “adequacy and proportionality” discussed in para 52 of the judgment, which where were not of themselves enough to discharge the government’s burden of showing compliance. (para 56(14)-(15))
- In conclusion, the court was satisfied that the government had discharged the burden of showing that it was not in breach of the reasonable adjustment duty: that it had, through in-screen provision, taken such steps as it was reasonable to have to take to provide BSL interpreters for the briefings.
Detriment to the claimant
Detriment: Data briefings with no BSL
The UK government had failed to make a reasonable adjustment of providing BSL interpretation for two data briefings (above Reasonableness: the two data briefings with no BSL). However, said the court (para 36), the claimant could only claim if she had suffered a “detriment” (step 2 at Reasonable adjustments by service providers>Two-stage legal test for anticipatory duty).
The court decided she had suffered a detriment (para 36). It accepted evidence that she had the reading understanding of a 7-year-old, and could not understand the data in the data briefings because there was no BSL interpretation – so the information was not in a language she could understand. The court said that it approached “detriment” having in mind the way in which it is described in the Services Code of Practice (§5.10): something that a reasonable person would complain about; not merely an unjustified sense of grievance. The court said detriment could arise even if “for … a limited period” (VC §175). There was “at least a real prospect” that BSL interpreters “would have made a difference” (VC §177) in the present case. Furthermore:
“36. … Saying that the Claimant could get the information in the end brings to mind telling a wheelchair user to wait for the first bus without a buggy in the wheelchair space [Paulley]; or telling Mr Roads to travel to Ely, change there and then travel home. The Claimant’s position is not like that of Mr Finnigan, for whom the absence of a BSL interpreter during police searches “did not have any effect” on communication (Finnigan §44).”
Detriment: In-screen versus on-platform
The court held that it was not reasonable (above) for the government to have to provide on-platform interpretation, rather than in-screen. However if an on-platform interpreter had been reasonably required, the court said the claimant would have suffered the “detriment” necessary to enable her to claim. Its reason for this was the nature of the four strands (above) in the case for on-platform provision, together with the concept of “detriment” as something which a reasonable person could complain about; not an unjustified sense of grievance (above). (para 58)
Remedy even if no detriment?
Even had the court found that the claimant did not suffer a detriment, and so was not entitled to damages under the Equality Act, the court would still have been “most reluctant to refuse a remedy”. The court would have wanted to consider whether it was precluded from a finding of breach and a declaration or mandatory order. (para 58)
Damages for lack of BSL at two data briefings
The claimant’s claim for damages in respect of lack of BSL at two data briefings was transferred to the county court for determination, following a period of three months to give the parties a chance to agree damages. (para 37, 59(A)(2))
Public sector equality duty (PSED): not having on-platform BSL
(Broadly the PSED requires public bodies to have due regard to the need to eliminate unlawful discrimination, advance equality of opportunity and foster good relations.)
Was the government in breach of the PSED in relation to ongoingly having an in-screen rather than on-platform BSL interpreter? The court held no (para 39-44).
The court said the PSED is concerned with “process” rather than “outcome” (or “result”) – although “process” here includes enquiry, thinking-process and reasoning-process. Enforcement of the PSED does not extend to the question of what “weight” should be given to the equality implications of a decision or action.
At a late stage (almost “at the door of the court”) the government had produced a PSED Assessment considering the issues and deciding in favour of in-screen interpreters. However no evidence before the court suggested that the writer of this assessment was reasoning backwards from a chosen policy position being defended before the court. The court accepted the assessment as what it purported to be, an objective and open-minded consideration of the issues. On that basis, the PSED Assessment was a rigorous evaluation which recognised the features of the statutory PSED duty and which could not, in any material respect, be said to be a failure of “due regard”. Among other things too, the court was satisfied that the government had taken reasonable steps to make enquiries about what may not yet be known to it.
Criticisms of the PSED Assessment made by the claimant did not relate to the “process” or “enquiry” but operated as weaknesses in the government’s reasoning when the court came to apply the objective standard in enforcing the reasonable adjustments duty (above). The government had the fruits of enquiry, sufficient to make a lawful decision, with the issues having been raised and views expressed.
Comments: Anticipatory duty – no knowledge needed of particular disabled person
The case illustrates how service providers are required to make adjustments for disabled people generally, without waiting for a particular disabled individual to present themself (Reasonable adjustments by service providers).
This ties in with the Equality Act not including a defence for a service provider who does not know of the particular disabled person. There was no suggestion in this case that the government’s lack of knowledge of this particular claimant gave it a defence to the reasonable adjustment claim. As discussed at that link, however, there are other circumstances where knowledge may be more of an issue.
Comments: Detailed analysis
The High Court gave a particularly detailed analysis of (its view of) the reasonable adjustment duty on service providers. Though not as authoritative as the various Court of Appeal decisions on this duty, some of its discussion may influence future cases, particularly in the county court.
Comments: Relevant class of disabled people, and being put at a disadvantage
I find the court’s discussion of what is the relevant class of disabled people (above) particularly interesting, as this has been very little considered by the courts. It tends to be uncontentious. The court held that the relevant class here was “Deaf BSL users”. My discussion on this – though not yet updated for this case – is Reasonable adjustments by service providers>Particular kind of disability.
For the reasonable adjustment duty to apply, this class of disabled people must be put at a more than minor or trivial disadvantage, compared with non-disabled people. What is perhaps more important than how the court defined the class, is that the court said it would have made no difference to the answers in this case if the relevant class were “people who are hearing-impaired”, as suggested by the UK government. This is a very much larger class than people who use BSL, and supports the idea that a class of disabled people may be disadvantaged even if only a relatively small proportion of those in it are disadvantaged: Reasonable adjustments by service providers>No need for all people who stammer etc to be disadvantaged. According to Facts and figures (rnid.org.uk) it is estimated that 151,000 people in the UK use British Sign Language, of which 87,000 are Deaf. That compares with, for example, an estimated 1.2 million people in the UK who have hearing loss greater than 65 dBHL. This is a level at which the person has very great difficulty hearing and taking part in a conversation (natsip.org.uk). Those 1.2m people at least – and indeed more – will have a hearing impairment which is a disability within the Equality Act (the effect of a hearing aid is disregarded in determining disability), and so presumably fall within the class of “people who are hearing-impaired”. Though it did not discuss the issue, the court evidently considered that this class of people was put at a more than minor or trivial disadvantage by the two data briefings not having a BSL interpreter, even though only a fairly small proportion of them used BSL.