Rhieni Dros Addysg Gymraeg, R (on the application of) v Neath Port Talbot County Borough Council & Ors

Schools in Wales are known as Welsh-medium schools if they deliver education through the medium of the Welsh language. If education is delivered through the medium of the English language, they are called English-medium schools. The claimant is a charitable body established in 1952, funded by the Welsh Government. I am told its name translates to English as Parents for Welsh-Medium Education. Its work is to increase Welsh-medium education in Wales.

The main question in this case is whether the defendant local authority (the council) lawfully or unlawfully decided in October 2021 (the decision) to close three English-medium primary schools in the Pontardawe area and establish one new English-medium primary school at Parc Ynysderw, without including in the prior consultation exercise an assessment of the impact of the proposal on the Welsh language (a Welsh language impact assessment or WLIA).

The claimant’s oral submissions were mainly delivered in Welsh but were rendered intelligible to me and others unable to understand the language by excellent simultaneous translators, for whose services the court is very grateful. The parties then provided, at my request, a written update on the factual position which was developing at the time of, and after, the oral hearing. This followed a change in political control of the council in May 2022.

It is now not clear whether the council will implement the decision even if it succeeds in defending this claim. It is seeking a dialogue with the Welsh Government and may decide to review the decision. Nonetheless, the parties agreed that the legality of the decision, though taken in October 2021 under the previous regime, has not become an academic question. The decision would not, if implemented, take effect until 1 September 2024. Implementation is on hold at present, but I am content to provide this judgment as the parties wish.

Changes in provision of public education in Wales are governed by Part 3 of the School Standards and Organisation (Wales) Act 2013 (the 2013 Act). A mandatory school organisation code (the Code) issued by the Welsh Ministers must be followed by local authorities, school governing bodies and others in connection with proposals for change (section 38), when exercising their functions. The current version of the Code dates from November 2018.

A local authority may make proposals for the establishment of a new community school (section 41(1)(a) of the 2013 Act) or to discontinue, i.e. close, a community school (section 43(1)(a)). Before publishing its proposals in accordance with the Code (as required by section 48(1)), it must “consult on its proposals in accordance with the Code” (section 48(2)). A person may object to the proposals within an objection period of 28 days from publication (section 49(1) and (2)).

Where the proposals come from a local authority and do not affect sixth form education, the authority itself may decide whether they should be implemented (section 53(1)). If the authority decides that they should be, it must implement the proposals within three years of the envisaged implementation date, unless it then decides that implementation would be unreasonably difficult or that circumstances have so altered since the proposals were approved that implementation would be inappropriate (section 55(1)(b), (2), (3) and (5)).

Section 1 of the Code is headed “Development and consideration of proposals”. Within it, paragraph 1.4 states (emphasis in original):

“Where proposals affect schools where Welsh is a medium of instruction (for subjects other than Welsh) for some or all of the time, local authorities should carry out a Welsh Language Impact Assessment.”

Section 3, headed “Consultation”, includes paragraph 3.1 stating the principles governing public consultation exercises generally (called the Sedley criteria after the distinguished judge of that name): the consultation process should be undertaken when the proposals are still at a formative stage; should include sufficient reasons and information for particular proposals to enable intelligent consideration and response; should provide adequate time for consideration and response; and should ensure that the product of consultation is conscientiously taken into account when the ultimate decision is taken.

The required “Consultation document” is addressed at paragraph 3.4. The published document must include a long list of items, including “Details of affected schools”. The details that must be included relate to those schools that are, in the words of paragraph 3.4, “likely to be affected by the proposals”, a phrase repeated several times. An example is given in parentheses:

“(for example, in the case of a proposal to close a school information should be provided about all the surrounding schools to which it might reasonably be considered that pupils may wish to transfer); …”.

Later, still within paragraph 3.4 under the sub-heading “Consultation details”, included in a long list of requirements is the following (emphasis in original):

“Where any school involved or affected provides teaching through the medium of Welsh the following information must be included in the consultation document:

 an assessment of the impact of proposal on the Welsh language (a Welsh language impact assessment must be included either in the main part of the consultation document or as an Annex); and

 an explanation of how the proposal forms part of the WESP.”

The WESP is the Welsh in Education Strategic Plan, which under the Welsh in Education Strategic Plans (Wales) Regulations 2019 (enacted pursuant to provisions in the 2013 Act) local authorities are required (section 85 of the 2013 Act) to submit to the Welsh Ministers for approval prior to the plan being published.

Various legal instruments promote use of the Welsh language in Wales, as part of the Welsh culture and heritage that needs to be preserved and nurtured. For example, public bodies must set and publish objectives towards the achievement of the “well-being goals” in Part 2 of the Well-being of Future Generations (Wales) Act 2015 (sections 3 and 4) (the 2015 Act). The goals include “a Wales of vibrant culture and thriving Welsh language”.

The Welsh Language Standards (No 1) Regulations 2015 (the 2015 Regulations) specify “policy making standards” (regulation 2(2) and Schedule 2). The written standards numbered 88 to 97 relate to “considering the effects of a body’s policy decisions on the Welsh language”. The gist is that bodies making policy decisions must consider how to make policies that have positive or increased positive effects on opportunities to use the Welsh language and on treating it no less favourably than the English language.

Three primary schools in the Pontardawe area currently provide English-medium education for children aged 3 to 11: the Alltwen, Godre’r Graig and Llangiwg primary schools (the three schools). Each was ranked “good” by the schools inspectorate body, Estyn, when last inspected in 2017 to 2019. Their combined capacity is 548 full-time pupils and 102 part-time nursery pupils.

On 21 October 2020, the council decided to consult on a proposal to replace the three schools with a single new English-medium school (the new school) from 1 September 2024, with a capacity of 630 full-time pupils and 140 part-time nursery pupils. The proposal was adopted on the basis of a report (the October 2020 report or the report) in the name of Mr Andrew Thomas, the council’s then head of transformation, but largely written by Ms Rhiannon Crowhurst, the council’s witness and then strategic school improvement programme manager.

The October 2020 report was long and detailed. It explained in depth the reasoning underlying the proposal. The new school would be newly built, with funding from the council and the Welsh Government, already approved in principle. I will refer mainly to parts of it dealing with Welsh-medium education and the Welsh language.

That was first addressed under the heading “Reasons for the Amended Proposal – Informal Consultation with Stakeholders”, referring to a series of meetings. Paragraph 17 recorded that “[c]omments were … received about possible impacts on the Welsh-medium schools in the area and the impact on the development of the Welsh language as a whole”. The report went on to set out lots of information about the three schools and the proposed new school.

The report then set out eight options, of which the sixth – replacement of the three schools by the new school - was the preferred option. There was then a heading “Impact on pupils and parents”. The new school, it explained, would serve the combined catchment areas of the three schools. It would have plenty of playing fields and green space. Pupils of the three schools could transfer to the new school.

The impacts on various things were then set out under headings including the word “impact”: on pupils and parents; on travel arrangements; on governors; on special needs provision; and financial impact. The next heading was “Integrated Impact Assessment”. Paragraph 222 recorded that a first stage “Integrated Impact Assessment” (the IIA) had been undertaken to help discharge the council’s duties under various legal instruments, including the 2015 Act and the 2015 Regulations.

Paragraph 223 then stated that the “first stage screening assessment” was attached as appendix A and “[t]his indicates that the proposal is likely to have an impact on pupils, staff and other potential users of the new school” and that “[t]he proposal may also impact on opportunities to use the Welsh language”. The report went on to explain that the IIA is “an ongoing process” which would be put before members in fully completed form before the final decision.

Appendix A was headed “Impact Assessment – First Stage”. The format of the first part was that of a standard equality impact assessment, examining and grading the likely impact on those with the familiar protected characteristics: age, disability, gender reassignment, marriage and civil partnership, pregnancy and maternity, race, religion and belief, sex and sexual orientation.

The next part asked, first, “Does the initiative impact on … [p]eople’s opportunities to use the Welsh language”? The answer was a tick against the word “Yes”. The impact was “M” for medium. The reasons given for that decision and “[h]ow it might impact” were stated as follows:

“The Swansea Valley area is a linguistically significant area as it contains the highest number and percentages of Welsh speakers in NPT, and is amongst the highest in Wales as a whole. There are concerns however that the number of Welsh speakers is declining and work has been ongoing to ensure the language is promoted and grows.

Alltwen, Godre’rgraig and Llangiwg primaries teach Welsh as a second language and this will also be the case for the new school. Currently 25% of staff across the three schools are fluent or fairly fluent Welsh speakers. By combining the three schools Welsh language provision could be improved as there could be greater impact on having a more concentrated group of Welsh speakers able to support pupil and staff language development skills through the medium of Welsh. This aspect will need to be explored in more detail as part of the full assessment to better understand possible impacts.”

The next question asked was whether the initiative impacts on “[t]reating the Welsh language no less favourably than English”. The answer was again yes. The impact was said to be “L” for low. The reasoning and “[h]ow it might impact” was stated thus (with my emphasis):

“A new build EM school could attract pupils currently attending WM schools, which could hinder development of Welsh as a first language in the area, and as such could be seen to be treating the Welsh language less favourably than English.

Significant work is being undertaken in a number of the WM schools in the area which will improve provision for teaching and learning and provide extra pupil places, which should mitigate against the impact of a new school.

Further exploration of this aspect will be needed as part of the full assessment.”

In the body of the October 2020 report, the next section (paragraphs 230-237) referred to the WESP. It was noted that the three schools teach Welsh as a second language and the new school would also do so. The Swansea Valley area has the highest number of Welsh speakers in Neath Port Talbot. It was not expected that the proposal would reduce Welsh language provision in the area. However, “further exploration of possible impacts on the Welsh language will be undertaken during consultation” (paragraph 237).

Following the cabinet’s decision to consult on the proposal, a consultation document was produced by Mr Thomas and Ms Crowhurst on or about 3 November 2020, the start of the consultation period (the consultation document). It repeated in large part the content of the October 2020 report. The schools “affected” by the proposal were named as the three schools, i.e. those that would close.

As for other schools that “might be affected by this proposal”, only Cwmtawe Community School was named. That is not a primary school, nor Welsh-medium; it provides English-medium education for 11 to 16 year olds at a site close to Parc Ynysderw, the proposed site of the new school. There was no mention of any Welsh-medium primary schools whose pupils could be attracted to the new, English-medium school.

The concern expressed in Appendix A to the October 2020 report that this could be seen as treating the Welsh language less favourably than English was not directly mentioned. Nor was there any mention of a need to “mitigate against the impact of a new school”, as expressed in Appendix A. The consultation document explained, as in the October 2020 report, that the sixth of the eight options considered was the preferred option. The expected “impact” of the proposal on pupils, governors, etc was repeated from the October 2020 report.

The IIA was then referred to. It had indicated the need for “a more in depth assessment” if the proposal were to proceed. The proposal was likely to have an impact on pupils, staff and other users of the new school; and “[t]he proposal may also impact on opportunities to use the Welsh language”. After the consultation, “greater information” would help to “inform the full assessment”.

There was then a heading: “What is the impact on the Welsh language?” The points made in the October 2020 report were repeated, with reference to the WESP. It was expected that the new school would become known for good teaching of the Welsh language. The consultation document then continued:

“Primary education through the medium of Welsh is available at Ysgol Gynradd Gymraeg Pontardawe and Ysgol Gynradd Gymraeg Trebannws. Additionally Ysgol Gymraeg Ystalyfera –Bro Dur, which is approximately 5 miles from the proposed site, provides all through 3-19 education through the medium of Welsh.

Significant investment has taken place at Ysgol Gymraeg Ystalyfera–Bro Dur through projects in both Band A (circ. £17.5m) and Band B (circ. £9m) of 21st Century Schools Capital Funding Programme.

In addition funding of £1.6m has been secured to improve provision at Ysgol Gynradd Gymraeg Pontardawe, including extended and enhanced childcare and Foundation Phase facilities. Work is expected to be completed in 2021.

It is not expected that this proposal would reduce Welsh language provision in the area, however a Welsh language impact assessment will be undertaken during consultation.”

The consultation was to run from 3 November 2020 to 19 January 2021. It would be followed by a consultation report, to be published on 26 February 2021. A statutory notice would be published on 13 April 2021, with the 28 day objection period running thereafter until 11 May 2021. A determination would be made within 16 weeks thereafter. This timetable was explained and set out in the consultation document.

A comment form at the end of the consultation document invited participants to answer various questions about the proposal, including whether the respondent believed it would have a positive, negative or neutral impact on opportunities to use the Welsh language; and whether the respondent believed the proposal would treat the Welsh and English languages equally. The consultation then took place.

While it was taking place, in January 2021 Ms Crowhurst commissioned a WLIA from a consultancy firm, Red Shiny Apple Limited (the WLIA), through its director, Ms Anna Morgan. This was a condition of Welsh Government grant funding for the project. By 18 February 2021 the council was “continually being hounded” by the Welsh Government to produce the council’s WLIA, as Ms Crowhurst put it at the time in an email to Ms Morgan.

The WLIA went through several drafts between February and April 2021. The drafts were edited and commented on by Ms Crowhurst in email exchanges with Ms Morgan. The final version owed much to the efforts of both. It was dated 16 February 2021 but finalised on 7 May 2021, with the date unchanged. It was eventually published in June 2021 as a council document, not under the banner of Red Shiny Apple Limited, though the council acknowledged that it was a report that had been commissioned from an independent consultant.

The WLIA noted that the Swansea Valley, where the three schools are located, is a “stronghold” of the Welsh language. The proposed reorganisation was of English-medium schools only. That did not mean there would be no impact on the Welsh language. The consultation responses were available to the authors of the WLIA. They revealed that nearly half those who responded thought the proposal would have a negative impact on the language; nearly half thought the impact would be neutral; and 3.1 per cent thought it would be positive.

The WLIA then discussed whether the new school would create an incentive for parents to withdraw their children from Welsh-medium schools in the catchment area. In general, that proposition was not accepted. Examples were cited of instances in which a move from Welsh to English-medium education had not led to an adverse impact on Welsh-medium education; though some catchment area parents to whom the new school was closest could opt for the nearest school.

The fourth section of the WLIA referred to recommended “mitigating factors in terms of Welsh-medium education, to balance against the proposals” regarding the three schools. The recommended actions were those subsequently set out in the consultation report (to which I am coming). The recommended measures consisted of monitoring actions, i.e. examining future parental choice to see whether the parents moved their children away from Welsh-medium education.

The results of the monitoring would then feed into the “WESP” process rather than influencing whether to implement the proposals. The rest of the fourth section referred in some detail to projects for investment in Welsh-medium schools in the locality, including at two of the Welsh-medium schools within the same geographical area as the new school.

Later, under “Social Deprivation Considerations”, the WLIA referred to the possibility that the proposals “could have an impact on the accessibility of Welsh-medium education”. Parents of Welsh-medium school children in the area could be attracted to the new school because it was “brand new”, on a “new build site” and offering “perceived positive opportunities” for children. This “possible issue”, the WLIA noted, “has been identified in the Council’s ‘Integrated Impact Assessment – first stage’ document”, i.e. in the IIA.

That was a reference to the passages in Appendix A to the October 2020 report, quoted above. The WLIA continued:

“There are both Welsh language and socio-economic considerations to consider therefore, however as shown in Section 4, a number of Welsh-medium proposals across the county borough could mitigate any adverse effects, in particular the investment in, and expansion of, Ysgol Gynradd Gymraeg Pontardawe, which is broadly in the same geographical area as the current Alltwen, and Llangiwg Primary Schools, and At Ysgol Gymraeg Ystalyfera-Bro Dur which is near to Godre'rgraig Primary School.”

The report on the consultation (the consultation report) was dated 26 February 2021 but published in May 2021. The conclusion was that the proposals should proceed. The consultation report explained that there were 234 responses and, in addition, an online petition with 413 signatures. The large majority of those who responded opposed the proposal. Among the concerns raised by, among others, the Undeb Cenedlaethol Athrawon Cymru (National Union of Teachers of Wales (UCAC)) was an adverse effect on use of the Welsh language.

Two of three Welsh-medium primary schools in the area, cited by UCAC, responded through their governing bodies: Ysgol Gynradd Gymraeg (YGG) Pontardawe and YGG Trebannws. According to the consultation report “[b]oth responses focus on the impact this proposal may have on Welsh-medium schools in the area and on the development of the Welsh language as a whole.”

The consultation report recorded UCAC’s concern that these schools would lose pupils to the new school, with its superior facilities, thereby increasing English-medium education and decreasing Welsh-medium education in the area. These arguments were countered in detail in the “Officer responses” section of the consultation report, under the heading “Impact on other Schools”:

“Comments have also been received which state that a new English-medium school will negatively impact on the Welsh-medium schools in the locality, and comparisons have been made with the impact on Ysgol Gyfyn Ystalyfera when Cwmtawe Community School opened in 1996. It has been stated that at this time greater numbers were attracted to Cwmtawe Community School due to it being a new building and subsequently pupil numbers dropped at Ysgol Gyfyn Ystalyfera.

The suggestion that the condition of the building and available facilities are a more important factor than the language medium of education when parents are choosing a school for their children is debatable. Data reveals that new build English-medium schools in Neath Port Talbot which have opened in the last 10 years have had no significant impact on nearby Welsh-medium schools. Awel-y-Mor Primary opened in 2013 and despite significantly more pupils attending than predicted, it is evident that these pupils did not migrate from YGG Rhosafan, the nearest Welsh-medium school. In fact numbers at YGG Rhosafan have steadily increased, from 336 in 2021 to 381 in 2020. If it were the case that new English-medium schools significantly attract pupils who would otherwise have attended a Welsh-medium school then YGG Rhosafan should be hugely impacted as the catchment area of the school has also seen the opening of Ysgol Bae Baglan and Ysgol Cwm Brombil in recent years but the school has still continued to grow year on year. Similarly Ysgol Tyle’r Ynn is a short distance away from the newly opened Ysgol Carreg Hir and again pupil numbers have risen not declined over 10 years, from 197 in 2010 to 230 in 2020, and there does not appear to be significant movement of pupils from this school to the new build Ysgol Carreg Hir.

Conversely Welsh–medium education is already being provided in a 21st century school build in the Swansea Valley at Ysgol Gymraeg Ystalyfera –Bro Dur, which opened in 2017, a 3-19 Welsh-medium middle school established in place of YGG Y Wern and Ysgol Gymraeg Ystalyfera, across two sites in Ystalyfera and Port Talbot. The primary phase of the school has not yet seen significant growth despite being in a new purpose built 21st Century building.

It is also the case that transition data indicates that pupils continue to transfer from the Welsh-medium primary schools to Cwmtawe Community School rather than to Ysgol Gymraeg Ystalyfera, despite Cwmtawe Community School now being over 20 years old. This would suggest that more than condition of buildings is being considered by parents/carers when decisions are made about which school children will attend.

Neath Port Talbot Council does not have an unlimited source of funding and as such it is inevitable that across the school estate, including all Welsh and English- medium schools, difficult decisions have to be made when identifying which schools are most suitable for inclusion in new build schemes.

It is also the case that where possible the Council will utilise any source of funding available in order to improve and enhance current schools, not just 21st century schools grants for new build projects.

In the last 2 years significant funding has been secured for YGG Pontardawe, with the aim of increasing the school capacity and providing enhanced teaching and learning environments across the school. The £1.62m Welsh Government funded grant has provided the school with remodelled and extended childcare and nursery facilities along with 2 new build Foundation Phase classrooms, a new entrance way, staffroom, meeting room and office and additional play areas, fencing and pathways, along with medium refurbishment of the hall. It will provide space for an additional 60 pupils, and the extended childcare facilities will allow the current provision to increase to 28 places. It is estimated that additional works identified as the scheme has progressed will bring the final cost of the scheme to circa. £2m. The scheme aims to strengthen Welsh-medium education in the Pontardawe area, enabling the school to further develop as a thriving and sustainable provision.”

In the officer response section dealing with “Impact on Welsh Language Development”, the consultation report revealed that the WLIA had been obtained. The timing of the drafting suggests it was the final version completed on 7 May 2021. The WLIA was not referred to in the consultation report as an appendix to it, though, according to the claimant, it appeared on the council’s website at about the same time as the consultation report.

The officer response, clearly drawing on the WLIA, was (with my emphasis):

“A Welsh Language Impact Assessment has been carried out by an independent consultant which recognises concerns that while this proposal relates to the establishment of an English-medium school there may be direct or indirect impacts on the development of the Welsh language in the area.

Alltwen, Godre’rgraig and Llangiwg primaries teach Welsh as a second language and this will also be the case for the new school. Currently 25% of staff across the three schools are fluent or fairly fluent Welsh speakers. By combining the three schools Welsh language provision could be improved as there could be greater impact on having a more concentrated group of Welsh speakers able to support pupil and staff language development skills through the medium of Welsh.

The WESP objective ‘To ensure that second language provision across all schools provides pupils with the skills and ability to become confident and sustained speakers of Welsh and that the provision develops a meaningful relationship between the language and the learner’ is of particular relevance to this proposal.

The impact assessment states ‘A common theme in the consultation comments was that a new English-medium school would negatively impact on the Welsh-medium schools in the locality. However, data/evidence held by the Council indicates that new build English-medium schools in Neath Port Talbot, that have opened in the last 10 years, have had no significant impact on nearby Welsh-medium schools which suggests there is low if any impact on Welsh Medium schools in the locality with the proposal.’

As a result of the consultation and consideration of existing data and evidence the following mitigating actions are recommended for consideration and implementation should the proposal be approved.

•

Longer term monitoring is required to ascertain any impact on wider community Welsh language issues; this could be planned for as part of the updated WESP actions (See Section 5.1)

•

Monitor parental choice of school i.e. Welsh/English medium”.

On 14 June 2021, Ms Elin Maher of the claimant wrote to the council complaining of the tardiness of the WLIA, not obtained in time to form part of the consultation. That would have been of benefit, she argued, “so that the public could respond appropriately with all the documentation available from the outset”. She also complained that the WLIA did not address “how this level of provision will impact on the Welsh language within the community” and that the council “are sending a message to the community that English is the default language” (claimant’s counsel’s translation from the original Welsh).

The council’s cabinet met two days later on 16 June with the WLIA among the documents appended to the officers’ report on the proposals. The report touched briefly on the “integrated impact assessment”, asserting that the proposals would have a positive impact on those with the protected characteristics of age and disability and a neutral impact on those with other protected characteristics.

The WESP and the Welsh language were mentioned: “[w]ith the implementation of mitigating actions the proposal should have at least a neutral impact on the Welsh language …”. Officers informed the cabinet that the appended WLIA “undertaken by an independent consultant … has concluded that the proposal will have a positive impact on people’s opportunities to use the Welsh language and a neutral impact on treating the Welsh and English languages equally”.

Officers recommended in the report that the council should publish a statutory notice under the 2013 Act, triggering the 28 day objection period. The cabinet decided to do so. The statutory notice was published the next day. Objectors were able to see the WLIA and frame their objections with knowledge of its contents, not available during the consultation phase. Of the 297 objectors who made written objections, 35 including the claimant expressed concerns about the impact on the Welsh language.

The Welsh Government was to provide part of the funding for implementing the proposals. The total cost of the project was estimated at about £22.7 million. In August 2021, the Welsh Government commissioned a report on the proposals (the Welsh Government report) from an independent consultancy called Meirion, to examine the following, in the words of the authors:

“• Defining and providing context to the term 'linguistic sensitivity'

• Setting out the principles for safeguarding and promoting language in such an area.

• Consider how these principles could be applied to Pontardawe, within the context of the Swansea Valley proposal.

• Provide options around mitigating actions to reduce negative impacts on the stability and future growth of the Welsh language in the short, medium and long term.”

The Welsh Government report was provided to the council as a “draft report” on 8 October 2021, shortly before cabinet was to meet to make its decision. The “draft” was not updated and is the only and latest version the council has received. The stated purpose of the Welsh Government report was set out in the sub-heading “Additional report to further inform the Welsh language impact assessment”. It was therefore intended to build on the WLIA.

After summarising the proposals, in the first part of the Welsh Government report the authors discussed at some length in broad general terms the sensitivity of languages and the Welsh language in particular; and discussed in general terms the need for strategies to preserve and enhance use of the Welsh language and avert any decline, especially in the Swansea Valley area.

The authors ventured the view (pages 24-31) that “if a new large high status English-medium primary school is established in an area where the Welsh language is in a fragile state … the numbers attending YGG Trebannws and YGG Pontardawe might decline, …”. They then noted that the existing WLIA “does not provide a full assessment of the likely impact of a major development in a linguistically sensitive area”.

With reference to the two monitoring proposals in the WLIA (quoted above), the authors commented that the WLIA “endorses a limited number of vague mitigating initiatives which would have no effect on the current proposal and refers to activity which might take place at some point in the future”; criticised the suggestion that some issues arising from the development could be dealt with through the WESP process; and noted that the WLIA was not commissioned in time to be considered during the consultation process.

The authors noted that YGG Trebannws is the nearest school to the new school site, closer than any of the three schools that would close; there is a path that leads directly from the village to the new school; the car journey takes only minutes; and “YGG Trebannws is not likely to fare well in such a situation, in terms of status planning and of retaining number in Welsh-medium education, apart from increasing the numbers, this is more than slightly problematic”.

The Welsh Government report went on to propose “some mitigating actions”, while noting that from the language perspective none would “compensate for continuing with this proposal as it stands”. To promote a “state-of-the-art” “super” primary school in a “prime location” would “imperil the linguistic nature of this area … of linguistic sensitivity”. Bilingualism “must be an easy choice” if it is to thrive; the proposal “takes away that easy choice”. The last section of the Welsh Government report went on to propose 12 “mitigating actions”.

They were: to reconsider the plan; to develop a “Powerhouse” for the Welsh language; a robust and comprehensive programme of Welsh-medium pre-school provision; developing a dual language stream in the new school; to ensure new leisure facilities (not managed by staff at the new school) are staffed with dual language skills; to establish a new centre for “latecomers to Welsh-medium education”, after age 7; provision of language awareness training; a review of building provision as part of the new WESP; further developing provision at the Bro Dur site of YGG Ystalyfera-Bro Dur; developing a “holistic” plan to promote Welsh-medium education in collaboration with three other local authorities; an enhanced programme for second language Welsh teaching in English-medium education; and an “Action Plan” to promote the Welsh language in the council’s area for the period 2018-2023.

The WLIA was then reissued and published by the council, still bearing the date 16 February 2021 but revised in October 2021 (the revised WLIA). A new section was added to the WLIA referring to the Welsh Government report, quoting its conclusion that “in terms of the language planning principles and processes [discussed earlier in the report], no mitigating actions in the context of the future of the Welsh language in Swansea Valley will compensate for continuing with this proposal as it stands”.

After that, the 12 “mitigating actions” suggested in the Welsh Government report were set out, stating each of them in more detail than I have and adding (in the right hand column, using a tabulated format) the comments of the WLIA authors. The comments argued broadly that the council still believed that the proposals should proceed; some of the mitigating actions were being taken forward through the WESP or had begun to develop; others should be further explored; and one (developing a dual language stream in the new school) was not recommended in the short term.

The council’s officers prepared an undated report on objections (the objection report) at some point in about the second or third week of October 2021. On the subject of Welsh language, among the objections was the criticism that the council had not properly considered the impact on the Welsh language during the consultation:

“Measuring the impact on the Welsh language should have taken place at the first stage of this process, during the formative stages of the proposals and not as a last minute consideration at the end of the process. The language impact study [WLIA] was published 4 months after the close of the original consultation and therefore insufficient time has been given to consider the full impact on the Welsh language.

....

The community was not given an opportunity to comment on the language impact assessment. The council has not discussed the effects with the community”.

In the “officer response” part of the objection report, they sought to rebut the criticisms:

“It is not the case that the Council has not considered the possible impacts of the proposal on opportunities for persons to use the Welsh language, or on treating the Welsh language no less favourably than the English language. In preparation for the consultation a first stage screening assessment was undertaken which identified possible causes for concern. A Welsh language Impact Assessment was then developed by an independent consultant, including comments which were received during the consultation period, and this has formed part of the documents used to support the decision making process. The impact assessment identifies potential impacts, both positive and negative, and also possible mitigating actions.

Further opportunities to comment explicitly on the Welsh Language Impact Assessment have been possible during the objection period. The Welsh Language Impact Assessment has always been a document which is developed as part of the process of school reorganisation. taking into account new information gained through consultation with stakeholders as the proposal progresses, not just in this instance but for all proposals brought forward in Neath Port Talbot. As a result of comments received during this period it was recognised that in some respects the Welsh Language Impact Assessment could benefit from additional information which was not received during the consultation period, and officers have therefore met with Welsh Government representatives to discuss the further development of the Welsh Language Impact Assessment in preparation for the final report.”

The objection report then went on to explain that the Welsh Government had commissioned a report in August 2021 (i.e. the Welsh Government report) and explained the scope of the report. The Welsh Government report (which described itself as a draft) was not provided to cabinet members. The objection report, however, went on to quote from it and to argue in some detail that the Welsh language issues were developing and would continue to inform the proposals and actions to mitigate any adverse impacts (with my emphasis):

“The Welsh Government commissioned report notes the following ‘.. .it should be clearly underlined that, in terms of the language planning principles and processes noted above, no mitigating actions in the context of the future of the Welsh language in the Swansea Valley will compensate for continuing with this proposal as it stands’. lt also notes that ‘In bilingual communities, languages increasingly become a matter of choice. To support bilingualism within these communities, bilingualism must be an easy choice. This proposal takes away that easy choice.’

However, the report does identify a 11 possible mitigations, many of which have already been identified as actions in the draft WESP. These actions, along with officer comments are included in the revised WLIA document.

The Welsh Government's School Organisation Code requires the Council to consult on its proposal and to publish a consultation report summarising any issues raised by consultees, the Council's response to those issues and Estyn’s view of the overall merit of the proposal. The Code does not require a Welsh Language Impact Assessment to be completed when proposals relate to English-medium schools. Consultation has taken place in strict adherence to the Code, from 3rd November 2020 to 19th January 2021.

Objections have been received which state that the community was not given opportunity to comment on the Welsh Language Impact Assessment. The consultation related to the proposal to establish a new school and a Welsh Language Impact Assessment was developed through the consultation to help support Members in the decision making process. The community were invited to make comment on the proposal specifically in relation to its impact on the Welsh language and opportunities to use it.

The Code specifies who should be consulted and all statutory consultees were informed. The consultation was undertaken bilingually, and consultees included the Welsh-medium schools of the Swansea Valley and preschool providers. The community councils of Cilybebyll, Cwmllynfell, Gwaun Cae Gurwen and Ystalyfera along with Pontardawe Town Council were consultees and the consultation document was also sent directly to the office of the Welsh Language Commissioner. Information regarding the consultation was widely shared across the Swansea Valley communities and the proposal was given a great deal of publicity both on social media and in the press.

It is recognised that a number of organisations in the area are concerned that the proposed new school will hinder their work on developing the Welsh language. If the proposal is approved and progresses, the Welsh Language Impact Assessment will continue to be an important document, not just to ensure that any mitigating actions are carried out but to continue to highlight any areas of concern and to further support the planning process. Significant actions will be included in Neath Port Talbot's Welsh in Education Strategic Plan which is expected to be submitted to Welsh Government in January 2022, following an eight week consultation period, providing further opportunities for stakeholders to comment and shape the future development of the language.

It is not the case that the process has not considered the impact that the proposal could have on developing Welsh language skills in the proposed new English-medium school. Suggestions that the proposed new school should be a Welsh-medium school or should consider transitioning from an English-medium to a Welsh-medium provision have also been received. It should be noted that the proposal seeks to replace three current English-medium schools, transferring staff and pupils from existing schools to the new provision, and as a result it has to ensure that the pupils and staff who are displaced from the current schools are able to easily transition into the proposed new school. Changing the language designation of the proposed new school would create further change for the school communities, and is more likely to attract pupils who may otherwise have attended YGG Trebannws, YGG Pontardawe or YG Ystalyfera-Bro Dur (primary phase) for Welsh -medium education, meaning possible change and disruption for these schools also.

The new curriculum for Wales emphasises that language development (in Welsh or English) is based on a continuum or framework of progression. Welsh is a mandatory element meaning that in all schools there is the requirement to teach Welsh to all learners up to 16 years old, and while this is not new (Welsh was included in the national curriculum following the Education Reform Act 1988, and became a compulsory subject for all learners in Wales in Key Stages 1, 2 and 3 in 1990), the 2021 Curriculum and Assessment Act has brought about changes to delivery, removing the current distinction between two programmes of study -Welsh and Welsh second language, and allowing for one continuum of learning Welsh to be taught in all schools in Wales as part of the Languages, Literacy and Communication Area of Learning.

Alltwen, Godre'rgraig and Llangiwg primaries have traditionally taught Welsh as a second language with currently 25% of staff across the three schools being fluent or fairly fluent Welsh speakers. With the requirements of the new curriculum and the additional benefits of having a more concentrated group of Welsh speakers able to support pupil and staff language development skills through the medium of Welsh, it would appear that if the proposal progresses and the school staff are combined, then progress in Welsh language development at the proposed new school could subsequently be improved.”

The director of education, leisure and lifelong learning, Mr Andrew Thomas, provided the officers’ report to cabinet, published on 14 October 2021 (the cabinet report) for its meeting on 20 October 2021. It acknowledged (at paragraph 67) that whilst the proposals directly affected English-medium provision, “there is also a potential effect (that may be positive, negative or neutral) on Welsh-medium schools in the county, which should also be considered”.

Among the appendices were the earlier cabinet reports for the meetings in October 2020 and June 2021; the IIA going back to October 2020; the objection report; and the revised WLIA, but not the (draft) Welsh Government report; though the latter was quoted from again, in similar vein to the quotes in the objections report, which was itself one of the appendices.

The cabinet met on 20 October 2021. It decided to adopt the recommendation of the director and to proceed with the proposal to close the three schools on 31 August 2024 and to open the new school on 1 September 2024. It is that decision which is challenged in this judicial review. The council then pressed ahead with the preliminary work towards implementation, issuing “mini-tender” documentation on 3 December 2021.

The present challenge was foreshadowed in a pre-action protocol letter dated 22 December 2021, relying on the same grounds of challenge as argued before me. The council responded on 7 January 2022, rejecting the challenge on the ground that it was without merit and because any claim would be out of time. In opposing the first ground, the council advanced the terse proposition: “[t]he three primary schools to close, and the new primary school to open, are English-medium schools. The above part of the Code does not apply to such proposals.”

I interject that in written argument before me, Mr Oldham submitted that “if ‘affected’ means ‘the subject of the statutory proposals’, then no WM school was ‘affected’. The closing schools and the opening school are all English medium”. But in oral argument, he did not support and endorse the formulation in the objection report (“[t]he Code does not require a Welsh Language Impact Assessment to be completed when proposals relate to English-medium schools”). The rest of his argument sensibly proceeded on the more realistic footing that the class of “affected” schools is wider than those “involved”.

The present claim was brought on 18 January 2022, just short of three months after the cabinet’s decision. Mini-tender returns were received on 11 February 2022, providing a timescale for completion in September to December 2024, with construction to start in April 2023. By early February 2022, the council had spent about £65,000 to £70,000 on surveys, assessments and investigations. Increased costs in the tens of thousands were being spent each quarter due to housing pupils of the three closing schools in temporary buildings.

The council filed its summary grounds of resistance on 7 February 2022. It maintained the primary contention that the schools “involved or affected” within paragraph 3.4 of the Code are those named in the proposal, i.e. the three schools and the new schools. Since they are or will be all English-medium schools, the part of paragraph 3.4 of the Code relied on by the claimant did not apply; it only applied where a Welsh-medium school is “involved or affected”.

Alternatively, the council contended that even if that construction was wrong and Welsh-medium schools could be “involved or affected” by the proposals, the council “believed that there were no such schools” and thus there was no obligation to include a WLIA within the consultation. The council was entitled to decide that no Welsh-medium schools were affected and did so.

The council further contended that the processes had all been properly and lawfully undertaken; there was no impropriety in the process; no obligation to reopen consultation either after the WLIA was obtained nor after the Welsh Government had commented on it in the Welsh Government report.

Further, the council contended, even if the conduct complained of had occurred, it was highly likely the outcome for the claimant would not have been substantially different. In addition, the claim had not been brought promptly; it included, among other things, a complaint about the consultation undertaken 14 months earlier. The grant of relief would cause substantial hardship to parents and others, after prolonged uncertainty.

The matter came before Steyn J on the papers. On 23 March 2022, she granted permission to apply for judicial review, commenting that the grounds are arguable. The council was not shut out from arguing delay points at the main hearing, but permission would not be refused on delay grounds alone; the proposals were not due to be implemented until September 2024, a lead time notably longer than in some other school closure cases.

In detailed grounds of resistance filed on or about 25 April 2022, the council relied on a further witness statement of Ms Crowhurst, her third, to support its contention that the council had lawfully formed the view, and stood by the view, that no Welsh-medium school was affected by the proposals, even if, contrary to the council’s primary contention, it were in principle possible for the Code to apply where the proposals only related to English-medium schools.

The rest of the detailed grounds broadly followed the same pattern as the summary grounds, as well as adding an assertion that the claimant has no standing to bring the claim; a contention not pursued before me (the claimant having by then obtained a costs capping order).

In her third witness statement made on 25 April 2022, Ms Crowhurst sought to address the position on the footing that the Code could, after all, apply where the proposals were to close or open only English-medium schools. I infer that the council had begun to accept that this may be so, contrary to its earlier understanding. She explained that she had drafted the consultation document with help from colleagues.

Ms Crowhurst said that in considering whether any schools were “involved or affected” she had been guided by the bullet points in paragraph 3.4 of the Code under the sub-heading “Details of affected schools”. She explained that she asked herself, whether the proposal would “directly affect the pupil roll of the school …” or “directly affect the capacity at the school”. These bullet points are the provisions determining what information about schools should be provided, but she maintained that they were relevant to what schools if any were “affected” for the purpose of the WLIA requirement later in paragraph 3.4.

Ms Crowhurst then added evidence in tabulated form not drawn up at the time but “which is for explanatory purposes in this claim”. In the table she posed the questions corresponding to the bullet points that guided her; most importantly for present purposes, “[w]ill the proposal directly affect the pupil roll of the school at the time of the proposed implementation and the 5 year projected pupil numbers?” The table states “unlikely” as the answer, supplemented by further reasoning, in the case of both YGG Trebannws and YGG Pontardawe.

Ms Crowhurst went on to produce detailed statistics by way of explaining that in the past loss of pupils had not materialised in the case of other school reorganisations, by way of comparison. This was a point made in some of the documents at the time, but the reasoning in the third witness statement supplements the reasoning deployed during and after the consultation process.

Ms Crowhurst said council officers had found no precedent for “an example of an EM school reorganisation in Wales which involved a consultation document with a WLIA”, despite internet searches. (In argument, the claimant disputed that this was correct and claimed to have found such a case, possibly postdating the consultation exercise in this case.) No browsing history documents evidencing these internet searches were produced. There is no evidence of when the searches were carried out, nor what search terms were used.

Ms Crowhurst went on to explain that the decision to commission the WLIA was for reasons unrelated to the terms of the Code. It is obvious from the correspondence at the time that the Welsh Government was insisting on a WLIA. As the proposed joint funder of the project, it was in a position to insist on one. Ms Crowhurst’s third witness statement says also that the WLIA was commissioned because it was recognised that the three schools and the new school (all English-medium) were “located in an area of linguistic sensitivity”.

That did not mean, Ms Crowhurst insists in her third witness statement, that the council thought there would be impacts on Welsh-medium schools or education. She did not accept much of the content of the Welsh Government report; for example, she deprecated the use of a term such as “super” school to describe the new school. The revised WLIA referring to new proposed “mitigating actions” was, however, put before cabinet; it was appropriate that cabinet should be aware of the Welsh Government’s thinking about the proposals.

The contractor and appraisal appointment for the purpose of constructing the new school was due to take place at about the time of Ms Crowhurst’s third witness statement, 25 April 2022. However, following elections in early May 2022, the political composition of elected members of the council changed from a Labour majority administration to one of no overall control. A new administration was formed on 7 June 2022.

Under relevant contract procedure rules (the value of the proposed contract exceeding £250,000) the relevant cabinet member of the new administration had to be consulted and approve the proposal before any contract for construction of the new school could be signed. So far, only the first stage of the two-stage contract has been authorised. That was on 24 June 2022.

On 27 June 2022, with the court’s permission, Ms Crowhurst explained these developments in a fourth witness statement. She also gave some further explanations in that witness statement about the evolution of the WLIA and the process of giving of certain disclosure to the claimant’s solicitors, which I have in part touched on above and need not set out further. In addition, she explained that the council had decided to consider whether to reopen the decision to close the three schools and open the new school.

On 29 June 2022, the council wrote to the Minister for Education and Welsh Language explaining the new administration’s wish to review the decision and seeking a meeting with him to establish what information the Welsh Ministers would require in order to consider giving “consent” to the review. It has since been clarified by the council that the Welsh Ministers’ consent is not legally required for any review, but the council wishes in any case to engage with the Welsh Government in the course of reviewing the decision.

A website announcement from the council on 30 June 2022 explained that the new administration “wishes to review the decision” and see whether “alternative ways” of bringing 21^st century standards to the Swansea Valley “can be achieved which would be more acceptable to the community”. The council would “now start discussions with Welsh Government Ministers to establish what information they might require from the council”. That would “inform the consultation process which the council will undertake with stakeholders”.

That was the position when the case was ably argued before me in Cardiff by leading counsel in mid-July 2022. At my request the parties agreed they would keep the court updated, letting me know of any relevant developments while judgment was reserved. I received a helpful update in the first half of September 2022. The council had written to the Minister again on 26 July 2022. The Minister responded on 28 July. No meeting had taken place by mid-September.

The current envisaged timescale is as follows. A meeting with the Welsh Government is expected to take place this autumn. A report to members will be presented to inform their decision whether to embark on the formal process that could lead to a decision not to implement the proposals. It is considered likely that the final determination whether or not to implement the proposals will be in about April 2023.

In line with current judicial thinking, I will summarise the parties’ main contentions only in briefest outline, although I have considered them in the much greater detail with which they were skilfully and eloquently presented to me in writing and orally.

The claimant says not including a WLIA within the consultation was wrong in law. Either the decision not to do so was based on a misinterpretation of paragraph 3.4 of the Code; or, there was a failure to consider whether any Welsh-medium school was “affected” by the proposals; or it was irrational not to commission a WLIA while accepting in the consultation document that the proposals could have an impact on opportunities to use the Welsh language.

Mr Gwion Lewis KC (as he now is) submitted that YGG Pontardawe and YGG Trebannws were Welsh-medium schools that were plainly “affected” by the proposal. It was a misdirection or irrational to conclude otherwise. The misdirection was adopted in the objection report which included the erroneous proposition: “[t]he Code does not require a Welsh Language Impact Assessment to be completed when proposals relate to English-medium schools.”

It was therefore no surprise, Mr Lewis submitted, that the WLIA, when it came, confirmed the potential effect on, at least, YGG Pontadarwe and YGG Trebannws. They could, as pointed out in the WLIA, lose pupils to the attractive new school. That assessment came too late; the consultation had already closed. Yet, paragraph 3.4 used the word “must”, making a WLIA mandatory at the consultation stage if any “affected” school provides teaching through the medium of Welsh.

Mr Lewis submitted that the court should reject as inadmissible or treat with scepticism and caution Ms Crowhurst’s ex post facto evidence, not given until 25 April 2022, of having considered at the time, without making any record, that the proposals would be unlikely to affect pupil numbers at the two Welsh-medium primary schools. That evidence was not given to elucidate the factual position but to “try to fill a complete vacuum in the evidence” in relation to Ms Crowhurst’s consideration of the issue.

He referred me to well known authority supporting that approach; most recently R. (United Trade Action Group Ltd) v. Transport for London [2021] EWCA Civ 1197, judgment of the court (Bean LJ, Sir Keith Lindblom, Senior President of Tribunals and Sir Stephen Irwin) at [125]. While the court could decide to admit the ex post facto evidence, it should be taken with a pinch of salt, viewed as self-serving and contrary to the reasoning in the contemporary documents, including the WLIA itself and especially the IIA appended to the October 2020 report.

Mr Lewis referred to Ms Crowhurst’s editing of the WLIA. He said she sought in her drafting contributions and exchanges with Ms Morgan to play down and minimise the impact on Welsh-medium schools, opportunities to use the Welsh language and equality of treatment as between the Welsh and English languages. Her editing and drafting contributions stood uneasily with her ex post facto witness evidence, he submitted; but her later witness evidence remained contradicted by the WLIA despite her contributions to it.

For the council, Mr Peter Oldham KC (as he now is) accepted that where a local authority concludes that a Welsh-medium school is affected by a proposal, the authority must include a WLIA in the consultation on the proposal. But he submitted that the council had made an unassailable finding to the contrary. It was not enough that a school might be affected; the question was whether it would be.

Mr Oldham submitted that it was a Welsh-medium school that had to be affected, not the Welsh language. These were two different things. He said that if “affected” means “the subject of the statutory proposals”, then no Welsh-medium school is affected. If it bore a wider meaning, it was for the council to determine which Welsh medium schools, if any, were affected. Ms Crowhurst’s explanatory evidence showed she had rationally concluded that no Welsh-medium school was “affected”.

The council submitted that Ms Crowhurst’s explanatory evidence was admissible, relevant and probative. The case was not one where reasons were given or required to be given at the time of the decision. Ms Crowhurst’s explanation of her thinking must be taken at face value; there is no good reason to dismiss it or disbelieve it.

In written and oral argument, Mr Oldham went through the documents in detail, starting with the consultation document. They did not show that any Welsh-medium school was affected by the proposals, he said. The view that they might be was not the council’s but that of others, such as consultees, and was speculative. The materials relied on post-dated the consultation document and were produced for purposes other than compliance with the requirement in the Code for a consultation to include a WLIA.

As for the IIA, it evolved over time, submitted Mr Oldham. It was not intended to address, and did not address, the question whether the proposals would affect Welsh-medium schools. Ms Crowhurst approached that question by reference to the bullet points in the earlier part of paragraph 3.4, referring to schools “likely to be affected by the proposals” for the different purpose of ascertaining what information needed to be provided in the consultation document.

Many of the passages in documents relied on by the claimant in fact addressed the impact of the proposals on the Welsh language, not on Welsh-medium schools. The fact that a WLIA was produced later, for purposes separate from the Code, was not probative of the council’s view at the time of the consultation. The arguments of the claimant sets the bar too low, speaking of Welsh-medium schools being “potentially” affected or saying they “could” be affected.

Nor does the later discussion, post-dating the consultation document, about the need for “mitigating” measures imply recognition of an effect on Welsh-medium schools, the council submitted. Longer term monitoring through the WESP and monitoring of parental choice was proposed in the WLIA not because Welsh-medium schools would be affected but so that “future trends could be checked to see if such impacts might arise”.

The claimant’s forensic points on the evolution of the draft WLIA do not assist the claimant, Mr Oldham submitted. For instance, the removal from the 26 February 2021 draft WLIA of the words “[t]hese proposals may therefore have an impact on numbers attending Welsh medium education” were replaced in the 8 April 2021 version with more specific and detailed language expressing the view that parents might prefer their local Welsh-medium school.

Turning to my reasoning and conclusions on this issue, I start with the interpretation of the Code and, in particular, paragraph 3.4. That is a matter for the court. The Code is a normative document, creating obligations given statutory force by the 2013 Act; but it is not itself a statute. It is intended to be operated by education professionals rather than by lawyers. It should be construed in a common sense and not unduly legalistic way.

I accept Mr Lewis’ submission that in the phrase “involved or affected”, the word “involved” refers to the schools that are directly the subject of the proposal in question, i.e. in this case the schools that are to close or to open. The wider class of schools “affected” refers to schools that are not directly “involved”, i.e. they are not going to close or open, but which the proposals affect, i.e. on which the proposals have an effect.

Whether a particular school is “affected” within paragraph 3.4 is primarily an evaluative factual judgment to be made by the council. It is not for the court to interfere with that judgment unless the council misdirects itself or otherwise errs on established public law principles. The council must, however, direct itself correctly in accordance with the true meaning of “affected”, which is for the court to determine, objectively.

I therefore consider next the scope of that word, in its context of identifying when a WLIA must be obtained while consulting upon a reorganisation proposal. To determine whether a particular schools is “affected” by a proposal requires consideration of, at least, the three matters identified by the Court of Appeal in R (Driver) v. Rhondda Cynon Taf County Borough Council [2020] EWCA Civ 1759, in the judgment of the court of Sir Geoffrey Vos, then Chancellor of the High Court, giving the judgment of the court at [60]:

“the effect on existing pupils, the availability of places to meet demand for Welsh medium primary education and the likelihood of parents in future opting for a closer English medium primary schools for their children rather than sending them to the proposed new Welsh medium school”.

The Code should not, in my judgment, be read as importing nice distinctions of the kind beloved by lawyers between, for example, being “affected” and being “likely to be affected” (a phrase used in parts of the Code dealing with provision of information about a proposal, and picked up by Ms Crowhurst in her deliberations). In my judgment, the common sense interpretation is that a school is “affected” by a proposal if it may undergo material change because of the proposal and therefore needs to address that potential for change.

A real possibility that the school may undergo material change is enough for the school to be “affected” because if the possibility exists, the school management must deal with it. The need to address the possible impact of a proposal is an effect on the school. The likelihood of change (such as losing pupils to a new school or changes to the demographic profile of the pupils) does not have to be a certainty. It does not have to be a very strong likelihood. A real possibility is enough because the school must address the possibility. The school cannot ignore the possibility even if, as it turns out, the change does not materialise.

It follows that so-called “mitigation” measures such as monitoring future trends (which is not truly a mitigation measure at all) are unlikely to form a proper basis for a judgment that a particular school is not “affected”. If the local authority needs to monitor future parental choice, so does the school. The need to monitor arises because there is a perceived possibility of losing pupils to the new school. That is an effect on the school which is thereby “affected”.

Real mitigation measures, such as proposals for investment in a Welsh-medium school to help retain pupils, point towards the school being “affected” by the proposal, not to the contrary conclusion. It is a wrong approach to paragraph 3.4 to determine that a school is not “affected” because mitigation measures are likely to avert any loss of pupils. Even if the measures are seductively attractive to parents at the school in question, the management of the school must still address the effects on it of the proposal to open a new school elsewhere.

The effect on the school in question is apparent in such a case because a strategy for communicating with parents and answering their queries will be needed. Budgets may have to be redrawn in case parents remove their children; priorities may have to be reassessed; recruitment decisions may be affected; class sizes may have to be revisited; and so forth. If new investment is available to mitigate the effects, the school must address the practicalities of the investment project; perhaps by deciding the configuration of new buildings, and so forth.

It is clear from the evidence that the council adopted a higher test, that of a likelihood or strong likelihood that the local Welsh-medium schools would lose pupils. Standing back from the detail, the gist of Ms Crowhurst’s account – both contemporary and ex post facto – is that she did not consider the two local Welsh-medium schools were “affected” because she did not think they needed to worry unduly about losing pupils to the new school and, even if they might lose some in the future, that issue could be dealt with through the WESP.

That is not the correct interpretation of paragraph 3.4 of the Code. It sets the bar too high, using a “strong possibility” or “strong likelihood” test which is different and narrower than the correct approach, which I have outlined. A real possibility which is more than fanciful is enough.

Ms Crowhurst was, moreover, under the misapprehension until after the claim was brought that paragraph 3.4 of the Code did not apply at all because these proposals did not involve opening or closing any Welsh-medium schools. Her witness statements do not explain how this influenced her thinking. Based on her mistaken understanding of paragraph 3.4 in 2020 and 2021, the evaluative judgment exercise she was at such pains to explain in her ex post facto evidence did not arise and there was no need for her to undertake it. I infer that this may explain the absence of any documentary audit trail.

But even accepting that she made the evaluative judgment later explained in her ex post facto explanatory evidence, still the test she applied was too high, as I have explained. If she had directed her mind to the immediate concern of those managing YGG Trebannws and YGG Pontardawe that a new English-medium “super school” (even if that term is objectively unjustified) might lure pupils away from them, she would have appreciated the need for those two schools to address that possibility and that they were therefore “affected” by the proposals.

The clear common sense finding made in the IIA in October 2020 was that “[a] new build EM school could attract pupils currently attending WM schools”. That finding was decisive of the applicability of the mandatory requirement for a WLIA, if the correct interpretation of paragraph 3.4 of the Code is adopted. If the new school could attract pupils from the local Welsh-medium schools, then manifestly the latter were “affected” by the proposal to open that new school, even if it later turns out that they do not lose a single pupil.

It is clear from the defensively drafted documents subsequent to the October 2020 report (from which I have quoted above, passim) that the council was concerned to allay fears that, in the linguistically sensitive Swansea Valley area, a parental preference for English-medium education would increase, with a detrimental impact on Welsh-medium educated pupil numbers and an adverse impact on opportunities to use the Welsh language.

These were the very concerns highlighted in the IIA in October 2020, which ought to have led to a WLIA being included in the consultation. They are essentially the same concerns that emerged from the WLIA, the Welsh Government report and the revised WLIA. They were played down in the consultation document. They were met with counter-arguments in the WLIA, the consultation report and the objections report.

I conclude that the council misdirected itself in law in interpreting and applying paragraph 3.4 of the Code. On its own findings in October 2020, a WLIA was mandatory. The council believed otherwise at the time because it thought paragraph 3.4 did not apply at all. The result was that the WLIA was not obtained until after the proposals had ceased to be at a formative stage. The first ground of challenge therefore succeeds.

It becomes unnecessary to consider further the claimant’s other ways of advancing the first ground: that the council failed to consider whether any Welsh-medium school was “affected” by the proposals; or that it was irrational not to commission a WLIA while accepting in the consultation document that the proposals could have an impact on opportunities to use the Welsh language.

For completeness, I should add that I do not see much force in the council’s point that the requirement was for a Welsh-medium school to be affected and not that use of the Welsh language must be affected. That is true, but the two concepts are very closely interlinked and, in practice, the former will often entail the latter or vice versa. The Welshness of a school’s educational provision is defined by the language in which it educates its pupils.

The claimant’s second ground of challenge is that when the WLIA did become available, the council should have reopened the consultation and not left comment on its content to be formulated as an objection, during the objections period.

The proposals, said Mr Lewis, had by then moved beyond the formative stage; they were approved by cabinet in June 2021, subject to the objections process. By not reopening the consultation once the WLIA had become available, the council deprived consultees of the opportunity to give intelligent consideration to the proposals and an intelligent response to them while they were still at a formative stage.

Mr Oldham countered that there was no obligation at common law or under the Code to reopen the consultation. The advent of the WLIA, created for other purposes, did not lead to the proposals being significantly re-cast, transforming the position to such an extent that the original consultation was superseded and invalidated. The WLIA did not say that the proposals would have an impact on Welsh-medium schools.

I can deal with this second ground of challenge more briefly. If I had found that a WLIA was not mandatory at the stage of undertaking the consultation, I would not have found that the WLIA, once obtained by the council at the behest of the Welsh Government, generated a legal obligation to reopen the consultation. I accept Mr Oldham’s submission that there would be no obligation to do so.

The WLIA was either mandatory ab initio, as I have found, or it was not mandatory at all. In the present context, the scope of the obligation to consult is delineated by the Code. There is no room for the common law to supplement the obligation to consult as defined in the Code.

The claimant’s case for a mandatory WLIA rests firmly on paragraph 3.4. As such, it succeeds. If that is wrong, it cannot be made good by the common law. The second ground of challenge fails.

The claimant’s third ground of challenge is that it was incumbent on the council to consult on the content of the Welsh Government report, once it was received in October 2021. It was not sufficient to provide only a brief summary of it in the objection report, not disclosing even to cabinet its full content.

Mr Lewis submitted that the Welsh Government’s decision to intervene in the process at the eleventh hour was wholly exceptional and confirmed that consultees had not had enough information about the Welsh language impact of the proposals. The only lawful course, on receipt of the Welsh Government report, was as a matter of basic fairness to revoke the statutory notice and reopen the consultation.

Mr Oldham submitted that the same reasoning applied as deployed in opposition to the second ground of challenge. The Welsh Government report, received as a draft and never confirmed as final, no more obliged the council to reopen the consultation than did the WLIA. Furthermore, the council did not have ownership of the Welsh Government report and had already taken the decision to publish its statutory notice before the report was received as a draft.

Nor was there any obligation to disclose the Welsh Government report in full, either to the consultees, the cabinet or anyone else. It was not the council’s document and was considered at the time to be a draft, although it later became evident that no final version was likely to be produced. The content of the report comprised mostly generalities. The council did not accept its unreasoned assertion that “no gesture … would be sufficient to mitigate the damage which could be caused by this new development”.

In my judgment, this ground of challenge fares no better than the previous one. I agree with Mr Oldham’s submission that the Welsh Government report did not change the position. It came very late in the process, after the publication of statutory proposals. It was described at the time as a draft and was not obviously given to the council for public consumption.

The Welsh Government’s education department was itself a consultee in the earlier consultation. I have not seen any response from it within the compass of the consultation, but its late intervention in the form of the Welsh Government report was not capable of generating a fresh common law obligation to consult, over and above the obligations to consult found in the Code.

I do not accept the claimant’s suggestion that the Welsh Government report so transformed the landscape that the previously undertaken consultation was wholly superseded and had to be replayed. The proposals remained exactly the same. The arguments for and against remained essentially the same. They were expressed particularly trenchantly in the Welsh Government report, but the substance did not break new ground, so much as come down firmly on one side of the argument rather than the other.

The council said it was highly likely the outcome would not have been substantially different for the claimant, had the conduct complained of not occurred. The conduct complained of was the omission to include a WLIA in the consultation. The outcome was, in Mr Oldham’s submission, the decision to approve the proposals. That is the decision challenged. The later decision of the new administration to reconsider whether to implement them was, he said, irrelevant. The current state of uncertainty is not part of the outcome.

Mr Oldham argued that there was no basis for supposing that the views expressed in the consultation would have been any different if consultees had received a WLIA. The claimant itself did not respond to the consultation at all. The council would have decided in October 2021 to implement the proposals even if the consultation materials had included a WLIA.

For the claimant, Mr Lewis directed me to the helpful review of cases on section 31(2A) in the judgment of Ms Kate Grange QC (as she then was) sitting as a deputy High Court judge in R (Cava Bien Ltd) v. Milton Keynes Council [2021] EWHC 3003 (Admin) and the 14 propositions she drew from the cases, at [52], which I do not repeat here. He submitted that the high threshold was not reached because the court could not say how consultees would have responded if the required WLIA had been included in the consultation materials.

Mr Lewis submitted that the requirement for a WLIA in the Code would be pointless if local authorities could ignore the requirement, commission a late WLIA after the consultation and then argue that the omission to produce one earlier did not affect the outcome. The Welsh Government clearly attached importance to obtaining a WLIA, or it would not have included the requirement for one in the Code.

Cabinet had made its decision with a late WLIA on which the public had not had an opportunity to comment properly, Mr Lewis submitted. An earlier WLIA would have prompted the Welsh Government, itself a consultee, to react strongly and influentially much earlier than it did, in the manner it did when it commissioned the Welsh Government report including the view that “no mitigating actions in the context of the future of the Welsh language in Swansea Valley will compensate for continuing with this proposal as it stands”.

Mr Lewis submitted further, after the hearing, that the decision of the subsequently elected administration to reconsider whether to implement the proposals further weakened and indeed rendered implausible the council’s argument that it was “highly likely” the outcome would have been the same with or without the conduct complained of, namely the exclusion of the WLIA from the consultation process.

In my judgment, the “outcome” is not to be viewed narrowly as confined to the decision under challenge, artificially ignoring the wider “outcome” which includes what actually happens, in the aftermath of the decision challenged. I cannot find in the statutory language any warrant for the council’s artificially narrow approach to what the outcome is, when measuring that against what it would have been had the conduct complained of not occurred.

I find it difficult to see how the council can say that it is highly likely the outcome would have been the same had the conduct complained of not occurred, when the council admits that it does not itself know what the final outcome of the exercise will turn out to be. In written submissions after the hearing the council explained that the court’s decision in this challenge would not be academic. The council would benefit from knowing whether the claim succeeded or failed because:

“If the claim fails, and

(1)

the Defendant decides not to proceed with a review of the Decision, the Defendant will know that it may proceed to implement the Decision; or

(2)

the Defendant decides to proceed with a review, the Court’s decision will help to determine the steps which the review procedure should, and should not, encompass…. .

If the claim succeeds, and

(1)

the Defendant decides not to proceed with a review of the Decision, the Defendant will know what if any steps it will have to undertake before proceeding; or

(2)

the Defendant decides to proceed with a review, the Court’s decision will likewise help to determine the steps which the review procedure should encompass.”

If the court’s decision has potential relevance to the eventual outcome, as the council there asserts, it is difficult to see how the conduct complained of can have made no difference. For the outcome to be the same, the council would have to have a closed mind in response to any fresh consultation exercise that may be required. That seems to me to be an end to the council’s argument.

The Welsh Government’s position raises another difficulty in the way of the council’s proposition that the outcome for the claimant would have been highly likely to have been the same had the conduct complained of not occurred. If a WLIA had formed part of the consultation exercise, it is not unreasonable to suppose that it might have quite closely resembled the actual WLIA later produced at the request of the Welsh Government (not the revised WLIA but the original version prepared by Ms Morgan and Ms Crowhurst).

The response of the latter to that WLIA was not positive. It led to the revised WLIA and the commissioning of the Welsh Government report which commented negatively on the proposals. There is a good chance that, had a WLIA been available earlier in the process, at the consultation stage, the Welsh Government would have made clear, perhaps in the same trenchant terms later used in the Welsh Government report, its strong reservations about the project.

That could well, in my judgment, have impelled cabinet members to vote against the proposals; particularly as the Welsh Government holds the purse strings and the new school project probably cannot proceed without its agreement to provide its share of the funding. Indeed, the project may now still not go ahead even if the council were to win this case, as I have already noted.

For those brief reasons, I am unable to say that it is highly likely the outcome would not have been substantially different for the claimant if the conduct complained of had not occurred. I do not know what the outcome would have been if that conduct had not occurred. I do not, indeed, yet know what the outcome will be and on the evidence before the court, nor does anyone else.

The council submitted that the claim was not made promptly. Mr Oldham relied on cases in which courts have emphasised the need for school reorganisation challenges to be brought quickly and early in order not to prolong a state of uncertainty among parents, school children and the education community locally; and so that effort and expense are saved and not unnecessarily wasted.

Here, Mr Oldham said the challenge ought to have attacked the alleged flaw in the consultation exercise, promptly after the claimant became aware that the council did not intend to include a WLIA within the consultation exercise, which began as long ago as January 2021. A timely challenge then would have avoided all the unnecessary expense and effort that went into airing the proposals after the consultation exercise. The claimant ought not to have awaited the decision of the cabinet in October 2021.

The relevant cases cited on delay and prejudice in the particular context of school reorganisations are R v. Secretary of State for Education and Science ex p. Threapleton, (12 May 1988, reported in the Times, 2 June 1988 (Divisional Court); in Nichol v. Gateshead MBC (1988) 87 LGR 435; R. v. Leeds City Council ex p. N [1999] ELR 324; R (Bandtock) v. Secretary of State for Education [2001] ELR 333; and R. (Jones) v. Denbighshire County Council [2016] EWHC 2074 (Admin).

Mr Oldham reminded me of the discussion in Collins J’s judgment on the permission application in Bandtock, noting the inconsistency of approach in the judgment of Woolf LJ (as he then was) in ex p. Threapleton and the Court of Appeal’s decision in Nichols v. Gateshead MBC. Collins J commended the approach of Mr Richard McManus QC in his book, Education and the Courts (Jordans, 1998) at paragraph 1-32. Mr McManus noted that the question when is the appropriate time to bring a judicial review was “controversial”.

According to Mr McManus, the “better view”, albeit not fully supported by the authorities, is that where a decision is fatally and incurably flawed ab initio, the challenge should be brought early on, to expose that flaw. Only where the defect can be cured should the challenger refrain from going to court and instead draw the defect to the decision maker (then, in England, the Secretary of State). He placed within the first category a case where “the statutory consultation had not on its face complied with the statutory requirements”.

In relation to prejudice and section 31(6) of the 1981 Act, Mr Oldham submitted that there had been substantial delay and that granting relief would cause hardship to and substantially prejudice the rights of pupils, families, school governors and staff as well as the council itself and would be detrimental to good administration. Mr Oldham relied on the steps taken towards implementing the proposals, including providing information to parents and commissioning some contractor work.

Delay in implementing the proposals would be detrimental, Mr Oldham submitted. The new school opening could be delayed beyond the planned opening date, 1 September 2024. If that happened, the impact would be particularly hard on pupils at Godre’r Graig school, who are in temporary buildings; and would delay required provision of more, and more suitable, places for pupils with autism.

The uncertainty caused by the new administration’s intention to reconsider whether to go ahead with the proposals was irrelevant, submitted Mr Oldham in post-hearing written submissions. The issue of delay “has to be judged by the [c]laimant’s actions”.

For the claimant, Mr Lewis submitted that the challenge was brought within three months of the cabinet’s decision. There was no undue delay during that period. The pre-action protocol process had to be gone through; documents had to be translated from Welsh to English; the three month period included the Christmas holiday period; the solicitor and counsel had to be fluent in Welsh.

The cabinet’s decision was the correct target for the challenge, not the earlier decision to consult without a WLIA, Mr Lewis said. The facts here were different from those in R. v. Leeds City Council ex p. N, relied on by the council. Here, there was no educational deficit; the three schools are not failing and the education they provide has not been found inadequate, as was the position in the Leeds case. When the challenge was brought, the three schools were continuing to take new pupils to start in the academic year 2022-2023.

An earlier challenge would have been regarded as premature, as in the Denbighshire case, (see at [90]-[91] in the judgment of Hickinbottom J (as he then was) giving the judgment of the Divisional Court). The issue raised is of national significance, concerning the correct interpretation of provisions of the Code relating to the Welsh language.

Furthermore, there is no prejudice to good administration, Mr Lewis submitted. The council has not yet started work in earnest on implementing the proposals. The current state of uncertainty is of the council’s own making and has nothing to do with any delay arising from the current legal challenge. The council has halted work towards construction and opening of the new school and is approaching the possible forthcoming review without any sense of urgency and without any fixed timetable.

I find the state of the authorities perplexing. They do not speak with one voice. The search for a coherent principle running through them is vain. The advantages of an early challenge are counterbalanced by the disadvantages of encouraging litigation. The pre-action protocol and cost capping have become a feature of this kind of litigation since the older cases mentioned above were decided, against the backdrop of differing statutory schemes.

The facts may also vary widely. I do not think there are special rules for school reorganisation cases. Reorganisation of health care cases and other cases involving consultation raise similar timing issues. Time starts to run when the “the grounds to make the claim first arose” (CPR rule 54.5(1)(b)). If a claim is premature, as defendants to judicial review claims often assert, it is at best doubtful whether grounds for making it have arisen.

Sometimes there may be a particular need for speed, as in the Leeds case where action to safeguard the quality of children’s education needed to be taken quickly. Often (but not always) the later the challenge is brought, the longer the period of uncertainty caused by the litigation. In other cases, the urgency may be less and the proper target decision may be a later one in the course of events.

It is well known among practitioners that a claimant may be reproached with prematurity if they claim early and with tardiness if they claim late. The tension between the two approaches is shown in the cases. There is no obvious temperate zone in between the Scylla of prematurity and the Charybdis of laches. Some claimants seek to avoid litigation. Others, with deeper pockets or of a more litigious disposition, claim early. Some may claim early and then seek a stay, but the defendant may contest a stay and the court may not grant it.

In the present case, I have come to the conclusion that I should not refuse relief on the grounds of delay and prejudice, for the following brief reasons. First, an early challenge to the consultation process would probably have been met with an allegation of prematurity. The council would have had available to it the Divisional Court’s decision in ex p. Threapleton and, closer to home, its decision in the Denbighshire case.

The council could have deployed the converse arguments to those it now advances. It could have contended that the claimant had “rushed off to court” (in the familiar phrase): the challenge would be academic unless cabinet later decided to adopt the proposals; if they were published, objectors might then yet win the counter-argument and defeat them; Welsh Government funding was indispensable and far from guaranteed; and elections were due over two years before the distant implementation date (1 September 2024).

Next, the claimant is a respected charitable organisation without substantial financial resources. It was appropriate for it to be cautious about litigating. It needed and later obtained a costs capping order. It made known its objections to the absence of a WLIA during the process. The council recorded in the consultation report the objection from some consultees that the absence of a WLIA prevented them from giving intelligent consideration to the proposals.

The council would have known from the Denbighshire and ex p. Threapleton decisions that the passing of three months from the start of the consultation did not immunise the consultation process completely from challenge. It decided to press ahead with publishing the proposals. If the claimant had sought to challenge the decision to publish them, taken in June 2021, it could have been met with the riposte that its timing was “the worst of all worlds”: too late for the consultation, too early for the final decision yet some way off.

Furthermore, at that stage, the proposals were still beset with uncertainty. Litigation might well be unnecessary, from the perspective of the claimant at the time. The Welsh Government, which held the purse strings, had insisted on a WLIA; its content was known to the claimant when Ms Elin Maher wrote to the council on 14 June 2021, complaining that it had come too late. Its status and likely impact was unclear. The council interpreted it benevolently in the consultation report but it was far from clear that the proposals would proceed and the opening of the new school was still over 3½ years away.

I accept that the claimant did not delay unduly in bringing its challenge to the October 2021 cabinet decision, for the reasons Mr Lewis gives. The claim was prepared largely bilingually and over the Christmas period, and brought within three months of the decision challenged.

There are no clear rules for determining the right time to bring a challenge. It is a question of fact and degree in each case. Practitioners know that defendant decision makers are likely to criticise the timing of a challenge whenever it is brought. I do not think that on the facts here, the claimant should be shut out because it chose to refrain from litigation in early and mid-2021. As Steyn J noted, the implementation dates in August and September 2024 are more distant than in many cases of this kind.

As for prejudice, I accept that some money has been spent. The sums involved are relatively modest because of the decision not to proceed yet with the second stage of implementing the design and build plans. If the project proceeds, much of that money will have had to be spent anyway. The possible review of the decision is relevant to the question of hardship and prejudice to good administration. While the assessment of delay requires an examination of the claimant’s conduct, the issue of prejudice caused by any delay is at large.

There is no continuing financial loss because the project is on hold. Uncertainty among parents and the education community is not a product of tardiness in bringing the claim but of the continuing discussions about the merits of the project. The pace of those discussions is not particularly fast; a meeting with the Welsh Government is anticipated during the autumn of 2022, that is to say, about now. It is said that the final decision on whether to proceed is likely to be taken in about April 2023. It is therefore now doubtful whether the new school could open in September 2024, irrespective of this litigation.

For those reasons, the claim succeeds on the first ground only. The council argued that a declaration to the effect that the council erred by not including a WLIA in the consultation would suffice. A quashing order was not needed because the declaration would make clear (subject to any appeal or stay pending appeal) that a fresh process would be needed before the proposals could be implemented.

The claimant submitted that a quashing order was necessary and appropriate. On reflection, I agree. In the current state of uncertainty surrounding the possible review of the cabinet’s decision, the more legal certainty there is about the status of that decision, the better. Non-lawyers and the public could be confused by the notion that the decision subject to review, while made in error, has not been expressly overturned by the court.