(with whom Lord Neuberger, Lord Mance, Lord Kerr, Lord Sumption and Lord Reed agree)
These appeals arise out of the decision of the government to promote the high speed rail link from London to the north known as HS2. The decision was announced in a command paper, "High Speed Rail: Investing in Britain's Future - Decisions and Next Steps" (Cm 8247, 10 January 2012). (It has been referred to in the proceedings as "the DNS".) The main issues, in summary, are, first, whether it should have been preceded by strategic environmental assessment, under the relevant European Directive, and, secondly, whether the hybrid bill procedure, as currently proposed, will comply with the procedural requirements of European law. The Court of Appeal decided both issues against the appellants, the first by a majority (Sullivan LJ dissenting). We also need to consider the possibility of referring either question to the European court ("CJEU").
In the first appeal, the HS2 Action Alliance is a not-for-profit organisation working with over 90 affiliated action groups and residents' associations in opposition to the HS2 scheme. The appellants in the Hillingdon appeal are local authorities along the proposed route of Phase 1 of HS2. They are all members of "the 51M group", a group of local authorities which joined together in a national campaign to oppose the HS2 rail proposals. In the third appeal, Heathrow Hub Limited ("HHL") has for many years promoted the concept of a multi-modal transport hub at Heathrow Airport, integrating Heathrow with road, conventional mainline railway and high speed rail services.
In January 2009, the previous government established a company called High Speed Two Limited ("HS2 Ltd") to advise on proposals for "a new railway from London to the West Midlands and potentially beyond". In December 2009 HS2 Ltd reported to the Secretary of State. The options for routes north of Birmingham include what became the preferred Y-shaped network and two others known as the "reverse S" and the "reverse E" configurations. On 15 December 2009, the Secretary of State made a statement to Parliament setting out his proposed next steps, including a White Paper by the end of March 2010, followed by a full public consultation in the autumn of 2010, leading to preparation of a hybrid Bill.
On 11 March 2010, the Department for Transport published a Command Paper entitled High Speed Rail (Cm 7827), along with HS2 Ltd's report and other technical reports. The initial core high speed network would link London to Birmingham, Manchester, the East Midlands, Sheffield and Leeds, and be capable of carrying trains at up to 250 miles per hour. It would take the form of a Y-shaped network of around 335 miles. The paper explained the government's reasons in the light of the HS2 Ltd studies for preferring it to the "reverse S" and "reverse E" configurations. The studies had shown that as a first step a high speed line from London to Birmingham would offer high value for money. There would be connections to existing tracks to allow direct high speed train services to destinations further north. The capacity so released would be used to expand commuter, regional and freight service on existing lines.
The first phase would run from a rebuilt Euston Station to a new Birmingham City Centre Station. Following further work by HS2 Ltd, formal public consultation would begin in the autumn. At the same time detailed planning work would begin on the routes from Birmingham to Manchester and Leeds, with a view to public consultation in early 2012. High speed access to Heathrow would be provided by a link with Crossrail and the Heathrow Express, but the government had appointed Lord Mawhinney to assess the options for a potential station at Heathrow.
As to the procedure it was stated, at p 9, point 17:
That powers to deliver this proposed high speed rail network should be secured by means of a single Hybrid Bill, to be introduced subject to public consultation, environmental impact assessment and further detailed work on funding and costs to feed into decisions to be taken in the next Spending Review. Depending on Parliamentary timescales and approval, this could allow construction to begin after the completion of London's Crossrail line, opening from 2017, with the high speed network opening in phases from 2026.
Following the general election in May 2010, the proposals were adopted by the new Coalition Government, but it was indicated that due to financial constraints it would be achieved in phases. In June 2010 Lord Macwhinney reported that there was no compelling case for a direct high speed link to Heathrow. In October 2010, following further work by HS2 Ltd, the Secretary of State announced the preferred option for north of Birmingham involving two separate corridors, one via Manchester and the other via the East Midlands (the "Y-network"). In December the Secretary of State published details of the proposed route for Phase 1 between London and Birmingham. The proposed route included provision for a spur link to Heathrow Airport, to be built later at the same time as the lines to Leeds and Manchester.
In February 2011, the government opened formal public consultation on the high speed rail proposals, including the proposed Y-network, and the preferred route for Phase 1 from London to the West Midlands. There was a consultation report entitled "High Speed Rail: Investing In Britain's Future", accompanied by an "Appraisal of Sustainability", and other economic and technical studies. The Secretary of State's foreword described the consultation as "one of the largest and most wide-ranging ever undertaken by Government". The government would announce the result of the consultation and final decisions on its strategy for high speed rail before the end of 2011.
Among other responses, the 51M group submitted an extensive consultation response objecting to the principle of HS2, challenging the government's case on business and capacity grounds, expressing concerns over the environmental impact, and arguing that the Appraisal of Sustainability had not been properly carried out or consulted upon with regard to other alternatives. In particular it submitted that any necessary increase in capacity could be provided more cost effectively by an alternative proposal, known as "the optimised alternative", based on improving existing lines and services. Camden Council submitted a separate response raising concerns about the impact on the community and infrastructure around Euston. HHL contended that the mainline of HS2 should run via Heathrow.
On 10 January 2012, the Department for Transport published the DNS. It included confirmation of the government's high speed rail strategy and a summary of its decisions, a review of the consultation responses, and statement of the next steps. With regard to alternatives, it was noted that "relatively few" responses had discussed the merits of the proposed Y-network, but so far as alternatives were put forward the government remained of the view for the reasons given previously that its proposal offered the most effective approach. Under the heading "Alternatives to high speed rail", the paper considered options for upgrading the existing network, including the "optimised alternative" proposed by the 51M group. It was concluded that the approach of upgrading the existing network would be "incapable of matching the scale of the benefits that could be provided by a new high speed rail line", although it accepted that such alternatives would be expected to have some advantages, such as "lower sustainability impacts than entirely new lines, including smaller impacts on noise, landscape and townscape". The overall conclusion was that "any sustainability and cost advantages are outweighed by the substantial disbenefits of enhancing existing lines" (paras 3.77- 3.92). The DNS set out the process by which the government intended to obtain development consent for HS2, namely through two Hybrid Bills in Parliament, the first for Phase 1 and the second for Phase 2.
The DNS also stated that following consultation safeguarding directions would be issued under the planning laws to safeguard the Phase 1 route corridor adopted by the DNS from incompatible development. Consultation on safeguarding was started in October 2012 and completed in January 2013. On 9 July 2013, the Safeguarding Direction was made. The effect is that the Secretary of State will be notified if a local planning authority is minded to grant planning permission for any development which HS2 Ltd considers would conflict with the Phase 1 route corridor, and the Secretary of State has power to give directions restricting the grant of planning permission, either indefinitely or during such a period as may be specified.
The making of the Safeguarding Direction also triggered the statutory blight procedures. Eligible property owners within the safeguarded area may serve a blight notice asking the Secretary of State to buy their property prior to it being needed for construction. A High Speed Rail (Preparation) Bill was introduced into the House of Commons on 13 May 2013 and received Royal Assent on 21 November 2013. It was described as a "paving bill" to enable the Secretary of State "to incur essential expenditure on preparatory works" to allow the construction programme to proceed as quickly as possible following Royal Assent for the main bill.
Meanwhile, work on Phase 2 continued. Public consultation on the detailed route for Phase 2 of the Y-network began on 17 July 2013. It took the form of a consultation paper ("Consultation on the route from the West Midlands to Manchester, Leeds and beyond"), with supporting documents. The proposals for Phase 2 were broadly in accordance with the Government's High Speed Rail Strategy as set out in the DNS. The paper states that the current intention is to bring forward a hybrid Bill for Phase 2 in the next Parliament, following the May 2015 General Election.
The present proceedings were commenced in April 2012. Following a ten day hearing in December 2012, Ouseley J gave judgment on 15 March 2013. The judgment is a tour de force running to 844 paragraphs, and dealing with a wide range of issues, most of which happily are no longer in dispute. He upheld the claim in relation to certain aspects of the consultation process, but dismissed it on the issues relevant to the present appeal. The Court of Appeal (Lord Dyson MR, Richards and Sullivan LJJ) following a hearing in June 2013, gave judgment dismissing the appeal on 24 July 2013. I will need to return to the reasoning of the judgments below when considering the submissions before the Supreme Court.
The issues before this court can be summarised as follows:
SEA whether the DNS in the circumstances of HS2 is a "plan or programme" which "sets the framework for development consent" and was "required by administrative provisions" within the meaning of articles 2-3 of Directive 2001/42/EC ("the SEA Directive").
Aarhus whether if the interpretation of the majority in the Court of Appeal is correct, article 3(2)(a) of the SEA Directive is inconsistent with article 7 of the Aarhus Convention, and if so with what consequences.
EIA/Hybrid Bill whether the Hybrid Bill procedure as proposed meets the requirements of Directive 2011/92/EU ("the EIA Directive"), taking account in particular that (a) issues of principle will be excluded from the Select Committee stage, and (b) the debate on the Bill at Second and Third Reading will be subject to a Government whip.
Timing whether the court should intervene at this stage, or whether the court should wait until the Parliamentary process is completed;
CJEU reference whether any of the above questions raise uncertain issues of European law on which a reference should be made to the European court.
Since the hearing the hybrid bill for Phase 1 has been introduced to Parliament and received its first reading on 25 November 2013. The issues relating to the parliamentary process (iii) and (iv) will be discussed by Lord Reed, with whose reasoning and conclusions I agree.
The SEA Directive
The relevant provisions of the directive and extracts from the authorities are quoted at length in the judgment of the Master of the Rolls. I can therefore be more selective. At issue is the interpretation of article 3 which provides:
HS2 is such a transport project.
By article 2(a) "plans and programmes" means plans and programmes -
Although not directly applicable, attention should be drawn also to articles 3.4 and 3.5, by which member states are required to determine whether plans and programmes, other than those referred to in paragraph 2, which set the framework for future development consent of projects, are likely to have significant environmental effects. In making that determination on a "case-by-case examination", they are required to take into account the criteria set out in Annex II. Those criteria include:
The characteristics of plans and programmes, having regard, in particular, to
We were referred to three relevant European authorities on the interpretation of the definition:
Terre wallonne ASBL and Inter-Environnement Wallonie ASBL v Région wallonne (Joined Cases C-105/09 and C-110/09)  ECR I-5611 ("Terre wallonne")
Inter-Environnement Bruxelles ASBL v Région de Bruxelles-Capitale (Case C-567/10)  2 CMLR 30 ("I-E Bruxelles")
Nomarchiaki Aftodioikisi Aitoloakarnanias v Ipourgos Perivallontos, Khorotaxias kai Dimosion Ergon (Case C-43/10)  Env. L.R. 21 (Grand Chamber) ("Nomarchiaki")
The debate in this court has centred on two parts of the definition: "required by .... administrative provisions" and "set the framework for future development consent...."
Required by administrative provisions
As explained by the CJEU, the word "required" in this context means no more than "regulated": I-E Bruxelles para 31. But it is less clear how that concept applies to administrative, as opposed to legislative or regulatory, provisions. In Walton v The Scottish Ministers  UKSC 44;  PTSR 51, at para 99, I said:
There may be some uncertainty as to what in the definition is meant by 'administrative', as opposed to 'legislative or regulatory', provisions. However, it seems that some level of formality is needed: the administrative provisions must be such as to identify both the competent authorities and the procedure for preparation and adoption.
The appellants submitted that the March 2010 Command Paper satisfied this part of the definition, since it determined the competent authority for adopting the plan and the procedure for preparing it. The majority in the Court of Appeal were inclined to agree, adopting a "broad and purposive interpretation", but found it unnecessary to reach a decision on this point (para 71). Sullivan LJ held that this part of the definition was satisfied: although there were some changes to the procedure set out in the 2010 Command Paper, the process there described was in substance followed by the new government, and to that extent "regulated" the preparation and adoption of the DNS (paras 180-182). Mr Mould was disposed to accept that the 2010 Paper was at least arguably "an administrative provision" within this part of the definition, but not that it "regulated" the procedure in the formal sense.
I am prepared to proceed on the assumption that Sullivan LJ was right on this point, or at least that there is a referable issue on the meaning of that part of the definition. I therefore turn to what emerged as the principal issue between the parties, that is the reference to a plan or programme which "sets the framework for future development consent".
Setting the framework
Terre wallonne concerned an "action programme" adopted under article 5(1) of Directive 91/676 concerning the protection of waters against pollution caused by nitrates from agricultural sources. The issue was whether it fell within article 3(2)(a) of the SEA Directive on the basis that it set the framework for future development consent of intensive livestock installations (listed in Annexes I and II to the EIA Directive). As Advocate General Kokott explained (paras 60-67), the main issue for the court was how strongly the requirements of the plans or programmes "must influence individual projects" in order to come within the definition. This was against the background of arguments by certain member states that the framework must "determine the location, nature or size" (her emphasis) of projects requiring environmental assessment. She rejected that view as too narrow. She concluded:
In deciding that the definition did apply to the instant case she noted that under article 8 of the EIA Directive consideration must be given not only to direct effects of the planned works, but also to effects on the environment arising from their use, including in this case the effects on water quality resulting from intensive lifestock installations, and that therefore disposal of manure arising had to be considered (para 80). She concluded (para 82):
In the context of such consideration, the framework set by the action programme has at least the effect that it must be possible for the installation to be operated in accordance with the provisions of the programme. At the same time, however, development consent can hardly be refused on grounds of the pollution of waters by nitrate from agriculture if the project complies with the rules of the programme. Certain alternatives, which are harmful to the environment as gauged by the objectives of the action programme, are thus excluded and others, which possibly afford water greater protection, do not have to be examined and taken into consideration ....
As I read her opinion, the references to "influence" in the earlier paragraphs were to indicate that something less than a specific determination of the nature of the project would suffice. On the other hand, the latter paragraph shows that "influence" as such might not be enough; the critical factor was that consideration of certain environmental effects would in practice be excluded altogether.
The court (paras 52-54) agreed with her as to the relevance of article 8 of the EIA Directive, and noted that under article 5(4) of Directive 91/676 action programmes must "provide for a set of measures compliance with which can be a requirement for issue of the consent", including requirements for storage of livestock manure. It concluded:
Accordingly, in answer to the relevant question, it held that an action programme adopted pursuant to article 5(1) of Directive 91/676 was in principle a plan or programme covered by article 3(2)(a) since it constituted a "plan" or "programme" within the meaning of article 2(a) and –
contains measures compliance with which is a requirement for issue of the consent that may be granted for carrying out projects listed in Annexes I and II to Council Directive 85/337.
In I-E Bruxelles, the court held that the repeal of a land use plan was capable of falling within the scope of the SEA Directive, even in the absence of any specific reference in its text to "repeal" (as opposed to "modification"). The court rejected a narrow interpretation as contrary to the objective of the directive to provide for a high level of protection of the environment (para 30):
That interpretation would thus run counter to the directive's aim of establishing a procedure for scrutinising measures likely to have significant effects on the environment, which define the criteria and the detailed rules for the development of land and normally concern a multiplicity of projects whose implementation is subject to compliance with the rules and procedures provided for by those measures.
The same formula (emphasised in the above quotation) was adopted by the Grand Chamber in Nomarchiaki. The case concerned a controversial project for the diversion of the River Acheloos in western Greece, to serve the irrigation and energy needs of the region of Thessaly. One of many issues was whether it should be regarded as a plan or programme within the meaning of the SEA Directive. Differing from the Advocate General, the court said no, and dealt with the issue very briefly (para 95):
It is not evident that the project concerned constitutes a measure which defines criteria and detailed rules for the development of land and which subjects implementation of one or more projects to rules and procedures for scrutiny (see, to that effect, [Inter-Environnement Bruxelles (Case C-567/10)  CMLR 909, para 30)] ....
The Court of Appeal
In the Court of Appeal a joint judgment was given by the Master of the Rolls and Richards LJ. Having referred to the "paradigm" case of a statutory development plan, and building on the Advocate General's discussion in Terre wallone, they spoke of the "different degrees of influence" which a plan might have (para 54):
At one end of the spectrum is the plan or programme which conclusively determines whether consent is given and all material conditions. Such a plan or programme clearly sets the framework. It is an example of legal influence of highest order. At the other end of the spectrum is the plan or programme which identifies various development options, but which states that the decision-maker is free to accept or reject all or any of the options.
In their view, however, it was not necessary for the plan to be legally binding (para 50):
We would not .... rule out the possibility that a plan or programme may set the framework where it has sufficiently potent factual influence, but (as we shall explain) not where the decision-maker is Parliament. If it is clear that the decision-maker will follow the recommendations contained in a plan or programme and the measures are likely to have significant effects on the environment, then the mere fact that the decision-maker is not legally obliged to make a decision in accordance with the plan or programme might not be a sufficient reason for holding that the plan or programme does not set the framework. But in our view, there must at least be cogent evidence that there is a real likelihood that a plan or programme will influence the decision if it is to be regarded as setting the framework ....
Applying the test as set out in the italicised words, they agreed with Ouseley J that the DNS was not within the definition. The DNS would have no legal influence on Parliament, which was not obliged to comply with it or even to have regard to it in reaching its decision. Nor was it appropriate or possible for the court to assess the degree of influence the DNS was likely to have as a matter of fact on Parliament's decision-making process (para 56):
Parliament is constitutionally sovereign and free to accept or reject statements of Government policy as it sees fit, and the court should not seek to second guess what Parliament will do. Moreover the decision whether to give consent to the project as outlined in the DNS is very controversial and politically sensitive. No final decision has yet been taken as to the form or length of debate that is to take place in Parliament.
Sullivan LJ was concerned that the majority's interpretation would leave an undesirable gap in strategic environmental protection; governments would be able to avoid the need for strategic environmental assessment by promoting specific acts of legislation (paras 154-7). He applied the same test as the majority but disagreed as to the result. He considered that there was "cogent evidence" of a "real likelihood" that the DNS would "influence" Parliament's decision. In the present context (by contrast with that of the "conventional" development control process), he rejected as unrealistic a distinction between the role of the government as promoter of the scheme and its role in the Parliamentary decision-making process (para 173):
When considering the status of the DNS in the hybrid Bill procedure it must be recognised that the Government has a dual role. Having devised the 'plan' the Government is not merely the promoter of the project, it will actively participate in the decision-making process under the hybrid Bill procedure. Parliament is constitutionally distinct from the executive, but members of the Government are members of Parliament .... The well-established convention of collective ministerial responsibility will ensure that the plan prepared by the Government (the DNS) will in fact have a very significant influence on Parliament's decision making process in respect of a Government Bill.
Drawing a parallel with the purposive approach of the CJEU to legislative decisions in the context of the EIA, he thought that the court should look at "the substance and not simply the constitutional formality of the entire decision-making process" (para 174).
The arguments in this court have broadly followed those summarised in the judgments of the Court of Appeal, and reflected in the respective views of the majority and minority. They have been developed at considerable length in the written and oral submissions to this court. I hope I will be forgiven for not attempting to summarise them further in this judgment. The difference between the parties in the end comes down to a relatively short point of construction of the directive and its application to the special facts of this case.
In Lord Reed's judgment in Walton v Scottish Ministers  PTSR 51, para 10ff, there is a detailed discussion of the evolution and general purpose of the relevant directives. It is unnecessary to repeat it here. He cited in particular (para 12 of that judgment) the helpful discussion by Advocate General Kokott (in Terre wallone, points 31-32) of the evolution of the SEA Directive to fill a perceived gap in the EIA regime:
The application of the EIA Directive revealed that, at the time of the assessment of projects, major effects on the environment are already established on the basis of earlier planning measures.... Whilst it is true that those effects can thus be examined during the environmental impact assessment, they cannot be taken fully into account when development consent is given for the project. It is therefore appropriate for such effects on the environment to be examined at the time of preparatory measures and taken into account in that context.
He referred also (para 14 of that judgment) to an extract from the European Commission's first report on the application of the SEA Directive ((COM(2009) 469 final, para 4.1):
The two Directives are to a large extent complementary: the SEA is 'up-stream' and identifies the best options at an early planning stage, and the EIA is 'down-stream' and refers to the projects that are coming through at a later stage. In theory, an overlap of the two processes is unlikely to occur. However, different areas of potential overlaps in the application of the two Directives have been identified.
In particular, the boundaries between what constitutes a plan, a programme or a project are not always clear, and there may be some doubts as to whether the 'subject' of the assessment meets the criteria of either or both of the Directives.
It should be borne in mind also that, although the expression "strategic" is commonly used in shorthand descriptions of the directive, it is not a word that appears in the text. The correct title is "Directive.... on the assessment of the effects of certain plans and programmes on the environment". It is not therefore to be assumed, as some of Mr Elvin's submissions seemed to imply, that because a project is "strategic" in nature (as HS2 undoubtedly is) the presumption must be in favour of assessment under this directive. The purpose is more specific, that is to prevent major effects on the environment being predetermined by earlier planning measures before the EIA stage is reached.
Against that background, and unaided by more specific authority, I would have regarded the concept embodied in article 3.2 as reasonably clear. One is looking for something which does not simply define the project, or describe its merits, but which sets the criteria by which it is to be determined by the authority responsible for approving it. The purpose is to ensure that the decision on development consent is not constrained by earlier plans which have not themselves been assessed for likely significant environmental effects. That approach is to my mind strongly supported by the approach of the Advocate General and the court to the facts of Terre wallone and by the formula enunciated in I-E Bruxelles and adopted by the Grand Chamber in Nomarchiaki.
In relation to an ordinary planning proposal, the development plan is an obvious example of such a plan or programme. That is common ground. Even if as in the UK it is not prescriptive, it nonetheless defines the criteria by which the application is to be determined, and thus sets the framework for the grant of consent. No doubt the application itself will have been accompanied by plans and other supporting material designed to persuade the authority of its merits. In one sense that material might be said to "set the framework" for the authority's consideration, in that the nature of the application limits the scope of the debate. However, no one would for that reason regard the application as a plan or programme falling within the definition.
In principle, in my view, the same reasoning should apply to the DNS, albeit on a much larger scale. It is a very elaborate description of the HS2 project, including the thinking behind it and the government's reasons for rejecting alternatives. In one sense, it might be seen as helping to set the framework for the subsequent debate, and it is intended to influence its result. But it does not in any way constrain the decision-making process of the authority responsible, which in this case is Parliament. As Ouseley J said:
With respect to Sullivan LJ, I do not think that position is materially changed by what he called the "dual role" of government. Formally, and in reality, Parliament is autonomous, and not bound by any "criteria" contained in previous government statements.
I have noted that the majority and the minority in the Court of Appeal adopted the same test, turning on the likelihood that the plan or programme would "influence" the decision. The majority referred to the possibility of the plan having "a sufficiently potent factual influence" (para 55). Although Mr Mould generally supported the reasoning of the majority, he submitted that "influence" in the ordinary sense was not enough. The influence, he submitted, must be such as to constrain subsequent consideration, and to prevent appropriate account from being taken of all the environmental effects which might otherwise be relevant.
In my view he was right to make that qualification. A test based on the potency of the influence could have the paradoxical result that the stronger the case made in favour of a proposal, the greater the need for strategic assessment. Setting a framework implies more than mere influence, a word which is not used by the court in any of the judgments to which we have been referred. It appears in annex II of the directive, but only in the different context of one plan "influencing" another. In Terre wallone Advocate General Kokott spoke of influence, but, as already noted, that was by way of contrast with the submissions before her which suggested the need for the plan to be "determinative".
Finally, Mr Elvin pointed to the fact that the DNS had specific legal consequences, notably in the safeguarding direction, and the consequent application of the related blight provisions, and also in providing the basis for the paving Bill, and for the allocation of resources under it. I accept that these points provide an arguably material distinction from the supporting material for a conventional planning application. However, they do not imply any further constraint on Parliament's consideration of the environmental impacts of the project as a whole, under the hybrid Bill procedure.
Sullivan LJ was concerned that the majority's interpretation would leave a gap in the environmental protection provided by the directives. It is helpful to consider this concern in the context of the facts of the present case. The government's case from the beginning has been that the SEA Directive has no application because neither the DNS, nor anything which preceded it, was a "plan or programme" as there defined. They accept however that as a "project" within the meaning of the EIA Directive it must be subject to environmental assessment in a modified form adapted to the proposed legislative procedure (as discussed by Lord Reed).
It is common ground, as I understand it, that the difference between the two procedures is significant principally in relation to the treatment of alternatives. The respective requirements are:
SEA Directive Article 5 provides that the environmental report must "identif[y] describ[e] and evaluat[e]" the likely significant effects on the environment of implementing both the plan or programme itself, and "reasonable alternatives". Annex 1 sets out the information to be given, including "an outline of the reasons for selecting the alternatives dealt with" and a description of how the assessment was undertaken.
EIA Directive Article 5 requires the statement to include the information specified in annex IV, which includes simply "an outline" of the main alternatives studied by the developer and an indication of the "main reasons for this choice, taking into account the environmental effects".
The reasons for this difference are not obvious. It may simply reflect the different stages at which the two exercises are carried out. At the earlier stage of strategic assessment neither the proposed plan nor the alternatives will need to have been worked up to the same degree of detail as will be appropriate at the EIA stage. At the latter stage to require an equivalent degree of detail for the rejected alternatives may be seen as unduly burdensome.
In any event, it was not in dispute between the parties that the treatment of alternatives required under the SEA Directive is more detailed than under the EIA, and that it was not satisfied in this case. It is also common ground that compliance with the SEA Directive at this stage would be possible, but that it would involve significant delay. Mr Mould on instructions, and without dissent, spoke of an added delay of six months to a year.
There is also a measure of agreement as to what such additional consideration would involve. Ouseley J considered whether, in spite of the government's position that such treatment was unnecessary, "substantial compliance" had been achieved (paras 160-172). In a passage the reasoning of which has not been challenged before this court, he concluded that it had not been achieved, for reasons "essentially related to the Y-network and its alternatives, and the spurs to Heathrow". On the other hand, as Mr Mould emphasises, he took a different view in relation to Phase 1 in respect of which, viewed on its own, he would have found substantial compliance with the SEA Directive (para 168).
Furthermore, in his view, even if the SEA Directive had applied, it would not have required more detailed consideration of alternative strategies based on improvements to the existing network, such as the optimised alternative (para 162):
The Government concluded that alternative strategies for motorways or a new conventional or enhanced existing rail network were not capable of meeting the plan objectives set for high speed rail. It is obviously a contestable view as to whether those objectives should be met, or can be met to a large extent by means other than a new high speed rail network. These alternative strategies could not, however, have constituted reasonable alternatives to the plan for assessment in the SEA, since they are incapable by their very nature of meeting all the objectives for a new high speed rail network. The sifting process whereby a plan is arrived at does not require public consultation at each sift. This whole process has been set out in considerable detail in the many published documents for those who wished to pursue it, but it did not all have to be in an SEA.
On that view, which was not challenged before us, application of the SEA Directive would result in more detailed consideration of alternatives such as the reverse S and reverse E configurations, but not of the optimised alternative. Since the optimised alternative is the only one for which the parties before us have expressed any positive support, the SEA process as such may not meet their particular needs (save possibly in respect of HHL's interest in the Heathrow Spur alternatives, although we were told that that aspect is affected by the current study of future airport capacity under Sir Howard Davies).
Miss Lieven suggested that the strategic significance of the optimised alternative might require reassessment in the light of more recent ministerial statements about the objectives of HS2. That cannot in my view affect our consideration of the present appeals, which are concerned with the procedural requirements for the DNS at the time it was made. On the other hand, nothing in the DNS prevents arguments and evidence relating to the government's present intentions being presented to Parliament within the current decision-making process. This indeed illustrates the practical importance of the distinction, in the context of the SEA Directive, between merely influencing subsequent consideration, and setting limits on the scope of what can be considered. Until Parliament has reached its decision, the merits of all aspects of the HS2 project, on economic, environmental and other grounds, remain open to debate.
It is convenient at this point to deal briefly with Mr Elvin's related argument under article 7 of the Aarhus Convention. That article requires provision to be made for the public to participate in the preparation of "plans and programmes relating to the environment". It is to be noted that this article refers to plans and programmes in general, without the qualifications found in the SEA Directive definition. It is not suggested, having regard to the extent of public consultation which has already taken place on the HS2 project, that there has been any breach of this requirement taken on its own, even assuming the DNS to be a "plan or programme" within the meaning of this article. Instead the argument, as I understand it, is that the SEA Directive must be interpreted in such a way as to ensure conformity with the Convention, which in turn requires that any plans or programmes covered by article 7 are also subject to the SEA procedure.
The majority of the Court of Appeal rejected this argument. They said (para 63):
.... our conclusion that the DNS is not a plan or programme setting the framework for future development consent does not in our view involve any incompatibility with article 7. If a plan or programme does not set the framework, it is difficult to see how article 7 can have been intended to apply to it. In such a case, the requisite degree of public participation can be achieved through compliance with the requirements of the EIA Directive in the development consent procedure for a specific project.
Sullivan LJ was unpersuaded by this reasoning. He thought that consultation under the EIA Directive was an inadequate response to article 7, because by that time "strategic alternatives will have been foreclosed by the legislative process and the pass will have been sold." (para 178)
To my mind there is a more fundamental objection to Mr Elvin's argument. There is no reason to assume that article 7 and the SEA Directive are intended to cover exactly the same ground. The differences in wording are clear and must be assumed to be deliberate. Indeed the UNECE guidance on the Convention (The Aarhus Convention: An Implementation Guide 2nd Ed 2013 p 118-119) accepts that its reference to plans and programmes relating to the environment is broader than the equivalent definition in the SEA Directive. The SEA Directive must be interpreted and applied in its own terms. If this falls short of full compliance with the Aarhus Convention, it does not invalidate the directive so far as it goes. It simply means that a possible breach of the Convention may have to be considered as a separate and additional issue. In the present case the point is academic because no such breach is alleged.
It will be apparent from what I have said that I do not find it necessary to make a reference to the CJEU in this case. I am conscious of the disagreement between the very experienced members of the Court of Appeal. However, they differed principally not on the formulation of the test, but on its application to the facts of the case, and in particular to the workings of the parliamentary process under domestic law. Although I have taken a rather different view of the appropriate legal test, that is because I have attached more importance to the guidance contained in the words of the court itself in the trilogy of cases to which I have referred.
This seems to me the kind of case which Advocate General Jacobs had in mind when (in Case C-338/95 Wiener S.1. GmbH v Hauptzollamt Emmerich  ECR I-6495, para 61) he referred to the emergence of a body of case law developed by the CJEU to which national courts and tribunals can resort in resolving new questions of Community law:
Experience has shown that that case law now provides sufficient guidance to enable national courts and tribunals - and in particular specialised courts and tribunals - to decide many cases for themselves without the need for a reference ....
That approach is also reflected in the recommendation issued by the court in September 2012, to which Lord Sumption has referred.
For these reasons, and those given by Lord Reed on the issues covered in his judgment, I would dismiss the appeals.
(with whom Lord Neuberger, Lady Hale, Lord Mance, Lord Kerr, Lord Sumption and Lord Carnwath agree)
Hybrid bill procedure and the EIA Directive
As Lord Carnwath has explained, the appeal brought by the London Borough of Hillingdon and nine other local authorities raises the question whether the hybrid bill procedure, under which Parliament is being invited to authorise the HS2 project by Acts of Parliament, is compliant with the requirements of the Environmental Impact Assessment Directive (Directive 2011/92/EU, OJ 2012, L 26/1, "the EIA Directive"). In particular, the appellants seek the quashing of the Government's decision, announced in the DNS, to pursue a hybrid bill for each phase of the Y network, and to introduce a hybrid bill by the end of 2013 to provide the necessary powers to construct and operate Phase 1. The question is also raised, on behalf of the respondents, whether it is appropriate for the court to consider the compatibility of the Parliamentary procedure with the EIA Directive at the present stage, or whether that issue should be considered only after the Parliamentary procedure has been completed. It is convenient to consider those questions together, as they are to some extent inter-related.
Hybrid bill procedure
It may be helpful at the outset to explain what is meant by hybrid bill procedure. A hybrid bill shares certain characteristics of a public bill and a private bill. The Speaker has defined a hybrid bill as "a public bill which affects a particular private interest in a manner different from the private interests of other persons or bodies of the same category or class" (Hansard (HC Debates), 10 December 1962, col 45). This hybrid character influences the Parliamentary procedure: a hybrid bill proceeds as a public bill, with a second reading, committee report and third reading, but with an additional select committee stage after the second reading in each House, at which objectors whose interests are directly and specifically affected by the bill (including local authorities) may petition against the bill and be heard. Parliamentary standing orders make provision for those persons who have standing to lodge a petition.
It is for Parliament and not the Government to determine the Parliamentary procedure for a hybrid bill laid before it. It is however a matter of agreement between the parties that, in the case of the hybrid bill for Phase 1 of HS2, the principle of the bill will be set upon the bill's receiving a second reading following debate, subject to the Government whip, in the House of Commons. It is expected that the principle of the bill will extend to a high speed rail line running between London, Birmingham and the West Midlands, with its central London terminus at Euston and a link to HS1 (ie the Channel Tunnel Rail Link). It is also common ground that the established convention is that a select committee for a hybrid bill cannot hear petitions which seek to challenge the principle of the bill, unless instructed to do so by the House at second reading (Erskine May's Treatise on the Law, Privileges, Proceedings and Usage of Parliament, 24th ed (2011), ed Jack, p 656). Under the Parliamentary procedures as currently envisaged by the Government, matters that go to the principle of the bill will not be considered by the select committee. Such matters would be expected to include the business case for HS2, alternatives to the high speed rail project and alternative routes for Phase 1. The principle of the bill could in theory be re-opened at third reading, but that debate also will be subject to the Government whip.
The relevant standing orders
In order to understand the arguments, it is also necessary to note the relevant Parliamentary standing orders ("SOs"). SO 27A for Private Business requires that a bill authorising the carrying out of works the nature and extent of which are specified in the bill must be accompanied by an environmental statement, which must be available for inspection and for sale at a reasonable price. The environmental statement must contain the information required by the Town and Country Planning (Environmental Impact Assessment) Regulations 2011 (SI 2011/1824), "the 2011 Regulations"), which transpose the requirements of the EIA Directive, so far as affecting applications for planning permission, into English law.
SO 224A, which was introduced in June 2013 after the hearing of the appeal in the Court of Appeal, requires that upon the deposit of the bill a notice must be published stating that any person who wishes to make comments on the environmental statement should send those comments to the minister responsible for the bill. The minister must publish and deposit the comments received, and submit them to an independent assessor appointed by the Examiner of Petitions for Private Bills. The assessor is then to prepare a report summarising the issues raised by those comments. The report must be submitted to the House at least 14 days prior to second reading. At third reading the minister must set out the main reasons and considerations upon which Parliament is invited to consent to the project and the main measures to avoid, reduce and if possible offset the project's major adverse effects. A written statement must be laid before the House not less than seven days before third reading. The House of Lords has made corresponding arrangements under SO 83A.
Finally, by way of introduction, it is necessary to consider the role of Government whips. In that regard, although the argument on behalf of the appellants was largely concerned with the implications, for the purposes of assessing compliance with the EIA Directive, of the fact that votes on the bill are intended to be subject to the Government whip, the court was not provided with any authoritative account of how the whip operates.
In general terms, the Government whips are ministers responsible for fitting the Government's programme of business into the time available during the session. The Opposition parties also have whips, who are members of either House, appointed by their party in Parliament to help organise their party's contribution to Parliamentary business. The term is derived from hunting: a whipper-in is a huntsman's assistant, who drives straying hounds back to the pack using a whip. One of the whips' duties is to see that their parties are as fully represented as possible at important votes or "divisions", and, in the Commons, to arrange "pairs" for members who wish to be absent (a "pair" being a member of the opposite party who also wishes to be absent). Each week they send a circular to their Members of Parliament or peers, detailing the forthcoming Parliamentary business. Items underlined once are considered routine and attendance is optional. Those underlined twice are more important and attendance is expected unless a "pair" has been arranged. Items underlined three times, such as second readings of significant bills, are highly important. The failure of Government backbenchers to attend a vote with a three-line whip, or their voting contrary to Government policy on such an occasion, may have disadvantageous consequences for them within their party, including in extreme circumstances the possibility of suspension from the Parliamentary party. In that event the member keeps his seat but sits as an independent until the whip is restored.
In practice, Members of Parliament have to consider a range of factors besides the guidance of the whips. For example, in relation to controversial developments affecting their constituencies, Members of Parliament have to consider the views of their constituents: if they fail to do so, they may lose their constituents' support, and may in consequence be liable to lose their seat at the next election.
Although Government backbenchers generally support Government policies, failures to vote in accordance with the whip headingare not infrequent. One recent study found that Members of Parliament on the Government benches had voted against the whip in 43% of divisions during the first 18 months of the current Government: P Cowley and M Stuart, "A Coalition with Two Wobbly Wings: Backbench Dissent in the House of Commons", (2012) Political Insight, 3, pp 8-11. It also has to be borne in mind that the apprehension of backbench dissent may result in changes to proposed legislation, so as to ensure that the Government will not be defeated. A study of the Government elected in 2001 carried out by Professor Philip Cowley of the University of Nottingham, for example, concluded that the fact that it had never suffered a defeat on a whipped vote could "hardly be seen as evidence of parliamentary impotence .... From the very beginning, the 2001 Parliament saw the Government give ground to its backbench critics on measure after measure, including on almost all major policy initiatives": Cowley, The Rebels: How Blair Mislaid His Majority (2005), pp 242-243. The same study commented that to focus on the weakness of Members of Parliament and the disciplinary power of the whips was "a quite monumental failure to understand the realities of parliamentary life" (op cit, p 48).
In some circumstances, it may in any event be impractical for the Government to proceed with a project without the support of the Opposition, as well as that of its own backbenchers. That may be the position, for example, where the period of time over which substantial Government resources require to be committed will extend beyond the Parliament during which the necessary legislation is enacted. In such a situation, there may be little purpose in obtaining Parliamentary approval for a project unless there is confidence that a future government, even if of a different party, will continue to support the project as so approved. Whether the HS2 project might be in that position, as has been suggested in public debate on the issue, was not addressed in the submissions.
In relation to voting in Parliament, it is also relevant to note the convention that members of the Government do not vote against Government legislation. If they do so, they are generally expected to resign, failing which they may be dismissed. Their loss of office does not affect their position as Members of Parliament.
The appellants' argument
At the hearing of the appeal, which was held before the bill for Phase 1 was introduced into Parliament, the appellants argued as follows. The Government intends to seek development consent for HS2 through hybrid bills in Parliament, without going through all the procedures required by the EIA Directive. The Government relies on the exemption granted by article 1(4):
This Directive shall not apply to projects the details of which are adopted by a specific act of national legislation, since the objectives of this Directive, including that of supplying information, are achieved through the legislative process.
As construed by the Court of Justice, however, that provision applies only where the objectives of the Directive are fulfilled by the legislative process.
In order to achieve the objectives of the EIA Directive, it is argued, the Parliamentary procedure must allow effective public participation, as required by article 6(4):
The public concerned shall be given early and effective opportunities to participate in the environmental decision-making procedures referred to in article 2(2) and shall for that purpose, be entitled to express comments and opinions when all options are open to the competent authority or authorities before the decision on the request for development consent is taken.
The procedure must therefore permit the public to produce information demonstrating why the HS2 project should not proceed, and that information must be capable of influencing the outcome of the decision-making process. In particular, these appellants must be able to provide information about their proposal for the optimised alternative, and Members of Parliament must be able to consider it and to be influenced by it.
It is however not possible, it is argued, for there to be effective public participation under the procedure envisaged. The Government has already taken the critical decision to accept the economic case for HS2 and to reject the optimised alternative. The bill for Phase 1 will reflect that decision. Parliament will be asked to approve the principle of the bill at second reading. It will have available to it an environmental statement prepared on behalf of the Secretary of State. It will also have available to it the comments on the environmental statement and the assessor's summary of those comments. The Secretary of State has however confirmed in correspondence that the vote at the conclusion of the debate will be subject to the whip. Ministers will face the alternatives of resignation or dismissal from office if they vote against the bill. Backbenchers will risk disciplinary sanctions.
The effect of the whip and of collective ministerial responsibility, it is argued, is therefore that ministers and backbenchers will be unable to give proper consideration to the environmental information when examining or debating the bill. Both the imposition of the Government whip, and collective ministerial responsibility, are incompatible with the EIA Directive, since they necessarily render public participation ineffective.
Furthermore, it is argued, the environmental information which will be provided to Parliament in respect of HS2 is so voluminous and complex that Members of Parliament cannot possibly consider it properly following the procedure envisaged. The draft environmental statement prepared on behalf of the Secretary of State extends to 27 volumes. The material produced by these appellants, relating to the optimised alternative, is also substantial and detailed. There is no requirement that Members of Parliament should read the environmental statement, the non-technical summary which it contains, or the summary of consultation responses, before voting. It is implausible to suggest that all or even a majority of Members of Parliament who vote on the bill will have done so. A second reading debate will not in any event allow a proper examination of the material to take place. Such a debate is likely to last only one or two days. There is no requirement that Members of Parliament should be present in the Chamber during the debate, and it would be implausible to suggest that the majority of Members who vote on the bill will have been present.
The subsequent examination of the bill by a select committee will not, it is argued, involve any consideration of the principle of the bill, and therefore will not involve consideration of alternatives to HS2 or of the economic case for the project. Any argument that the environmental impact as set out in the environmental statement should outweigh the need for HS2 will not be considered by the select committee. The final vote on the bill at third reading will again be subject to the whip.
Put shortly, it is argued that the effect of (1) the whipping of the vote at second and third readings, (2) the limited opportunity which is provided by a debate in Parliament for the examination of the environmental information, and (3) the limited remit of the select committee following second reading, is to prevent effective public participation, contrary to article 6(4) of the EIA Directive.
At the least, it is argued, the question whether the proposed procedure is compliant with the EIA Directive is not acte clair, and should therefore be the subject of a reference to the Court of Justice.
It is argued that this is a matter on which the court should rule now. The critical decision which is subject to challenge is not any decision of Parliament, but the decision of the Government to promote a hybrid bill and to impose a whip upon its progress through Parliament. Although the Parliamentary procedure is capable of being changed, the Government has no intention of seeking any such change. It is in addition obligatory under article 6(2) of the EIA Directive that the public should be informed early in the decision-making process of the procedure to be followed. That provision states:
The public shall be informed, whether by public notices or by other appropriate means such as electronic media where available, of the following matters early in the environmental decision-making procedures referred to in article 2(2) and, at the latest, as soon as information can reasonably be provided: ....
It is argued that it is in any event preferable, and consistent with the objectives of the Aarhus Convention, that this issue should be considered by the court before the substantial time and expense involved in the Parliamentary proceedings have been incurred by the appellants and others. Judicial scrutiny of the Parliamentary procedure cannot be avoided in the event that a bill is passed following the procedure proposed, since the courts cannot in that event avoid determining whether the requirements of the EIA Directive have been satisfied. That follows from article 11(1), which provides:
Member states shall ensure that, in accordance with the relevant national legal system, members of the public concerned:
have access to a review procedure before a court of law or another independent and impartial body established by law to challenge the substantive or procedural legality of decisions, acts or omissions subject to the public participation provisions of this Directive.
The respondents' argument
On behalf of the respondents, it was contended that the court could not determine in advance whether the quality of the Parliamentary proceedings would meet the requirements of article 1(4) of the EIA Directive, as construed by the Court of Justice. In view of the requirements of SO 27A and SO 204A, the hybrid bill procedure was in principle capable of meeting those requirements. The appellants' contentions to the contrary were based merely on assertion. Whether Members of Parliament had in fact given effective scrutiny to the bill and taken account of public representations and comments would fall to be considered by the court, if called upon to do so, after the legislation had been enacted, in the light of the Parliamentary debates. At the same time, if the court considered that there were inherent defects in the procedure currently contemplated, it would be helpful for it to say so.
The argument presented on behalf of the appellants as to the implications of the EIA Directive, if well founded, impinges upon long-established constitutional principles governing the relationship between Parliament and the courts, as reflected for example in article 9 of the Bill of Rights 1689, in authorities concerned with judicial scrutiny of Parliamentary procedure, such as Edinburgh and Dalkeith Railway Co v Wauchope (1842) 8 Cl & F 710; 1 Bell 252, Lee v Bude and Torrington Junction Railway Co (1871) LR 6 CP 576, Pickin v British Railways Board  AC 765 and Wilson v First County Trust Ltd (No 2)  UKHL 40;  1 AC 816, and in other cases concerned with judicial scrutiny of decisions whether to introduce a bill in Parliament, such as R (Wheeler) v Office of the Prime Minister  EWHC 1409 (Admin). Neither the Bill of Rights nor any of the authorities I have mentioned was however referred to in the parties' printed cases; nor was this issue mentioned before us until it was raised by the court. Nevertheless, it follows that the appellants' contentions potentially raise a question as to the extent, if any, to which these principles may have been implicitly qualified or abrogated by the European Communities Act 1972.
Contrary to the submission made on behalf of the appellants, that question cannot be resolved simply by applying the doctrine developed by the Court of Justice of the supremacy of EU law, since the application of that doctrine in our law itself depends upon the 1972 Act. If there is a conflict between a constitutional principle, such as that embodied in article 9 of the Bill of Rights, and EU law, that conflict has to be resolved by our courts as an issue arising under the constitutional law of the United Kingdom. Nor can the issue be resolved, as was also suggested, by following the decision in R v Secretary of State for Transport, Ex p Factortame Ltd (No 2)  1 AC 603, since that case was not concerned with the compatibility with EU law of the process by which legislation is enacted in Parliament. In the event, for reasons which I shall explain, it is possible to determine the appeal without requiring to address these matters.
National legislation and the EIA Directive
The EIA Directive was adopted in its original form in 1985 (Directive 85/337/EEC, OJ 1985, L 175/40). Since then it has been significantly amended by further directives, including the Public Participation Directive (Directive 2003/35/EC, OJ 2003, L 156/17). I shall refer to the EIA Directive as codified in 2011 (Directive 2011/92/EU, OJ 2012, L 26/1).
Article 1(4) of the EIA Directive (originally numbered 1(5)) has already been quoted, but it is convenient to remind oneself of its terms:
This Directive shall not apply to projects the details of which are adopted by a specific act of national legislation, since the objectives of this Directive, including that of supplying information, are achieved through the legislative process.
It might have been thought, in the light of that provision, that the Directive would not apply to the HS2 project if its details were adopted by a specific Act of Parliament. The Court of Justice has however given article 1(4) what might diplomatically be described as a purposive interpretation: in effect, "since" has been construed as meaning "provided that".
A Grand Chamber of the Court explained how article 1(4) was to be understood in Case C-43/10 Nomarchiaki Aftodioikisi Aitoloakarnanias  Env LR 27. After quoting article 1(4), the court continued:
In the present case, there is no dispute that the first of these conditions, as explained in greater detail by the court at paras 80-82 of its Nomarchiaki judgment, is capable of being satisfied through the proposed Parliamentary procedure. The appeal focuses upon the second condition. It is argued that the hybrid bill procedure is not capable of achieving the objectives of the EIA Directive.
The Court of Justice has considered the second condition on a number of occasions, and in its judgments has often repeated the same paragraphs, with minor variations. The most recent exposition is contained in the Nomarchiaki judgment. In paragraph 83, the court identified the fundamental objective of the EIA Directive:
The fundamental objective is thus the assessment of the environmental effects of projects before consent is given.
In paragraph 84, the court identified the basis upon which the assessment must be conducted:
The assessment must therefore be based upon appropriate information.
In paragraphs 85 and 86, the court explained how that approach applied to national legislatures:
The legislature must therefore have appropriate information at its disposal at the time when the project is adopted.
In paragraphs 88 and 89 the court made some additional observations:
The references in those paragraphs to article 1(5) refer to the provision in Directive 85/337 corresponding to article 1(4) of the codified EIA Directive. It is clear from paragraph 88 that article 1(4) requires a substantive legislative process, rather than the mere ratification of an administrative decision. In other words, the decision must in reality be that of the legislature: its role must not be merely formal. Paragraph 89 reiterates the requirement that appropriate information should be available to the members of the legislative body.
Finally, in relation to this judgment, the court made clear in paragraph 90, as it had in its previous judgments, the responsibility of national courts to apply these principles to the legislation passed by their national legislatures:
It is therefore for national courts, not the Court of Justice, to determine whether the conditions laid down by the court in that judgment have been satisfied in a particular national context: in particular, in relation to the second condition, whether the decision to adopt the project was the outcome of a substantive legislative process, and whether appropriate information was available to the members of the legislature at the time when the project was adopted.
In relation to the requirement that there should be the possibility of review by the national court, it is also relevant to note the Grand Chamber judgment in Case C-135/09 Boxus  ECR I-9711;  Env LR 14. In its judgment, the court stated at para 50 that neither the EIA Directive nor the Aarhus Convention applied to projects adopted by a legislative act satisfying the two conditions laid down by the court. The obligation imposed by article 11 of the Directive applied to other projects: that is to say, those adopted either by an act which was not legislative in nature or by a legislative act which did not fulfil those conditions (para 51). It follows that the reliance placed by the appellants in the present case upon article 11 of the Directive begs the question whether the proposed procedure would be incompatible with article 1(4): it is only if it is incompatible with that provision that article 11 applies.
At the same time, the Court of Justice also made it clear in its Boxus judgment that there must be the possibility of review of whether the conditions laid down by the court are satisfied. In that regard, the court stated at paras 54-55:
Articles 1(5) and 10a of Directive 85/337 correspond to articles 1(4) and 11 respectively of the EIA Directive in its codified form.
The acceptance by the Court of Justice in para 50 of its Boxus judgment that the EIA Directive does not apply to projects adopted by a legislative act satisfying the two conditions set by the court has important implications for the submissions made on behalf of the appellants in the present case. Those submissions took as their premise that Parliamentary procedure must comply with the requirements of the Directive, and in particular with the requirements of article 6. They appeared to overlook the fact that the whole point of article 1(4) is to exempt legislation falling within its scope from those requirements. Exemption from the requirements of the Directive cannot be conditional upon compliance with them: otherwise, there would be no exemption.
In particular, article 6(4) cannot apply to projects which are exempted from its requirements by article 1(4). As Advocate General Sharpston explained in the case of Boxus at point 56 of her opinion:
Where a decision is reached by a legislative process, however, such public participation already exists. The legislature itself is composed of democratically-elected representatives of the public. When the decision-making process takes place within such a body, it benefits from indirect, but nevertheless representative, public participation.
Is the application premature?
In my view it is appropriate to consider the appellants' contentions at the present stage, rather than waiting until legislation may have been enacted. In taking that view, I do not however accept all the arguments advanced in that regard on behalf of the appellants: in particular, the arguments based on articles 6(2) and 11(1) of the EIA Directive.
The principal advantages of considering the appellants' contentions at the present stage are practical. It is not in dispute that the Parliamentary procedure will be costly and time-consuming. It is plainly convenient to have the point of law as to the effect of the EIA Directive, as a matter of EU law, decided before further time and expense are incurred on the basis, if the appellants are correct, of a mistaken understanding by Government.
I am mindful of the importance of refraining from trespassing upon the province of Parliament or, so far as possible, even appearing to do so. The court can however consider the effect of the Directive under EU law without in my opinion affecting or encroaching upon any of the powers of Parliament. The Parliamentary authorities have not thought it necessary to seek to intervene in these proceedings, although the court was told that they have been kept informed of the parties' cases. No bill or draft bill has been placed before the court. Nothing the court does or says at this stage will affect the supremacy of Parliament in respect of any bill presented to it; nor will it affect the power of the Secretary of State, or any other Member of Parliament, to present to Parliament whatever bill he thinks fit. Nor is it necessary for the court to express any view, let alone take any action, concerning any decision to lay any bill before Parliament or concerning Parliament's approving such a bill. The court can in my opinion resolve the issue raised by the appellants by performing its ordinary duty to interpret legislation. The only unusual feature is that the court is arguably doing so before any action falling within the ambit of that legislation has yet been taken.
As was observed by Sir John Donaldson MR in R v HM Treasury, Ex p Smedley  QB 657, 667, in relation to a challenge to a draft Order in Council:
In many, and possibly most, circumstances the proper course would undoubtedly be for the courts to invite the applicant to renew his application if and when an order was made, but in some circumstances an expression of view on questions of law which would arise for decision if Parliament were to approve a draft may be of service not only to the parties, but also to each House of Parliament itself.
The present case is of course concerned with a proposal to seek Parliamentary approval of a bill rather than a draft Order in Council; and there is in consequence a prior constitutional question, as I have explained, as to whether the issues of law raised by the appellants would indeed arise for decision in the event that the bill were to be enacted. In that respect, this case is distinguishable from Ex p Smedley, and from the analogous case of R v Electricity Commissioners, Ex p London Electricity Joint Committee (1920) Ltd  1 KB 171. The approach adopted in those cases nevertheless has much to recommend it in the present circumstances. Adopting the words of Younger LJ in the case of the Electricity Commissioners at p 213, the interference of the court in such a case as this, and at this stage, so far from being even in the most diluted sense of the words a challenge to its supremacy, may be an assistance to Parliament.
If, on the other hand, the appellants' contentions were not considered until after legislation had been enacted, those contentions would necessarily be directed against the compatibility with the Directive of the proceedings in Parliament which had led to the enactment. In particular, the appellants' arguments, if reflecting those which the court has heard in this appeal, would focus upon the adequacy of the consideration of the environmental information by Members of Parliament, and the factors which might have influenced the way in which they voted. Unless authorised by the legislation to undertake such scrutiny, the court would then have to consider the conflict between such contentions and long-established constitutional principles before such an argument could even be entertained. That is not to say that the possibility of a future challenge can be foreclosed, since the compatibility with the Directive of Parliamentary proceedings which have not yet been completed cannot be definitively determined in advance. Any future challenge could not however be based on contentions which had already been considered and rejected in the present proceedings.
The compatibility of the procedure envisaged with the EIA Directive
Turning then to the appellants' contentions, there is no doubt that the procedure by which the Secretary of State proposes to seek Parliamentary authorisation for the HS2 project is a substantive legislative process. Parliament's role is not merely formal. It will be asked to give its consent to a bill which may undergo amendment during its passage through Parliament, and not merely to give formal ratification to a prior administrative decision.
There is equally no reason to doubt at this stage that appropriate information will be available to the members of the legislature at the time when decisions are taken as to whether the project should be adopted: the procedures laid down in SOs 27A and 224A of the House of Commons, and in the corresponding standing orders of the House of Lords, are apt to ensure that such information is made available. In those circumstances, it is unnecessary for the purposes of this appeal to consider the question whether it can ever be constitutionally permissible for the courts to enquire into the adequacy of the information placed before Parliament during the passage of a bill.
The appellants did not seek to argue that appropriate information could not be made available. As I have explained, their primary objection was to the fact that the decision whether to approve the principal elements of the project would be subject to the whip and thus to party oversight. Although this was not spelled out, the implication of their argument is that a decision by Parliament would be compatible with the EIA Directive only if Members of Parliament were allowed a free vote, regardless of their party allegiance or of their membership of the Government.
There is however nothing in the case law of the Court of Justice to suggest that the influence of Parliamentary parties, or of Government, over voting in national legislatures is incompatible with article 1(4). As I have explained, the court has identified the two conditions that must be satisfied in order for the exemption conferred by article 1(4) to apply, and has left it to national courts to judge whether those conditions are met in particular circumstances. As I have explained in paras 98 and 99 above, there is no reason to doubt that those conditions are capable of being met in the present case.
One of the ideas underlying the submissions on behalf of the appellants appears to be that members of the legislature must act independently and impartially when voting on whether to approve a project falling within the scope of article 1(4) of the EIA Directive, rather than being influenced by Parliamentary party politics. That idea appears to me however to be based on a misunderstanding of the constitutional role of the legislature. In that regard, there may be some value in referring to domestic cases where analogous issues have been considered.
The case of Franklin v Minister of Town and Country Planning  AC 87, for example, concerned the decision of a government minister to confirm a draft new town order following a public local inquiry. One of the grounds on which the decision was challenged was that the minister could not consider the report and the objections without a pre-disposition to favour the confirmation of the draft order, since it took forward a government policy to which he was necessarily committed. That argument was rejected. The minister's decision-making function was not of a judicial or quasi-judicial character: the purpose of the report was to provide him with information, and the only question was whether he had genuinely considered the report and the objections when they were submitted to him. As Lord Thankerton explained, there is no universal rule requiring that decision-makers must possess the independence and impartiality required of a court or tribunal: it is necessary to take account of the constitutional position of the decision-maker, and of the nature of the decision.
A similar approach can be seen in more recent cases concerned with the role of government policy in decisions concerned with infrastructure projects and development control, such as Bushell v Secretary of State for the Environment  AC 75 and R (Alconbury Developments Ltd) v Secretary of State for the Environment, Transport and the Regions  UKHL 23;  2 AC 295. In the latter case, for example, Lord Hoffmann explained that "in a democratic country, decisions as to what the general interest requires are made by democratically elected bodies or persons accountable to them" (para 69), and that such a decision "is not a judicial or quasi-judicial act", but is "the exercise of a power delegated by the people as a whole to decide what the public interest requires" (para 74).
The interpretation of the EIA Directive is of course a question of EU law, and cannot be determined by principles of our domestic law. EU law nevertheless draws inspiration from the constitutional traditions of the member states; and the observations in these decisions reflect constitutional principles which apply in other member states besides the United Kingdom.
Furthermore, Parliamentary parties are recognised as playing a legitimate role in democratic decision-making in other member states besides the United Kingdom (see, for example, article 53a of the German Basic Law, and the Federal Constitutional Court's judgment of 10 May 1977 on the Weapons Act 1972, BVerfGE 44, 308, paras 35-37). Their role at European level is expressly recognised in article 10(4) of the Treaty on European Union, which provides that "political parties at European level contribute to forming European political awareness and to expressing the will of citizens of the Union". Article 12(2) of the Charter of Fundamental Rights of the European Union is in similar terms. The role of political parties in democratic decision-making at national level is no less important.
It would be surprising if the EIA Directive required the adoption of a radically different approach. The fundamental objective of the Directive is, as the Court of Justice has explained, to ensure that the environmental effects of projects are assessed before consent is given. The achievement of that objective requires that appropriate environmental information should be available for consideration before consent is given. It does not require that the decision whether to give consent should be influenced solely or decisively by that information.
In particular, the question whether it is in the public interest to proceed with a project of national importance, such as HS2, may be a matter of national political significance. It is partly for that reason that such decisions may be considered appropriate for determination by the national legislature rather than by the ordinary processes of development control. The national legislatures of the member states are of course political institutions, whose decisions are likely to be influenced, possibly decisively, by the policy of the dominant Parliamentary party or parties. Article 1(4) of the EIA Directive is nevertheless based on the premise that the objectives of the Directive can be achieved where the decision is made by a body of that kind. That is not difficult to understand: the influence of party and governmental policy does not prevent the members of national legislatures from giving careful and responsible consideration to the information, including environmental information, which is relevant to the matters that they have to decide.
The contention that the procedure currently envisaged by the Government will not permit an adequate examination of the environmental information to take place appears to me to be equally unpersuasive. I observe in the first place that there is nothing either in the text of article 1(4) of the EIA Directive, or in the exegesis of that text by the Court of Justice, to suggest that national courts are required not only to confirm that there has been a substantive legislative process and that the appropriate information was made available to the members of the legislature, but must in addition review the adequacy of the legislature's consideration of that information, for example by assessing the quality of the debate and examining the extent to which members participated in it. These are not matters which are apt for judicial supervision. Nor is there anything to suggest the inevitable corollary: that national courts should strike down legislation if they conclude that the legislature's consideration of the information was inadequate.
There is a further difficulty with the contention that EU law requires the internal proceedings of national legislatures to be subject to judicial oversight of this nature. The separation of powers is a fundamental aspect of most if not all of the constitutions of the member states. The precise form in which the separation of powers finds expression in their constitutions varies; but the appellants' contentions might pose a difficulty in any member state in which it would be considered inappropriate for the courts to supervise the internal proceedings of the national legislature, at least in the absence of the breach of a constitutional guarantee.
Against this background, it appears unlikely that the Court of Justice intended to require national courts to exercise a supervisory jurisdiction over the internal proceedings of national legislatures of the nature for which the appellants contend. There is in addition much to be said for the view, advanced by the German Federal Constitutional Court in its judgment of 24 April 2013 on the Counter-Terrorism Database Act, 1 BvR 1215/07, para 91, that as part of a co-operative relationship, a decision of the Court of Justice should not be read by a national court in a way that places in question the identity of the national constitutional order ("Im Sinne eines kooperativen Miteinanders zwischen dem Bundesverfassungsgericht und dem Europäischen Gerichtshof ... darf dieser Entscheidung keine Lesart unterlegt werden, nach der diese offensichtlich als Ultra-vires-Akt zu beurteilen wäre oder Schutz und Durchsetzung der mitgliedstaatlichen Grundrechte in einer Weise gefährdete ..., dass dies die Identität der durch das Grundgesetz errichteten Verfassungsordnung in Frage stellte").
Counsel for the appellants relied however upon a statement made by Advocate General Sharpston in the Boxus case at point 84:
In my view, in order to assess whether that has happened in any particular case, the national court will need to examine the following aspects ....
Similarly in the Nomarchiaki case Advocate General Kokott referred at points 136-137 to a requirement that the legislature "performs its democratic function correctly and effectively", and to the need to clarify whether the legislature "was able properly to examine and debate the environmental effects of the project". I observe however, first, that those statements were not endorsed by the Court of Justice, and secondly, that their focus is upon the ability of the legislature to examine and debate the proposed project, rather than upon a qualitative assessment of the legislature's actual consideration of the proposal.
In the present case, there is in any event no reason to suppose that Members of Parliament will be unable properly to examine and debate the proposed project. Although the environmental statement made available to Members of Parliament may be of a size which reflects the scale of the project and the complexity of its impact upon the environment, it can be expected to include a non-technical summary of the information, in accordance with the 2011 Regulations (which transpose, in this respect, Annex IV to the EIA Directive). That can be expected to include information about the reasons for choosing HS2 rather than the main alternatives, as required by Annex IV to the Directive. Members of Parliament can also be expected to be provided with a summary of the comments received on the environmental statement, prepared by an independent assessor, in accordance with SO 224A. That summary can be expected to encompass any comments made by the appellants which advance the case for their optimised alternative.
Members of Parliament can be expected to have that information well in advance of the second reading debate on the bill: as I have explained, the summary of the comments received must be submitted to the House at least 14 days prior to the bill's receiving its second reading; and it is implicit in SO 224A that the environmental statement must itself have been submitted at least three months or so earlier (since the public must be allowed a period of at least 56 days to comment on the statement, and the assessor must be allowed at least 28 days to prepare the summary).
It is in any event unrealistic for the appellants to focus solely upon the second reading debate, as if it were the only opportunity for Members of Parliament to consider the environmental information. Active political debate on the HS2 project, including its environmental impact, has already been under way for some time, and it is reasonable to expect that Members of Parliament have been, and will continue to be, contacted about it by their constituents and lobbied by interested organisations, such as the appellants. As the bill proceeds through Parliament, and political interest in the project becomes more intense, Members of Parliament will have even more reason to be, and to wish to be, well informed about the project. As counsel for the respondents observed in relation to the opportunities for Members of Parliament to consider and discuss the proposal, the second reading debate is in reality the tip of the iceberg.
Without therefore considering the fundamental constitutional objection to this line of argument – that the court would be presuming to evaluate the quality of Parliament's consideration of the relevant issues, during the legislative process leading up to the enactment of a statute – I conclude that the argument is based on an incorrect interpretation of the EIA Directive, and is in addition unsupported by the evidence as to the procedure which might be followed.
I do not consider that the court is under any obligation to make a preliminary reference to the Court of Justice before reaching a decision on this matter. The Court of Justice has already given a clear account of the relevant principles: the passages which I have cited from its Nomarchiaki judgment have been repeated time and again, in judgments which include two given by a Grand Chamber of the court (in Nomarchiaki itself, and in the earlier case of Boxus  ECR I-9711,  Env LR 14). The court has made it clear that it expects national courts to apply the principles which were established in those judgments in the context of their own national systems.
For these reasons, and those given by Lord Carnwath and Lord Sumption, I would dismiss the appeals. I wish also to express my agreement with the additional observations of Lord Neuberger and Lord Mance.
(with whom Lord Neuberger, Lord Mance, Lord Kerr and Lord Reed agree)
I agree that this appeal should be dismissed for all the reasons given by Lord Carnwath and Lord Reed. I also agree with the additional observations of Lord Neuberger and Lord Mance. My purpose in adding yet another judgment is to explain why, like Lord Carnwath, I regard the proposition that the Government's command paper falls outside the scope of the SEA Directive as acte clair and as such unsuitable for a preliminary reference to the Court of Justice of the European Union.
The starting point is that the SEA Directive plainly does not require an environmental assessment to be carried out for all "plans or programmes" whose implementation would have a major impact on the environment. Even on the footing that a plan or programme is required (or regulated) by legislative, regulatory or administrative provisions within Article 2(a) and has a "significant environmental effect" within Article 3.1, an environmental assessment is still not required unless the plan or programme in question "set[s] the framework for future development consent" within Article 3.2(a).
The rationale for this is straightforward. It is common for development consent for specific projects to be affected by modern schemes of development control at different levels of generality. For example, in England planning policies may currently be laid down by the National Planning Policy Framework or local development plans or by national policy statements for nationally significant infrastructure projects. Areas may be zoned for more or less intensive standards of planning control, or for particular types of development or none, or for the application by planning authorities of special criteria in defined areas. The legal effect of these general policies may be weaker or stronger. In some cases development consent must be given or refused in accordance with the policy subject to limited exceptions, while in others the obligation of the planning authority is only to have regard to or take account of it. Similar approaches to planning policy can be found in other European countries.
The effect of the EIA Directive is that subject to limited exceptions an environmental impact assessment is required before development consent can be granted for any specific project of a kind specified in the Annexes which is likely to have a significant environmental impact. The effect of the SEA Directive is that where the grant or refusal of development consent for a specific project is governed by a policy framework regulated by legislative, regulatory or administrative provisions, the policy framework must itself be subject to an environmental assessment. The object is to deal with cases where the environmental impact assessment prepared under the EIA at the stage when development consent is granted is wholly or partly pre-empted, because some relevant factor is governed by a framework of planning policy adopted at an earlier stage.
None of this means that the only policy framework which counts is one which is determinative of the application for development consent, or of some question relevant to the application for development consent. What it means is that the policy framework must operate as a constraint on the discretion of the authority charged with making the subsequent decision about development consent. It must at least limit the range of discretionary factors which can be taken into account in making that decision, or affect the weight to be attached to them. Thus a development plan may set the framework for future development consent although the only obligation of the planning authority in dealing with development consent is to take account of it. In that sense the development plan may be described as influential rather than determinative. But it cannot be enough that a statement or rule is influential in some broader sense, for example because it presents a highly persuasive view of the merits of the project which the decision maker is perfectly free to ignore but likely in practice to accept. Nor can it be enough that it comes from a source such as a governmental proposal or a ministerial press statement, or a resolution at a party conference, or an editorial in a mass circulation newspaper which the decision-maker is at liberty to ignore but may in practice be reluctant to offend.
All of this is inherent in the concept of a "framework" and in the purpose of the Directive. It is consistent with the requirement of Article 2(a) that the plan or programme must be regulated by legislative, regulatory or administrative provisions, for whatever may precisely be meant by that, it clearly indicates a degree of prescription. It corresponds to the test adopted by the CJEU in Terre Wallonne ASBL and Inter-Environnement Wallonie ASBL v Région Wallonne (Joined Cases C-105/09 and C-110/09)  ECR I-5611 at para 55 ("contains measures compliance with which is a requirement for issue of the consent"); in Inter-Environnement Bruxelles ASBL v Région de Bruxelles-Capital (Case C-567/10)  2 CMLR 30 at para 30 ("define the criteria and the detailed rules for the development of land and normally concern a multiplicity of projects whose implementation is subject to compliance with the rules and procedures provided for by those measures"). That test was adopted and restated by the Grand Chamber in Nomarchiaki Aftodioikisi Aitoloakarnanias v Ipourgos Perivallontos Khorotaxias kai Dimosion Ergon (Case C-43/10)  Env. LR 2453, at para 95 ("defines criteria and detailed rules for the development of land and which subjects implementation of one or more projects to rules and procedures for scrutiny"). The two last-named cases considered the point in the context of an issue about the definition of plans and programmes in Article 2(a). But the required content of a plan or programme was clearly relevant to that definition, and it was to this that the Court was referring. These statements echo the language of the Commission's Guidance paper, Implementation of Directive 2001/42 on the assessment of the effects of certain plans and programmes on the environment (2001) at paras 3.5 and 3.6, where Article 3.2(a) is described, at para 3.23, as referring to a plan or programme which:
contains criteria or conditions which guide the way the consenting authority decides an application for development consent. Such criteria could place limits on the type of activity or development which is to be permitted in a given area; or they could contain conditions which must be met by the applicant if permission is to be granted; or they could be designed to preserve certain characteristics of the area concerned (such as the mixture of land uses which promotes the economic vitality of the area).
The main reason why the command paper cannot require an environmental assessment under the SEA is that it is nothing more than a proposal. Naturally, the fact that it is a government proposal and appears in a command paper makes it influential in the broader sense that I have mentioned above. It means it is politically more likely to be accepted. But the command paper does not operate as a constraint on the discretion of Parliament. None of the factors which bear on the ultimate decision whether to pass the hybrid bill into law have been pre-empted, even partially. I accept that this means that governments may in some cases be able to avoid the need for an environmental assessment by promoting specific legislation authorising development. But that is not because the SEA has no application to projects authorised in that way. It is because (i) the SEA Directive does not require member states to have plans or programmes which set the framework for future development consent, but only regulates the consequences if they do; (ii) where development consent is granted by specific legislation there are usually no plans or programmes which set the framework for that consent; and (iii) legislative grants of development consent are exempt from the EIA Directive by virtue of Article 1(4), subject to conditions which replicate some of the benefits of a requirement for an environmental impact assessment, and which like every other member of the Court, I consider to be satisfied by the proposed hybrid bill procedure.
We have not heard argument on the possibility floated by Lady Hale that the hybrid bill itself, if passed, might "set the framework for future development consent". It is not an issue on this appeal and it would be premature to raise it at a stage when the bill has only recently been published and the final form of any legislation is not known. I therefore express no final view on the point. But as at present advised, I think it clear that the bill if passed will not set the framework for future development consent. Clause 19 deems planning permission to be granted and authorises the development. An Act in these terms would not be part of the process by which the development consent is granted. It would be the ultimate decision. It would not set the framework for future development consent, because it would itself be the development consent.
The decision whether to make a preliminary reference under Article 267 of the Treaty on the Functioning of the European Union is for the national court alone. As the court of final appeal for the United Kingdom, the Supreme Court must make a reference unless "the correct application of Community law [is] so obvious as to leave no scope for any reasonable doubt as to the manner in which the question raised is to be resolved": Srl CILFIT v Ministry of Health  ECR 3415 at para 16. A proposition may be obvious because on a question of interpretation the meaning of the text is beyond reasonable dispute, or because the CJEU has decided the question. In formal Recommendations issued following upon the adoption of new rules of procedure in September 2012 (OJ C338/1, 6.11.2012), the Court of Justice put the matter in this way:
The question whether the command paper is within the scope of the SEA is in my view wholly unsuitable for a preliminary reference to the CJEU. In the first place, although there may be room for argument about what constitutes a plan or programme "required by legislative, regulatory or administrative provisions", article 3.2(a) of the SEA Directive is couched in plain, untechnical language. The concept of setting a "framework for future development consent" is perfectly straightforward against the undisputed background that modern systems of development control commonly lay down criteria for granting development consent for specific projects. Second, in two recent decisions, one being by a Grand Chamber, the CJEU has ruled that the Directive covers measures which define the criteria and detailed rules for development. It is now for national courts to apply the law as the CJEU has declared it. The relevant statements of principle are brief, as they commonly are in judgments of the CJEU. But there is nothing unclear about them, and nothing in the context of the command paper which makes the application of the test uncertain or problematical. No purpose would be served by referring to the CJEU the question whether they really meant it. Third, even if we were to refer that question, the nature and status of a command paper of this kind is a question of national law and practice. The outcome of such a reference would be determined by our finding that the command paper was no more than a proposal and not a framework for decision-making.
On the question whether to refer the question of the application of the EIA Directive to the hybrid bill procedure. I am content like the rest of the Court to adopt the observations of Lord Reed.
I have not found this an easy case. HS2 will be the largest infrastructure project carried out in this country since the development of the railways in the 19th century. Whatever the economic and social benefits it may bring, it will undoubtedly have a major impact upon the environment. There has never been a full environmental assessment of HS2 as against the alternative ways of developing the railway system, including ways which do not involve constructing new railway lines capable of carrying trains travelling at 250 miles per hour, such as the so-called "optimised alternative" favoured by the 51M, the body to which the local authorities involved in this case (and others) belong.
One might have thought that it was the object of Directive 2001/42/EC, commonly called the Strategic Environmental Assessment Directive (the "SEA Directive"), but actually entitled "on the assessment of the effects of certain plans and programmes on the environment", to ensure that such an assessment took place. Recitals (4) and (5) to the SEA Directive say this:
As Advocate General Kokott explained in her opinion in Terre wallonne ASBL and Inter-Environnement Wallonie ASBL v Région Wallonie (Joined Cases C-105/09 and C-110/09)  I-ECR 5611, the environmental assessment for which the SEA Directive provides "is .... carried out as part of decision-making procedures which precede the procedures for granting consent for individual projects, but may affect them" (para 2). Experience with Directive 85/337/EEC, the Environmental Impact Assessment Directive (the "EIA Directive"), had shown that, at the time when projects came to be assessed, major effects upon the environment had already been established on the basis of earlier planning measures. It was "therefore appropriate for such effects on the environment to be examined at the time of preparatory measures and taken into account in that context" (para 32). She gave this example, which has a clear resonance with this case (para 33):
An abstract routing plan, for example, may stipulate that a road is to be built in a certain corridor. The question whether alternatives outside that corridor would have less impact on the environment is therefore possibly not assessed when development consent is subsequently granted for a specific road-construction project. For this reason, it should be considered, even as the corridor is being specified, what effects the restriction of the route will have on the environment and whether alternatives should be included.
If a strategic environmental assessment is required, the SEA Directive stipulates a report "in which the likely significant effects on the environment of implementing the plan or programme, and reasonable alternatives taking into account the objectives and the geographical scope of the plan or programme, are identified, described and evaluated" (article 5). This has to be carried out during the preparation of the plan or programme and before its adoption (article 4). This evaluation of alternatives is of a different order from that required for projects covered by the EIA Directive, which only requires an outline of the main alternatives studied by the developer and an indication of the main reasons for the choice. Ouseley J decided that, if an SEA was required for the HS2 scheme as a whole, it has not yet been done, and the Court of Appeal agreed. This is now common ground between the parties. However, Ouseley J also decided that such a report would not have to cover non-high speed alternatives as well as alternative high speed routes and that decision is not under appeal. It would appear, therefore, that even if successful, this litigation would not succeed in achieving what the parties really want.
It is clear that the Directive does not require a strategic environmental assessment for all schemes which are likely to have significant effects upon the environment before they are formally adopted. It would have been so much simpler if it did. But, as its title and recital (4) say, it only applies to "certain plans and programmes". Two requirements are relevant for our purposes. First, the scheme must fall within the definition of "plans and programmes" in article 2(a). As far as relevant to this case, this reads:
'plans and programmes' shall mean plans and programmes .... as well as any modifications to them
Secondly, the relevant provision requiring an environmental assessment is that contained in article 3(2)(a):
.... an environmental assessment shall be carried out for all plans and programmes (a) which are prepared for .... transport .... and which set the framework for future development consent of projects listed in Annexes I and II to Directive 85/337/EEC.
Construction of lines for long distance railway traffic is one of the projects listed in Annex I to Directive 85/337/EEC, now consolidated in Directive 2011/92/EU.
This litigation is only concerned with the Government's command paper, High Speed Rail: Investing in Britain's Future – Decisions and Next Steps (2012, Cm 8247) (the "DNS"). The appellants argue that it was a plan or a programme and that it set the framework for future development consent of a project for the construction of lines for long distance railway traffic. However, the DNS is only one part of a long and complex process and it is entirely possible that no part of that process constitutes a plan or programme within the meaning of the Directive.
For our purposes, the process began with the Command Paper, High Speed Rail (2010, Cm 7827), published by the Labour Government in March 2010. This set out the twin goals of "new capacity and improved connectivity", to be achieved sustainably, without unacceptable environmental impacts; declared that "high speed rail is the most effective way to achieve these goals, offering a balance of capacity, connectivity and sustainability benefits unmatched by any other option"; and announced a Y shaped network, linking London to Birmingham, Manchester, the East Midlands, Sheffield and Leeds, connecting onto existing tracks going further north. Formal public consultation on the detailed recommendations for the first phase, from London to Birmingham, on the strategic case for high speed rail, and on the core scheme, would begin in the autumn of 2010. A further consultation would take place in 2012 on the detailed proposals for phase 2. Only after both consultations were completed would the Government make its final decisions, with a view to promoting a hybrid bill to cover the whole of the Y shaped network.
When the coalition Government came into power following the general election in May 2010, it declared that it was committed to taking forward high speed rail, but that this would have to be achieved in stages. Formal public consultation on the matters proposed in the March 2010 command paper was opened in February 2011. The Government promised to announce the outcome and its final decisions on its strategy for high speed rail before the end of 2011. The DNS was in fact published in January 2012. This reported the Government's conclusions, both as to the high speed rail strategy and as to the detailed proposals for phase 1 of the route. But instead of a single hybrid bill covering the whole of the core network, it was now proposed to have two bills, the first for phase 1 and the second for phase 2. Steps would also be taken to safeguard the phase 1 route from incompatible development by a Direction under article 25 of the Town and Country Planning (Development Management Procedure) (England) Order 2010 (SI 2010/2184). Following consultation, this was done in July 2013. Public consultation on the preferred route for phase 2 also began in July 2013. The proposals are broadly in line with the strategy set out in the DNS. A hybrid bill for phase 2 is proposed to follow the general election in May 2015.
The hybrid bill for phase 1 received its first reading in the House of Commons on 25 November 2013. As expected, it does three main things. First, it gives legal authority to carry out the works, which are described in detail in schedule 1. Second, it gives the necessary planning consents to carry out the development, subject to a few details which are left to local planning authorities. And third, it authorises the Secretary of State compulsorily to purchase or acquire other rights over the land needed to complete the works, modifying the usual procedures for doing so. The bill is accompanied by a massive environmental statement (apparently designed to comply with the requirements of the EIA Directive, should this be necessary).
We are being asked in this case to consider whether there should have been an SEA before the DNS was adopted. This raises two questions. First, was the DNS "required by legislative, regulatory or administrative provisions" within the meaning of article 2(a)? Second, did it "set the framework for future development consent of" a project for the construction of lines for long distance railway travel, within the meaning of article 3(2)(a)?
In Inter-Environnement Bruxelles ASBL v Région de Bruxelles-Capitale (Case C-567/10)  2 CMLR 30, the European Court of Justice (4th Chamber) held that "required" included "plans and programmes whose adoption is regulated by national legislative or regulatory provisions, which determine the competent authorities for adopting them and the procedure for preparing them" (para 31). As Lord Neuberger and Lord Mance point out, this was an expansive interpretation of a phrase which had been introduced to limit the scope of the plans and programmes covered by the obligation. But I do not find either word, "required" or "regulated", easy to apply to the situation we have here.
A massive scheme such as this cannot take place without comprehensive authorisation of the works, the land development they entail, and the compulsory purchase and other measures which will be required to do them. It is obviously completely impracticable to undertake such a scheme depending upon multiple individual applications for planning and related consents, together with the use of ordinary compulsory purchase powers. As Sullivan LJ pointed out in the Court of Appeal, "development consent for a nationally important project such as HS2 could have been sought by way of the development consent procedure for nationally significant infrastructure projects under the Planning Act 2008, or by way of an order for a scheme of national significance under the Transport and Works Act 1992". As he also observed, the Government accepted that if either of these procedures had been adopted, "the DNS would have 'set the framework' because it would have been a material consideration which the decision-maker under the 2008 Act or the 1992 Act would have been under a legal obligation to take into account" (para 151). He found it difficult to accept that the fact that the Government has chosen instead to adopt the hybrid bill procedure makes all the difference. I have considerable sympathy with him.
Can it be said that the command paper which preceded the Bill was "required" or "regulated" by our administrative provisions? It is, of course, how one would expect such a scheme to proceed – with a succession of government proposals, consultations and conclusions. A complex Bill like this does not suddenly spring onto the Parliamentary stage without any prior consultation with the public. Formal consultations such as this are governed by the Government's Code of Practice on Consultation. The 2010 command paper, High Speed Rail, promised formal consultations on each phase, followed by decisions and a hybrid bill. The only change since then has been to split the scheme into two phases. In that sense, the 2010 command paper did "determine the competent authorities for adopting them and the procedure for preparing them", but of course it did not do so in any legally binding sense.
In the Court of Appeal, both the majority and the minority judgments regarded the two questions as inter-linked. Both agreed that, if the DNS did indeed "set the framework" within the meaning of article 3(2)(a), it would be difficult to say that it was not "required by administrative provisions" for the purpose of article 2(a):  EWCA Civ 920;  PTSR 1194, paras 71 and 180. This is akin to the process of reasoning adopted by the CJEU adopted in Inter-Environnement Bruxelles ASBL v Bruxelles: because the Court regarded the measure as setting the framework, they reasoned backwards that it was regulated even if not legally required and thus fell within article 2(a).
But in any event did the DNS "set the framework"? It is clear from the Terre wallonne case, that a scheme adopted by national legislation can indeed be a "plan" or a "programme" within the meaning of article 2(a). The debate in that case was about how binding its provisions had to be upon subsequent planning decisions. Advocate General Kokott pointed out that "plans and programmes may, however, influence the development consent of individual projects in very different ways and, in so doing, prevent appropriate account from being taken of environmental effects. Consequently the SEA Directive is based on a very broad concept of 'framework'" (para 64). The term had to be construed "flexibly. It does not require any conclusive determinations, but also covers forms of influence that leave room for some discretion" (para 65). She concluded (para 67):
To summarise, it can therefore be said that a plan or programme sets a framework in so far as decisions are taken which influence any subsequent development consent of projects, in particular with regard to location, nature, size and operating conditions or by allocating resources.
In Walton v Scottish Ministers  UKSC 44,  PTSR 51, para 17, Lord Reed quoted paragraphs 64 and 65 of Advocate General Kokott's opinion. He also pointed out that, in cases where an SEA is not automatically required by article 3(2)(a), but may be required by article 3(3) or 3(4), Member States are required by article 3(5) to take into account the criteria set out in Annex II. These include "the degree to which the plan or programme sets a framework for projects and other activities, either with regard to the location, nature, size and operating conditions or by allocating resources." It is implicit in this, he observed, that a framework can be set without the location, nature, or size of the projects being determined. This supports the view that influence will do.
The actual question in Terre wallonne was whether action programmes required by Directive 91/676/EEC "concerning the protection of waters against pollution caused by nitrates from agricultural sources" were "plans and programmes" for the purpose of the SEA Directive. Neither the Advocate General nor the Court of Justice had any difficulty in deciding that they were. Hence the Court did not address itself to the degree of influence upon later decisions which the plan or programme had to have. It answered the question thus (para 55):
.... an action programme adopted pursuant to article 5(1) of Directive 91/676 is in principle a plan or programme covered by article 3(2)(a) of Directive 2001/42 since it constitutes a 'plan' or 'programme' within the meaning of article 2(a) of the latter directive and contains measures compliance with which is a requirement for issue of the consent that may be granted for carrying out projects listed in Annexes I and II to Directive 85/337.
Nor did the Grand Chamber answer the question directly in Nomarchiaki Aftodioikisi Aitoloakarnanias (Case C-43/10)  Env LR 27. It quoted a phrase used by the Court in Inter-Environnement Bruxelles ASBL v Bruxelles, when rejecting the submissions of the Belgian, Czech and UK Governments that the Directive did not apply to measures which were not required by rules of law (para 30):
That interpretation would thus run counter to the Directive's aim of establishing a procedure for scrutinising measures likely to have significant effects on the environment,which define the criteria and the detailed rules for the development of land and normally concern a multiplicity of projects whose implementation is subject to compliance with the rules and procedures provided for by those measures.
Thus in Nomarchiarcki, the Grand Chamber stated that (para 95):
It is not evident that the project concerned constitutes a measure which defines criteria and detailed rules for the development of land and which subjects implementation of one or more projects to rules and procedures for scrutiny.
It is, however, obvious that both the Chamber in Inter-Environnement Bruxelles and the Grand Chamber in Nomarchiaki were addressing the question of whether the project fell within the definition of a "plan and programme" in article 2(a). They were not directly addressing the further question of whether it "set the framework" within the meaning of article 3(2)(a) at all. Rather, the Court in Inter-Environnement Bruxelles was reasoning that certain plans and programmes which clearly did set the framework for the purpose of article 3(2)(a) in the sense described would be excluded if a narrow view were taken of "required" for the purpose of article 2(a). The upshot, it seems to me, is that the Court has not yet adopted the expansive view of setting the framework proposed by Advocate General Kokott in Terre wallonne. But neither has it expressly rejected it, because the question simply did not arise in that case.
Nevertheless, it could be said that the words used in Nomarkchiaki are consistent with what the Court did say in Terre wallonne, when addressing the "framework" question (emphasised at para 147 above). It could also be said that, when the Grand Chamber referred in Nomarchiaki to the "Directive's aim" it must have been referring to recitals (10) and (11), each of which refers to setting "a framework for future development consent of projects". This lends support to the view, so cogently explained by Lord Sumption, that the "framework" has to be one which those granting such consent must at the very least take into account.
But even if we do regard the emphasised words in both cases as an exhaustive definition, I originally found it hard to see how they would not apply to the hybrid Bill. Once passed, it would be at the end of the spectrum, referred to by the Court of Appeal in the passage quoted by Lord Carnwath at para 29 above, where the plan or programme "conclusively determines whether consent is given and all material conditions" (their para 54). Article 2(a) expressly contemplates that a plan or programme may not only be prepared and adopted by a national, regional or local authority, but may be prepared by an authority for adoption through a legislative procedure by Parliament or Government. In Terre wallonne, Advocate General Kokott explained that this did not mean that every possible law should be subject to an environmental assessment. The requirement that it be "required" meant that "freely taken political decisions on legislative proposals are not therefore subject to the obligation to carry out assessments" (para 41). But that requirement is at least arguably fulfilled in our case.
This litigation is solely about the DNS. We have not heard any argument about the Bill itself. There may be answers to the question of the Bill, as there so often are when issues are exposed to the glaring light of adversarial argument. The most obvious one is that the Act will do more than "set the framework": it will grant the necessary planning consents, and a great deal more, itself. It may well be, therefore, that we leap straight from a mere Government proposal in the DNS to the full blown development consent in the Act without anything which "sets the framework" in between.
Hence I have considerable sympathy with Sullivan LJ (para 164):
An interpretation of 'framework' in article 3(2)(a) which would enable the governments of member states to carve out an exemption from the SEA Directive for those projects for which they choose to obtain development consent by 'specific acts of national legislation' would be contrary to the purposive approach to the interpretation of the Directive adopted by the Court of Justice in the Terre wallonne and Inter-Environement Bruxelles cases.
I also think that the disagreement in the Court of Appeal was more properly characterised by him as a disagreement as to (para 188)
whether the fact that a member state chooses to adopt a process of granting development consent for a major project which will have a significant environmental effect by way of an act of national legislation is sufficient, of itself, to place the Government's adoption of its plan or programme outwith the scope of the European-wide strategic environmental protection conferred by the SEA Directive.
I was therefore attracted to the suggestion that we should refer a question along the following lines to the CJEU: "are plans and programmes which 'set the framework' for the purpose of article 3(2(a) of the SEA Directive limited to those 'which define criteria and detailed rules for the development of land' or may they include measures which will have a powerful but not necessarily constraining or determinative effect upon the ultimate decision-maker?" This is not the same as asking the CJEU whether the DNS is such a plan or programme. I entirely agree that that is a matter for us. The question is what test we should adopt when considering it.
I have, however, been persuaded that such a reference is unnecessary, for four main reasons. First and foremost, however briefly, the Grand Chamber in Nomarchiaki has adopted a definition which is consistent with the aims of the Directive as set out in the Recitals. Secondly, the aim of the Directive is not to ensure that all development proposals which will have major environmental effects are preceded by a strategic environmental assessment; rather, it is to ensure that future development consent for projects is not constrained by decisions which have been taken "upstream" without such assessment, thus pre-empting the environmental assessment to be made at project level. Thirdly, it is by no means clear that quashing the DNS would de-rail the Bill process, at least now that the Bill has received its first reading; we have no power to prevent Parliament from considering it; it is a matter for Parliament whether there is a sufficient risk that the Act itself would fall within the Directive to justify taking the precautionary step of commissioning a strategic environmental assessment at this stage (we are told that it would take six to 12 months to complete). Fourthly, however, it has been decided that such an assessment need not do what many, if not all, of the parties and the objectors would wish it to do, which is to compare the environmental effects of HS2 with those of the alternative methods of increasing capacity and connectivity on our railways. Mere delay is not a good reason for making a reference (although some might think it a good reason for asking for one).
I have not, however, entertained similar doubts in relation to the EIA Directive. In that respect I am in full agreement with the judgment of Lord Reed. But the conclusion is that I, too, would dismiss these appeals.
Lord Neuberger and Lord Mance
(with whom Lady Hale, Lord Kerr, Lord Sumption, Lord Reed and Lord Carnwath agree)
We agree that these appeals should be dismissed.
The issues arising under the Directives
However, the issues that have had to be addressed only arise as a result of decisions of the European Court of Justice, which we have found problematic and which call for some further observations.
The first decision concerns the word "required" in the phrase "required by legislative, regulatory or administrative provisions" in article 2(a) of the SEA Directive.
The Fourth Chamber of the Court of Justice has on 22 March 2012 held, disagreeing with Advocate General Kokott's Opinion, that "required" means, not required, but "regulated" (Inter-Environnement Bruxelles Asbl v Région de Bruxelles-Capitale Case C-567/10,  2 CMLR 30).
The second set of decisions concerns the exclusion in article 1(5) of the Directive 85/337/EC (amended by Directive 2003/35/EC) - the predecessor of article 1(4) of the EIA Directive - of
projects the details of which are adopted by a specific act of national legislation, since the objectives of this Directive, including that of supplying information, are achieved through the legislative process.
The Court has held that this word "since" means, in effect "provided that" - so that the exclusion only operates subject to satisfaction of a condition that "the objectives of the Directive, including that of supplying information, must be achieved through the legislative purpose" (World Wildlife Fund (WWF) v Autonome Provinz Bozen Case C-435/97,  ECR I-5613, para 57, Nomarchiaki Aftodioikisi Aitoloakarnanias Case C-43/10  Env LR 453, para 79) and only applies "where the legislative process has enabled the objectives pursued by the Directive .... to be achieved" (State of the Grand Duchy of Luxembourg v Linster Case C-287/98  ECR I-6917, para 59). We return to the implications below.
The constitutional basis of European Union legislation
Under the European Treaties, the Council of Ministers, now acting jointly with the European Parliament, serves as the European legislator; the Commission proposes legislative measures and oversees the application of European law; and the Court of Justice is charged to "ensure that, in the interpretation and application of the Treaties, the law is observed" and to give rulings on the interpretation of Union law as well as the validity of acts adopted by the European institutions: articles 220 and 234 of the pre-Lisbon EC Treaty and now articles 16 and 19(1) TEU and article 267 TFEU.
The Council of Ministers is composed of representatives of the elected national governments of Europe, and the European Parliament adds an additional democratic element to the enactment of European legislation. It was and is at the heart of the Community and now Union's legitimacy that decisions reached by the Council of Ministers, in conjunction with the European Parliament, are given effect.
The principle of legal certainty is also a fundamental principle of European law: Edward and Lane on European Union Law, para 6.134, citing R (International Association of Independent Tanker Owners (Intertanko)) v Secretary of State for Transport Case C-308/06  2 Lloyd's Rep 260, para 69 where the Court said:
The general principle of legal certainty, which is a fundamental principle of Community law, requires, in particular, that rules should be clear and precise, so that individuals may ascertain unequivocally what their rights and obligations are and may take steps accordingly (see [Belgium v Commission (Case C-110/03)  ECR I-2801, para 30, and IATA and ELFAA (Case C-344/04)  ECR I-403, para 68]).
The principle is "part of the legal order of the Community", now the Union, as the Court said in Deutsche MilchKontor GmbH v Federal Republic of Germany Joined Cases 205-215/82  ECR 2633, para 30. Union citizens and others need to know and are entitled to expect that the legislation enacted by their European legislator will be given its intended effect.
That does not exclude the vital role of case-law in interpreting legislation. But interpretation is only necessary when legislation, construed in the light of its language, context and objectives, is unclear. A national court is required under European law to refer to the Court of Justice any question of interpretation unless it reaches the conclusion that the answer is
so obvious as to leave no scope for any reasonable doubt as to the manner in which the question raised is to be resolved (CILFIT Srl v Ministry of Health (Case 283/81)  ECR 3415, para 16).
The well-known principles are stated in CILFIT and were reiterated in Junk v Kühnel (Case C-188/03)  ECR I-885.
The Court of Justice warned national courts in CILFIT that:
before coming to such a conclusion, the national court or tribunal "must be convinced that the matter is equally obvious to the courts of the other Member States and to the Court of Justice" (para 16),
"the existence of such a possibility must be assessed on the basis of the characteristic features of Community law and the particular difficulties to which its interpretation gives rise" (para 17),
"To begin with, it must be borne in mind that Community legislation is drafted in several languages and that the different language versions are all equally authentic. An interpretation of a provision of Community law thus involves a comparison of the different language versions" (para 18),
"It must also be borne in mind, even where the different language versions are entirely in accord with one another, that Community law uses terminology which is peculiar to it. Furthermore, it must be emphasized that legal concepts do not necessarily have the same meaning in Community law and in the law of the various Member States" (para 19).
The Court concluded with a more general observation about the principles according to which it will interpret European legislation (para 20):
Finally, every provision of Community law must be placed in its context and interpreted in the light of the provisions of Community law as a whole, regard being had to the objectives thereof and to its state of evolution at the date on which the provision in question is to be applied.
In Edward and Lane on European Union Law, 3rd Ed (2013), the authors, after setting out the passages quoted in paras 167(iii) and (iv) above, continue as follows (para 6.24):
In the event of discrepancy, real or apprehended, the court will therefore have recourse to all versions, within the context of the purpose and general scheme of the rules of which they form part, in order to determine their correct (and uniform) meaning. To assist it will also have recourse to travaux préparatoires (of legislation – none exists for the Treaties) to a much greater extent than is normally the case in national law. For these reasons, a literal interpretation of Union law texts is inappropriate.
The starting point in interpretation is therefore the different language versions of the text, to understand their purpose and scheme. But regard must be had to the objectives of the relevant provisions of Union law as a whole and its state of evolution, and recourse to the travaux préparatoires may be an important aid to identification of the correct meaning.
It is a common place in legislation that objectives may not be fully achievable or achieved. Compromises or concessions have to be made if legislators are to achieve the enactment of particular provisions. This is perhaps especially so at the international European level, in the case of measures agreed by the Council of Ministers where different Member States may only have been prepared to go part of the way with a Commission proposal (or Parliamentary proposal for amendment) and qualifications may have to be introduced to arrive at any agreement. The structure of the European Union involves a balance of interests which must be respected if the structure is to be stable.
When reading or interpreting legislation, it can never therefore be assumed that particular objectives have been achieved to the fullest possible degree. Limitations on the scope or application of a legislative measure may have been necessary to achieve agreement. There may also have been good reasons for limitations, of which courts are unaware or are not the best judge. Where the legislature has agreed a clearly expressed measure, reflecting the legislators' choices and compromises in order to achieve agreement, it is not for courts to rewrite the legislation, to extend or "improve" it in respects which the legislator clearly did not intend.
There are important practical consequences, if citizens and other users of the law cannot be confident that European legislation will be given its intended and obvious effect. First, there is a risk of loss of confidence at national level in European Union law, and a risk of impairment of the all-important dialogue between national courts and the Court of Justice, with its vital role of interpreting and consolidating the role of European law.
Second, national courts will find it much more difficult to decide whether a point of EU law is acte clair or not. This would lead to the risk of the Court of Justice's already heavy caseload becoming over-loaded with references, and many more cases where the parties have to face the delay and expense of a reference.
Third, it will make more difficult the drafting of Directives and Regulations, and, still more importantly, make it more difficult in future to achieve agreement upon such measures. Where a member state has, like the United Kingdom in some areas, the right not to opt into a measure, unpredictability about the meaning which might be attached to it may also encourage caution about opting in.
The SEA Directive
In Inter-Environnement Bruxelles Asbl v Région de Bruxelles-Capitale Case C-567/10,  2 CMLR 30, the Opinion prepared by Advocate General Kokott contains a careful analysis of the different language versions of the SEA Directive, of the legislative history and travaux préparatoires and of the legislative intent [AG14-AG30]. On that basis, she concluded that the word "required" meant what it says, that is that "it covers only plans or projects which are based on a legal obligation". She also cited logical reasons for the Council of Ministers' decision to that effect. She was supported in her conclusion by all three Governments (Belgian, Czech and United Kingdom) represented before the Court, as well as by the European Commission: see the Court's judgment, paras 26 and 27.
Far from casting any doubt on the literal and natural meaning of the word "required", all the factors discussed by the Advocate General provide strong – one would have thought conclusive – confirmation of its clear meaning: based on a legal obligation.
It would be unhelpful fully to repeat or to try to improve on the Advocate General's logical and impeccable analysis. As she noted, all the language versions with one exception are unequivocal ("exigés par des dispositions législatives, réglementaires ou administratives"; "que sean exigidos por disposiciones legales, reglamentarias o administrativas"; "die aufgrund von Rechts-oder Verwaltungsvorschriften erstellt werden müssen"; "die door wettelijke of bestuursrechtelijke bepalingen zijn voorgeschreven". etc). The one possible exception is Italian ("che sono previsti da disposizioni legislative, regolamentari o amministrative"). But "previsti" is quite capable of being understood as meaning "required", and, as Advocate General Kokott concluded, it must in the context of the other language versions and of the legislative history be so understood.
The legislative history is particularly striking. Advocate General Kokott summarised it as follows:
With this elucidation, the matter came before the Fourth Chamber of the Court. The Chamber recited that the applicants (seeking to annul certain provisions of a Belgian planning code) had argued that a mere literal interpretation of the relevant phrase in article 2(a) "would entail the dual risk of not requiring the assessment procedure for land development plans which normally have major effects on the territory concerned and of not ensuring uniform application of the directive in the Member States' various legal orders, given the differences in the formulation of the relevant national rules" (para 25).
The Fourth Chamber went on to recite that various Governments had on the other hand submitted that both the language of article 2(a) and the travaux préparatoires showed that "required" meant "required" (para 26) and that the Commission also considered that the test of being "required" was met where an authority was "subject to a legal obligation" (para 27).
The Chamber then simply said that "It must be stated" that "an interpretation" which would exclude from the scope of the SEA Directive plans and programmes regulated by rules of law in the various national legal systems, "solely because their adoption was not compulsory in all circumstances, cannot be upheld" (para 28).
The Chamber no doubt used the phrase "in all circumstances" because the position, under the relevant national law, was that "in certain cases" (among them the case before the Chamber) the municipal authority might refuse to prepare a specific land use plan (para 18). Cases in which the authority had no option but to prepare such a plan would on any view obviously fall within the word "required".
However that may be, the Chamber concluded that "required" means "regulated", so as to catch even cases where no plan was required to be prepared. The only reasons it gave were that to read "required" as meaning "required" would "have the consequence of restricting considerably the scope of the scrutiny" (para 29) or "compromise, in part, the practical effect of the directive, having regard to its objective, which consists in providing a high level of protection of the environment" and "thus run counter to the directive's aim of establishing a procedure for scrutinising measures likely to have significant effects on the environment ...." (para 30).
If, instead of "required", one must read the word "regulated", the question arises what it means. Is it sufficient that legislative, regulatory or administrative provisions grant powers to some authority wide enough to permit a plan or programme to be prepared? Or must such provisions actually refer to a possibility that such a plan or programme will be prepared? Or must they specify points and/or conditions that such a plan or programme, if prepared, must address and/or fulfil? The Chamber referred to provisions which "determine the competent authorities for adopting them [i.e. the relevant plan or programme] and the procedure for preparing them" (para 31).
If this is what is meant by "regulated", then not all plans and programmes can on any view be covered by the SEA Directive, and the desire for comprehensive regulation of plans and programmes "likely to have significant effects on the environment" cannot be met. In any event, it follows from the fact that the SEA Directive only applies to plans and programmes "which set the framework for future development consent of projects", that it is not exhaustive and does not cover every form of plan and programme simply because it could be said to be likely to have significant environmental effects: see Lord Carnwath and Lord Reed's judgments. The SEA Directive and its terms must be read as a whole.
Any condition attached to the scope or application of a legislative measure is capable of affecting its impact. As we have already noted, legislators cannot always agree everything that the most ardent supporters of its general objectives would like them to have achieved. On the Court's own approach, the SEA Directive cannot and does not cover all plans and programmes. They must be "regulated" by legislative, regulatory or administrative provisions.
Had the meaning of article 2(a) come before the Supreme Court without there being any European Court of Justice decision to assist, we would unhesitatingly have reached the same conclusion as Advocate General Kokott, and for the reasons she (as well as the Governments and the Commission represented before the Fourth Chamber) so convincingly gave. We would, like her, have concluded that that "the legislature clearly did not intend" plans and programmes not based on a legal obligation to require an environmental assessment, even though they might have significant effects on the environment [AG20].
We would also have regarded this as clear to the point where no reference under the CILFIT principles was required. The reasons given by the Fourth Chamber of the Court of Justice would not have persuaded us to the contrary. While they allude, in the briefest of terms, to the fact that the Governments made submissions based on the clear language of article 2(a) and on the legislative history, they do not actually address or answer them or any other aspect of Advocate General Kokott's reasoning.
In the result, a national court is faced with a clear legislative provision, to which the Fourth Chamber of the European Court of Justice has, in the interests of a more complete regulation of environmental developments, given a meaning which the European legislature clearly did not intend. For this reason, we would, had it been necessary, have wished to have the matter referred back to the European Court of Justice for it to reconsider, hopefully in a fully reasoned judgment of the Grand Chamber, the correctness of its previous decision.
The EIA Directive
We turn to the exclusion in article 1(4) of the EIA Directive (formerly article 1(5) of Directive 85/337/EC) of "projects the details of which are adopted by a specific act of national legislation, since the objectives of this Directive, including that of supplying information, are achieved through the legislative process". Again, one might have thought that the language of the Directive was clear enough.
The Court of Justice has however read the word "since" as if it said "provided that"; and has treated article 1(5), the predecessor of article 1(4), as subject to a pre-condition that the legislative process must have enabled the objectives pursued by the Directive to be achieved. The case-law has on this basis developed a set of detailed requirements which any national legislature must satisfy. The following propositions emerge:
The legislature must have "available to it information equivalent to that which would be submitted to the competent authority in an ordinary [planning] procedure"- meaning that the minimum information to be supplied must be "in accordance with article 5(2) of Directive 85/337/EC and Annex III thereto" (after the amendments made by Directive 2003/35/EC, these provisions became article 5(3) and Annex IV): Luxembourg v Linster, at paras 54 and 55, Boxus v Région Wallonne Cases C-128/09-C-131/09, C-134/09 and C-135/09,  Env LR 14, at paras 41-43, Solvay v Région Wallonne Case C-182/10,  Env LR 27, at para 37, Nomarchiaki at para 85.
"[T]he legislative act adopting a project must include, like a development consent .... all the elements of the project relevant to the environmental impact assessment":Boxus, , Solvay, , Nomarchiaki, .
The national court must be able to verify that such conditions have been satisfied, taking account both of the content of the legislative act adopted and of the entire legislative process, which led to its adoption, in particular the preparatory documents and parliamentary debates: Boxus, [at paras 47and 48, Solvay, at para 59,Nomarchiaki, at para 90.
According to the Fourth Chamber in Solvay, at paras 53 and 59-61, it follows that "all the reasons for [the] adoption" of the relevant legislative act must either be contained in it or communicated separately.
Lord Reed has quoted substantial passages from Nomarchiaki, the Court's most recent decision in the field. The case-law was decided under Directive 85/337/EC, the language of which is the same in material respects as that of the EIA Directive. Directive 85/337/EC reads:
Article 1(5) gives effect to the eleventh and article 2(3) to the twelfth recital. A materially identical position exists under the EIA Directive, in which article 1(4) gives effect to recital (22), while article 2(4) gives effect to recital (23).
Looking back over the case-law, it is possible to see how the Court in Luxembourg v Linster, at paras 53 and 54, sought to justify its conclusion that the exclusion in article 1(5) only applied where the legislature had available to it information equivalent to that which would be submitted to the competent planning authority in an ordinary procedure. It did so by reference to the sixth recital of Directive 85/337/EC. But the Court failed in this connection to address recital (para 11), now (para 22), recording expressly that the Directive as a whole was not to apply to projects the details of which were adopted by a specific act of national legislation.
It was not until twelve years later, when the Court came to consider the Directive once again in Boxus v Région Wallone that Advocate General Sharpston analysed article 1(5) and recognised some of the difficulties which different readings might present. She identified as alternatives that it connoted (a) an automatic presumption that the adoption by specific act of national legislation process met the Directive's objectives or (b) a disapplication of the Directive provided that the legislative process achieved such objectives, and went on:
It was only after examining the cases of World Wildlife Fund v Autonome Provinz Bozen and Luxembourg v Linster that she concluded [AG72] that they,
appear[ed] clearly to favour construing article 1(5) as containing a prior condition that the objectives of the EIA Directive must be achieved by the legislative process, rather than a presumption that they are so achieved.
The Court in its Grand Chamber in Boxus endorsed this without further discussion, feeding in the express requirement of judicial review of "the entire legislative process" (paras 37-48). The most recent decision, Solvay v Région Wallone, reiterates this position.
The case-law does not identify any textual or contextual basis for the conclusions reached in respect of article 1(5) (now 1(4)). Its reasoning was based exclusively on the "objectives" of the Directive. But the extent to which the European legislature concluded that these general objectives could and should be met, must be gathered from the Directive. On its face, the word "since" explains why specific legislative acts are excluded. It does not introduce a condition to their exclusion. In the light of the representation of the governments of Europe in the Council of Ministers and the mutual trust between them upon which Europe is founded, it is difficult to see why it should be supposed that the Council of Ministers as the European legislator intended a condition, or intended the word "since" to have anything other than its ordinary meaning.
As it stands, the European Court of Justice's case-law in respect of article 1(5) (now 1(4)), raises the question what is meant by the condition that "the legislative process must have enabled the objectives pursued by the Directive to be achieved". As Lord Reed has observed, the appellants treat the condition as if the Court had in effect read back into the Directive in relation to specific legislative acts the provisions of the Directive governing projects subject to the ordinary planning process. There is some support for this in the Court's own invocation of, and insistence on compliance with, the information requirements of article 5(2) and Annex III of Directive 85/337/EC: para 191(i) and (ii) above.
The statement in Solvay (para 191(iv) above) that all the reasons for the adoption of the relevant legislative act must be contained in it or communicated separately finds inspiration in article 9 of Directive 85/337/EC. The basis for requiring a possibility of judicial review of the "entire legislative process" is article 10a of Directive 85/335/EC as inserted by Directive 2003/35/EC, or now article 11 of the EIA Directive. In Solvay (para 59) reference was also made to a passage in R (Mellor) v Secretary of State for Communities and Local Government Case C-75/08  ECR I-3799,  under Directive 85/335/EC as amended, which repeats a passage from Union Nationale des Entraineurs v Heylens Case 222/86  ECR 4097, (para 15), on the general right to judicial review of decisions for their legality:
Effective judicial review, which must be able to cover the legality of the reasons for the contested decision, presupposes in general that the court to which the matter is referred may require the competent authority to notify its reasons.
In Boxus Advocate General Sharpston thought that it followed that a court must not only assess whether the legislature had before it sufficiently detailed and informative information, but must also consider whether the legislative process respected the appropriate procedure and allowed "[sufficient] preparation and discussion time .... for it to be plausible to conclude that the people's elected representatives were able properly to examine and debate the proposed project" [AG84]. Advocate General Kokott in Nomarchiaki took this up and said that it had "made clear that the EIA Directive is not about formalism, but is concerned with providing effective EIAs for all major projects", and that it "must therefore be clarified in the main proceedings whether the legislature .... was able properly to examine and debate the environmental effects of the project" [AG136-137].
Not perhaps surprisingly in the light of the above, the present appellants have argued that it is for the Supreme Court now not only to consider the adequacy of the information placed before members of both Houses of Parliament, but also to take the step of scrutinising the likely adequacy or otherwise of their procedures and debates, including the extent to which individual members are likely to direct attention to and understand, and apply an independent mind to, any issue falling for decision by the legislature within article 1(4) of the EIA Directive. We have as a result had to give careful consideration to where European law has gone and might yet go, and whether in particular it is necessary for us to make a reference to Luxembourg in order for us to decide the issues which are before us under the EIA Directive. We have however come to the conclusion that this is not necessary.
The European Court of Justice was itself careful to use a general formulation, invoking the "objectives" of the Directive, when it re-interpreted "since" to mean "provided that" in article 1(5). It did not say that the Directive or its provisions applied to a specific legislative act. It said that it was a condition of their disapplication that their "objectives" were met by the legislative process. The Court was careful not to endorse the very wide formulae, used by the two Advocates General in Boxus and Nomarchiaki, which suggested close scrutiny by national judges of the legislative process to see whether "the people's elected representatives" had been able "properly" to examine and debate the proposal or had "perform[ed] their democratic function correctly and effectively".
There was good general reason for this. Whatever other adjustments in meaning it might make by way of interpretation, the Court was here concerned with the fundamental institutions of national democracy in Europe. It was concerned with a provision which deliberately distinguished projects approved by legislative process from projects approved by the ordinary planning process. It is not conceivable, and it would not be consistent with the principle of mutual trust which underpins the Union, that the Council of Ministers should, when legislating, have envisaged the close scrutiny of the operations of Parliamentary democracy suggested by the words used by Advocates General Sharpston and Kokott. The Court will also have been well aware of the principles of separation of powers and mutual internal respect which govern the relations between different branches of modern democracies (as to which see, in the United Kingdom context, R (Jackson) v Attorney General  UKHL 56;  1 AC 262, (para 125, per Lord Hope of Craighead). The Court cannot have overlooked or intended to destabilise these. In a not so dissimilar context, the German Federal Constitutional Court noted in its judgment of 24 April 2013 – 1 BvR 1215/07, (para 91) – that decisions of the European Court of Justice must be understood in the context of the cooperative relationship ("Im Sinne eines kooperativen Miteinanders") which exists between that Court and a national constitutional court such as the Bundesverfassungsgericht or a supreme court like this Court.
In the case of the United Kingdom, the approach suggested by the two Advocates General would raise a particular issue of a kind which article 1(4) (formerly 1(5)) was no doubt intended to avoid. It is, we recognise, one that may be specific to the United Kingdom. Article 9 of the Bills of Rights, one of the pillars of constitutional settlement which established the rule of law in England in the 17thcentury, precludes the impeaching or questioning in any court of debates or proceedings in Parliament. Article 9 was described by Lord Browne-Wilkinson in the House of Lords in Pepper v Hart  AC 593, 638, as "a provision of the highest constitutional importance" which "should not be narrowly construed". More recently, in the Supreme Court case of R v Chaytor  1 AC 684, para 110, Lord Rodger of Earlsferry said this:
[I]n his Commentaries on the Laws of England, 17th ed (1814), vol 1, Bk 1, chap 2, p 175, under reference to Coke's Institutes, Blackstone says that the whole of the law and custom of Parliament has its original from this one maxim: 'that whatever matter arises concerning either House of Parliament, ought to be examined, discussed, and adjudged in that house to which it relates, and not elsewhere.'
The principle enshrined in article 9 is recognised and buttressed by a series of constitutional cases which Lord Reed has mentioned in his para 78 above. To take just one example, in Edinburgh and Dalkeith Railway Co v Wauchope (1842) 8 Cl & F 710, the trial judge had suggested that a statute was inoperative if, in breach of the rules of Parliament, no notice had been given to the defendant. Lord Campbell (in observations echoed by Lord Cottenham and Lord Brougham) said, at p 725, that:
I cannot but express my surprise that such a notion should ever have prevailed. There is no foundation whatever for it. All that a Court of Justice can do is to look to the Parliamentary roll: if from that it should appear that a bill has passed both Houses and received the Royal assent, no Court of Justice can inquire into the mode in which it was introduced into Parliament, nor into what was done previous to its introduction, or what passed in Parliament during its progress in its various stages ....
It may well be that it would involve no breach of this well-established principle for the courts to inquire into the information which the executive collects and places before Parliament in connection with a Bill falling within the ambit of Article 1(4) of the EIA Directive – see for instance Bank Mellat v Her Majesty's Treasury (No 2)  3 WLR 179, (paras 38-49), per Lord Sumption. However, even in that connection, we note that Bank Mellat was not concerned with primary legislation, but with a statutory instrument where different considerations apply. For present purposes, and in the light of Lord Reed's conclusions as to the adequacy of the information which will be put before Parliament, it is unnecessary to go further into that particular aspect.
Under the European Communities Act 1972, United Kingdom courts have also acknowledged that European law requires them to treat domestic statutes, whether passed before or after the 1972 Act, as invalid if and to the extent that they cannot be interpreted consistently with European law: R v Secretary of State, Ex p Factortame Ltd (No 2)  1 AC 603. That was a significant development, recognising the special status of the 1972 Act and of European law and the importance attaching to the United Kingdom and its courts fulfilling the commitment to give loyal effect to European law. But it is difficult to see how an English court could fully comply with the approach suggested by the two Advocates General without addressing its apparent conflict with other principles hitherto also regarded as fundamental and enshrined in the Bill of Rights. Scrutiny of the workings of Parliament and whether they satisfy externally imposed criteria clearly involves questioning and potentially impeaching (i.e. condemning) Parliament's internal proceedings, and would go a considerable step further than any United Kingdom court has ever gone.
The United Kingdom has no written constitution, but we have a number of constitutional instruments. They include Magna Carta, the Petition of Right 1628, the Bill of Rights and (in Scotland) the Claim of Rights Act 1689, the Act of Settlement 1701 and the Act of Union 1707. The European Communities Act 1972, the Human Rights Act 1998 and the Constitutional Reform Act 2005 may now be added to this list. The common law itself also recognises certain principles as fundamental to the rule of law. It is, putting the point at its lowest, certainly arguable (and it is for United Kingdom law and courts to determine) that there may be fundamental principles, whether contained in other constitutional instruments or recognised at common law, of which Parliament when it enacted the European Communities Act 1972 did not either contemplate or authorise the abrogation.
We are not expressing any view on whether or how far article 9 of the Bill of Rights would count among these, but the point is too important to pass without mention. We would wish to hear full argument upon it before expressing any concluded view. It is not a point upon which the parties before us proposed to make any submissions until it was raised by the Court. We were then told that the attention of the Parliamentary authorities (and we deliberately use a vague expression) had been drawn to this appeal, and they elected not to be represented. If and when the point does fall to be considered, the Parliamentary authorities may wish to reconsider whether they should be represented, and, particularly if they still regard that course as inappropriate, it may well be the sort of point on which the Attorney General should appear or be represented. Important insights into potential issues in this area are to be found in their penetrating discussion by Laws LJ in the Divisional Court inThoburn v Sunderland City Council  EWHC 195 (Admin),  QB 151, (The Metric Martyrs case), especially paras 58-70, although the focus there was the possibility of conflict between an earlier "constitutional" and later "ordinary" statute, rather than, as here, between two constitutional instruments, which raises yet further considerations.
As it is, however, we have come to the conclusion that the step mentioned in paras 200 and 206 above is one which the European legislature in this instance clearly provided by article 1(4) (formerly 1(5)) need not and should not be taken, and which the European Court of Justice has not endorsed. The Court of Justice's more limited approach in this respect is also borne out by an examination of the objectives of the Directive. Nothing in the Directive suggests that it is aimed at excluding either political involvement or reasoning based on political policy decisions from planning decisions. On the contrary, the recognition that projects may legitimately be approved by specific legislative act constitutes express recognition of the legitimacy of such factors.
The appellants' case, that the Parliamentary process will be tainted by considerations such as whipping or collective ministerial responsibility or simply by party policy, amounts to challenging the whole legitimacy of Parliamentary democracy as it presently operates. There would doubtless be a similar problem in most, probably all, the democracies of the Union. Finally, we note that article 10 TEU itself recognises that, in a Union "founded on representative democracy", whose citizens are directly represented in the European Parliament, "[p]olitical parties at European level contribute to forming European political awareness and to expressing the will of citizens of the Union." This, though it may at present be largely aspirational in so far as it refers to pan-European political parties, undermines any suggestion that the ordinary workings of political democracy can or should be seen as suspect under article 1(4) of the EIA Directive.
In the upshot, there is, as Lord Reed concludes, no basis under the EIA Directive for the wide-ranging review of Parliamentary process advocated by the appellants, whether this be to assess the quality of the consideration given in Parliament or the extent to which the members of either House will be free of party influence when deciding how to vote; and we further agree that this is clear to a point where is no need for a reference to the European Court of Justice.
It follows, as stated above, that the appeals should be dismissed under both Directives.
David Elvin QC & Charles Banner (instructed by King & Wood Mallesons LLP) for appellant.
Tim Mould QC, James Maurici QC, Jacqueline Lean & Richard Turney (instructed by Treasury Solicitors) for respondent.
all rights reserved