(Lord McGhie, Mr D J Houston, Mr A Macdonald)
(Application RN SLC 81/10 – Order of 5 May 2011)
AGRICULTURAL SUBSIDIES APPEALS – SINGLE FARM PAYMENT – REFERENCE PERIOD – DEROGATION PROVISIONS – CALCULATION OF REFERENCE AMOUNTS – REFERENCE TO EUROPEAN COURT – ACTE CLAIRE – COMPLETE CONFIDENCE – EUROPEAN COUNCIL REGULATION (EC) NO.1782/2003 ARTS 36 TO 43 – EUROPEAN COMMISSION REGULATION (EC) NO.795/2004 ART 16.3 – EUROPEAN COMMISSION REGULATION (EC) NO.1974/2004 – TREATY ON THE FUNCTIONING OF THE EUROPEAN UNION, ART 276
In terms of Art 43 of Reg 1782/2003, any farmer whose production was adversely affected during the normal reference period (years 2000 to 2002) by exceptional circumstances or the results of involvement in agri-environmental commitments was entitled to request that the reference amount be calculated on the basis of a different period. If the whole primary reference period was affected, the reference amount was to be calculated on the basis of the 1997 to 1999 period. Art 16.3 of Reg 795/2004 provided, in broad effect, that the application of Art 43 should be on the basis of the individual schemes.
A farmer whose business was affected by various factors including commitments under an agri-environmental scheme within the primary reference period was found entitled to adopt the earlier reference period for the purposes of a sheep scheme (SAPS). He contended that the same period had to be used for his cattle schemes (SCPS and BSPS). HELD that the plain meaning of Art 16 was that separate reference periods were to be used for each scheme even if that might give rise to some anomalous results and the case was sent for proof on the question of whether production under the cattle scheme had been affected throughout the primary reference period.
The note appended to the Court’s order is as follows:
This is an appeal by Mr Gladwin under the Agricultural Subsidies (Appeals) Scotland Regulations 2004 from a decision of Scottish Ministers relating to the determination of the appellant’s entitlement under the Single Farm Payment Scheme (“SFPS”). The principal issue was the determination of the appropriate reference period for calculation of reference amounts in relation to the Beef Special Premium Scheme (“BSPS”) and the Suckler Cow Premium Scheme (“SCPS”). We first heard parties in debate on 14 February and later, in the circumstances explained below, on 13 April 2011. The appellant was represented by Sir Crispin Agnew, QC, and the respondents by Mr Donald Cameron, Advocate.
Case C-61/09 – Landkreis Bad Durkheim v Aufsichts - und Dienstleistungsdirektion (Niedermair-Schiemann); 11 May 2010, Opinion of Advocate General
Case C-61/09 - Landkreis Bad Durkheim v Aufsichts und Dienstleistungsdirektion; 14 October 2010, Decision of the European Court of Justice
Case C-152/09 – André Grootes v Amt für Landwirtschaft Parchim; 8 July 2010, Opinion of the Advocate General
Case C-152/09 - André Grootes v Amt für Landwirtschaft Parchim; 11 November 2010, Decision of European Court of Justice
Case C-170/08 – Nijemeisland v Minister van Landbouw, Natuur en Voedselkwaliteit; 11 June 2009, Decision of European Court of Justice
Case C-115/10 – Babolna Mezogazdasagi Termelo, Fejleszto es Kereskedelmi Zrt. v Mezogazdasagi es Videkfejlesztesi Hivatal Kozponti Szerve, 11 February 2011, Opinion of Advocate General
Case C-283/81 Cilfit and ors  ECR 3415 Decision of European Court of Justice
Bulmer Ltd v Bollinger S.A. in  1Ch. 401
Matadeen & Another v Pointu and Others and Minister of Education and Science and another  AC 98
R. v Stock Exchange Ex p. Else Ltd  QB 534
R. (on app T.A. Gwillim & Sons) v Welsh Ministers  EWHC (Admin) 2946
R. (on app T.A. Gwillim & Sons) v Welsh Ministers  EWCA (Civ) 1048
European Council Regulation (EC) No.1782/2003
European Commission Regulation (EC) No.795/2004
European Commission Regulation (EC) No.1974/2004
In this decision references to art 16.3 are to art 16 in Regulation 795/2004, as amended by Regulation 1974/2004. Article 16.3 is in the following terms:
“3. Article 40 of Regulation (EC) No.1782/2003 shall apply on the basis of each direct payment referred to in Annex VI of that Regulation”.
The remaining references are to articles in Regulation 1782/2003. It is appropriate to set out the full text of art 40 and parts of art 43 at this point, other articles are quoted where appropriate. Article 40 is in the following terms:
1. By way of derogation from article 37, a farmer whose production was adversely affected during the reference period by a case of force majeure or exceptional circumstances occurring before or during that reference period shall be entitled to request that the reference amount be calculated on the basis of the calendar year or years in the reference period not affected by the case of force majeure or exceptional circumstances.
2. If the whole reference period was affected by the case of force majeure or exceptional circumstances, the Member State shall calculate the reference amount on the basis of the 1997 to 1999 period. [There follow certain provisions which have no bearing on the present case]
3. A case of force majeure or exceptional circumstances, with relevant evidence to the satisfaction of the competent authority, shall be notified by the farmer concerned in writing to the authority within a deadline to be fixed by each Member State.
4. Force majeure or exceptional circumstances shall be recognised by the competent authority in cases such as, for example:
(a) the death of the farmer;
(b) long-term professional incapacity of the farmer;
(c) a severe natural disaster gravely affecting the holding’s agricultural land;
(d) the accidental destruction of livestock buildings on the holding;
(e) an epizootic affecting part or all of the farmer’s livestock
5. Paragraphs 1, 2 and 3 of this Article shall apply, mutatis mutandis, to farmers who, during the reference period were under agri-environmental commitments [Specific details of such commitments were then given. It is not disputed that the ESA scheme referred to below was such a commitment].
In the case where the measures referred to in the first sub-paragraph covered both the reference period and the period referred to in paragraph 2, Member States shall establish, according to objective criteria and in such a way as to ensure equal treatment between farmers and to avoid market and competition distortions, a reference amount in accordance with the detail rules to be laid down by the Commission in accordance with the procedure referred to in Article 144(2).”
Article 43 appears under the heading “Determination of the payment entitlements” and paragraphs (1) and (2) provide:
“1. Without prejudice to Article 48, a farmer shall receive a payment entitlement per hectare which is calculated by dividing the reference amount by the three-year average number of all hectares which in the reference period gave right to direct payments listed in Annex VI.
The total number of payment entitlements shall be equal to the above mentioned average number of hectares.
[Provision was then made for new enterprises]
2. The number of hectares referred to in paragraph 1 shall further include:
[Provision was made for a number of special crop types which do not concern us]
(b) all forage area in the reference period.”
 This case turns on provisions which operate by way of derogation from the main scheme for transfer from support based on specific elements of production to support decoupled from production. Derogating provisions can be expected to turn on their own terms and will not necessarily fall to be given a meaning consistent with the provisions from which they are to derogate. However, they will require to be consistent with the operation of the scheme as a whole. It is necessary to understand the basic scheme to place the derogation provisions in proper context.
 With effect from 1 January 2005 the SFPS replaced most of the former production based subsidy schemes operating across the European Community. The “historic basis” for allocating entitlements was adopted in Scotland. That meant that from 2005 the quantum of annual payments provided by way of support to farmers depended on claims made during a specified reference period. Article 38 provided that: “The reference period shall comprise the calendar years 2000, 2001 and 2002”. We refer to that reference period as “the primary reference period”.
 In terms of art 33, farmers were entitled to have access to the single payment scheme if they were granted a payment under at least one of the support schemes set out in Annex VI of Regulation 1782/2003. There is no dispute as to Mr Gladwin’s eligibility to access the SFPS.
 The first stage, in the calculation of how much the single payment should be, required calculation of “the reference amount”. In terms of Article 37.1:
“The reference amount shall be the three-year average of the total amounts of payments, which a farmer was granted under the support schemes referred to in Annex VI, calculated and adjusted according to Annex VII, in each calendar year of the reference period referred to in Article 38”.
 Other provisions of art 37 made special provision for various different crops and fruits. There was also provision for agricultural activity which commenced during the reference period. It was not suggested that these subsidiary provisions had any bearing on the present case.
 Annex VII made provision for calculation of the appropriate reference amount for each of a range of different schemes. In relation to livestock payments the amount was calculated by multiplying the sum of the number of determined animals for which a payment was granted respectively in each year of the reference period by amounts established in respect of each scheme for the year 2002 and dividing the result by three. In terms of art 43.1, the “reference amount” was converted to “payment entitlements” by dividing it by the 3-year average number of all hectares which in the reference period gave right to direct payments of the type listed in Annex VI. The payment entitlement was a cash figure. The number of payment entitlements was the number of hectares used in the calculation.
 Under the SFPS, as it operated from 2005, a farmer would receive payment of support calculated on the basis of each of his payment entitlements for which he had an eligible hectare available to him for agricultural purposes on specified dates: art 44.1. In short, the main aim was to provide a single annual payment calculated by reference to the average amounts paid by way of support under individual schemes in the years 2000 to 2002. In the usual case, subject to issues such as deductions to fund the National Reserve and modulation, the annual payments were expected to continue at a level similar to receipts in that period but without direct regard to the nature of the farmer’s production. Put broadly, the important features of the scheme are the need to identify a reference period, or possibly reference periods, for calculation of the reference amount and the need to the establish the number of hectares to be used as a divisor in order to establish the value of each entitlement. The number of hectares used would normally also be the number of entitlements. In succeeding years payment would be triggered when the farmer had available a hectare to match each entitlement.
 The factual background against which the relevant issues arise is not of critical importance to the questions of law and we set it out briefly just to show the context. We have not heard evidence and the facts established after proof may differ to some extent from this summary.
 Mr Gladwin bought the relevant land on 31 October 1996. He took over some 180 suckler cows and finishing cattle and a substantial number of sheep. After the purchase, he increased the cattle numbers to over 260. He had claims at or about that level under the SCPS and BSPS in the years 1997 and 1998. He declared 1238, 1394 and 1408 eligible sheep under the Sheep Annual Premium Scheme (“SAPS”) in the years 1997, 1998, and 1999 respectively. In his first year of operation, he applied to join the Central Southern Uplands Environmentally Sensitive Area Scheme (“the ESA scheme”). This was approved on 5 January 1998. At that time he expected to have to reduce his sheep numbers by 300 to comply with the various requirements of the ESA scheme. It is said that he did not expect it to have any impact on his cattle.
 For reasons not yet established there was a substantial drop in the claims made under the cattle schemes in 1999 and no claims were made 2000 and 2001. His claims under SAPS were 845 and 773 in 2000 and 2001, respectively. Mr Gladwin was sequestrated on 12 June 2001 and an interim trustee was appointed. There was no claim under any scheme in 2002.
 The ESA Scheme agreement was due to last for a period of 10 years from January 1998 although there was provision for it to be terminated at the end of 2002. However, it was a matter of agreement at debate, that the applicant’s participation in the scheme, and his commitment under it, ended in 2001.
 After a separate appeal procedure, the Scottish Ministers accepted that Mr Gladwin’s production in relation to sheep had been adversely affected by the agri-environmental commitments imposed by his ESA Scheme. It appears that they accepted that this affect persisted in 2002. Accordingly, the whole primary reference period was affected and for calculation of that part of the reference amount based on the sheep meat and goat meat schemes, it was accepted that the effect of art 40.1 and 40.2 was to allow the payment under SAPS to be included in the calculation of payment entitlement on the basis of the 1997 to 1999 period.
 However, it was the respondents’ position on all the written material before us and, initially at least, at debate, that whether the decision in relation to the SAPS was right or wrong, the fact that there was no ESA scheme in 2002 meant that the appellant’s production under any cattle schemes in 2002 could not have been adversely affected by the ESA scheme and that, accordingly, the calculation of the reference amount could make no allowance for cattle because no payment had been made in that year.
 Both parties addressed the detail of certain of the relevant EU Regulations and we were referred to a number of cases, as set out above. Both parties made reference to aspects of the facts but it became apparent that their references were, on the one hand, in support of the appellant’s contention that the cattle business had been adversely affected by the ESA throughout the years 2000, 2001 and 2002 and, on the other, in support of the respondents’ contention that the dramatic downturn in the cattle side of Mr Gladwin’s operations was not due to the scheme. As both sides were agreed that a proof before answer would be required if this issue was to be addressed, it is unnecessary to say more about these submissions.
 It became clear that, essentially, the debate raised two distinct issues:
(a) The first arose from the Scottish Ministers’ having accepted that the appellant’s production was adversely affected by the ESA Scheme for the years 2000 to 2002 inclusive in relation to sheep and that the reference period for sheep should, accordingly, be the years 1997 to 1999. It was contended by the appellant that any change in reference period affected all calculations. In other words, that the Scottish Ministers, if requested, were obliged by the provisions of Article 40 to base all calculations as to entitlement to the single farm payment by reference to the period 1997 to 1999. The Ministers, on the other hand, contended that the effect of art 16.3 was to require them to determine a reference period for each individual type of direct payment, so that the BSPS or SCPS and SAPS had to be dealt with as entirely separate elements in the calculation.
(b) The second issue, which only arose if the first was answered in the Scottish Ministers’ favour, was a contention on their behalf that as Mr Gladwin had accepted that his commitments under the ESA Scheme ended in 2001, it could not be said that his cattle based schemes were affected by the ESA in 2002. This meant that the whole of the primary reference period could not be said to have been affected and it was contended that the reference amount for cattle accordingly fell to be calculated by reference to his 2002 claim – which, as we have seen, was nil.
 In relation to the first point, Sir Crispin accepted that art 16.3 could be read in support of the Scottish Ministers’ approach whereby a separate reference period was to be taken for each different type of support scheme, but his main submission was that the provisions should be understood, on a purposive approach, as intended to show that art 40 applied only to the schemes set out in Annex VI. That Annex provided only a basis of calculation and Annex VII showed different methods of calculating the “reference amount”. Neither Annex had any bearing on the reference period. Article 16.3 had to be construed in a purposive way and, in particular, had to be construed in a way which was consistent with the principles of equal treatment and avoidance of market distortion. In essence, it was simply saying that it applied the lists in Annex VI to Article 40 in the same way as they applied to art 37. The purpose was just to make it clear that art 40.1 was not to be read as if it only applied if the whole production was adversely affected. He submitted that art 16.3 had nothing to do with the question of re-dating of a reference period. It was to be read as a provision making it clear that an adverse effect on any one scheme brought the farmer within art 40.1. But for that provision it might have been thought necessary to establish an adverse effect on production as a whole.
 It should be noted, for completeness, that Sir Crispin also attempted to support the contention that only one reference period could be used, by a submission that such a result was necessary on the basis of an administrative obligation of equality of treatment. He contended that if the calculation of the sheep scheme was rebased, the cattle schemes should also be rebased because “like cases” should be treated in the same way. However, we understood him ultimately to accept that this would turn on proof of causation. If the sheep had been affected by agri-environmental commitments and the cattle had not, that would be a justification for different treatment. The Scottish Ministers’ acceptance in relation to the sheep was simply a decision on fact relating to that scheme. It could not govern the factual issues arising in relation to cattle.
 For the Scottish Ministers it was contended, without much elaboration, that art 16.3 clearly required art 40 to be applied looking at each scheme separately. We were told that was the approach that Scottish Ministers had always taken.
 Following the first day of debate, we attempted to consider these competing arguments. However, for the reasons set out in our Note of 1 March 2011 we came to have a concern that the implications of the arguments had not been fully discussed at debate. It was not obvious to us how the scheme would work, in certain foreseeable circumstances, if more than one reference period was to be adopted.
 The particular circumstances of the present case risked clouding the issue. It is an extreme example of circumstances where the appellant would clearly benefit from application of a single reference period applying to his whole production. It may be worth observing that we recognise that no question of unfair advantage arises. If he is to be assessed on the basis of his operations in 1997 to 1999, this can be assumed to be a period when the farm was being operated in a perfectly normal way. Whatever the reason for the absence of cattle in the primary reference period, the circumstances were clearly unusual. Neither he, nor the interim trustee, had reason to know that events in that period would be the measure of continuing annual entitlement.
 However, our concern was not with the facts of the present case but with the question of construction. We were satisfied that the start point was to follow the plain meaning of the language used. But we recognised the need for a purposive approach to construction of European legislation and that this might justify a departure from the plain meaning in certain circumstances. Putting matters very broadly, we thought it might be necessary to depart from the more obvious construction if we were satisfied that such construction would lead to inappropriate results or be “unworkable” in the sense that it was not consistent with the practical working of other provisions. Our particular concern was with the provisions of art 43. We were satisfied that art 43 could be applied without difficulty if Sir Crispin’s approach was correct. We were not sure how it would work if each scheme had to be considered individually. We accordingly invited further submissions and gave examples, in simplistic terms, of different circumstances which might arise and which might usefully illustrate the effect of the competing constructions.
 At the second hearing neither party attempted to address us on what sort of result might justify a departure from the primary meaning of the language used. Sir Crispin submitted, in effect, that the meaning was clear enough and beyond that contented himself with the submission that the respondents’ construction gave rise to anomalies which would be avoided by his construction. Mr Cameron also submitted that the meaning was clear and contended that there were provisions in the Regulations which meant that any perceived anomalies could be overcome.
 In relation to art 43, Sir Crispin submitted that while the article could probably work in most circumstances under either interpretation, there was a risk that, if there had been a change of farming practice, a rebasing in respect of individual schemes could result in a greater number of hectares having to be taken into account than the actual area of the farm. He suggested that it was not entirely clear how this would be dealt with by Scottish Ministers. He accepted that a construction was possible which allowed a calculation to be made within the terms of the Regulations which could accommodate the situation but he did not accept the Scottish Ministers’ approach, particularly as shown in their calculations in respect of forage areas. He contended that forage should be apportioned between schemes for the purpose of calculating areas. He accepted that the provisions of art 43(1) were wide enough to allow that to be done.
 In response to the submissions for Scottish Ministers as to how the problem of excess hectares should be addressed, Sir Crispin contended that their apparent reliance on arts 42, 58, etc, discussed below, was misplaced. He submitted that these were provisions which only applied after the stage of determining the number and value of entitlements.
 For the Scottish Ministers, Mr Cameron dealt in detail with various calculations under art 43. He pointed out that the English approach to calculation of Single Farm Payment as a whole was different from the Scottish approach. The English had taken advantage of a different option. Accordingly, their approach to detailed aspects of the scheme might not provide sound guidance although he advised us of his understanding that all four jurisdictions within the UK had taken the same approach to art 16.3. He also said that enquiries as to how other countries had administered the scheme, had produced no useful information. It was, of course, recognised that the problem was one of construction. How other countries had construed the relevant provisions would not necessarily have provided better guidance than evidence of the Scottish Ministers’ own approach. It was clear from the productions that the Scottish Ministers had consistently applied the separate schemes approach. He accepted that the examples posed by the Court were realistic but stressed that the Ministers had not in fact encountered any problems. It appeared that, in practice, where one scheme was adversely affected this had not normally been followed by any great increase in other schemes. Mr Cameron said that any difficulties which might arise in relation to areas in excess of the actual farm could be dealt with by recourse to the provisions of arts 42(5), and 58 of Regulation 1782/2003 and art 7(1) of Regulation 795/2004. However, he did not suggest that the Scottish Ministers had experience of this.
 He explained how the Ministers had treated calculation of forage areas and referred to art 43(2)(b). The Scottish Ministers took the view that the forage area did not necessarily go with any particular headage based scheme. The area farmed reflected the “primary situation”; what the farmer was actually doing in the primary reference period. It would only be changed if all schemes were rebased. Usually when changing a base year for a specific scheme they would not have required to change the area.
 In relation to the second issue, it was accepted that a proof would be required to determine whether the ESA had had an adverse impact on the cattle schemes in the primary reference period. We go on to discuss the issue below but it is unnecessary to set out the parties’ submissions at this stage.
 It is clear that, if art 40 was absent, it would be necessary to read all references to the “reference period” as meaning the single period defined in art 38. It is also fair to say that the overall scheme of the relevant provisions of the Regulation appears to be consistent with use of a single reference period. As we have seen, the scheme requires a total “reference amount” to be found by reference to years in “the reference period” and that sum requires to be divided by the number of hectares, again in “the reference period”, in order to derive a figure of “payment entitlements”.
 However, it is clear that art 40 operates by way of “derogation”. It is plainly intended to allow for change or changes to the reference period. Where other articles refer to a definite reference period, there is no difficulty in reading all references as implicitly qualified by the derogation provisions. If we leave aside the possibility of the article allowing more than one reference period, the provisions of art 40 seem tolerably clear. Although we may have introduced some confusion at debate, by raising the question of a possible alternative construction of the latter part of art 40.5, we are satisfied that, on a fair reading of art 40 as a whole, the appropriate reference period to be taken when there has been an adverse affect on production under any scheme has been spelled out carefully. If the adverse affect has been limited to one or two years of the primary reference period, the relevant reference period will be the remaining year or years. Where the affect has covered every year in the primary period, a new reference period comes into play. That, in terms of art 40.2 is the period 1997 to 1999. Where the adverse affect has covered all of that period, as well as the primary reference period, the Member States are obliged to find another way of determining the “reference amount” in accordance with objective criteria and with rules to be established by the Commission: art 40.5. In the present case, it is not suggested that the whole period 1997 to 1999 was adversely affected; the ESA scheme did not start until 1998, and it is unnecessary to consider how such objective criteria were or might have been applied by Scottish Ministers in other cases.
 But the critical question is whether the derogation is to be applied to all schemes together or only to the scheme or schemes adversely affected. That turns on the meaning and effect of the amendment made by the introduction of art 16.3. Sir Crispin pointed out that this was an amendment introduced in October 2004. He contended that a purposive construction was required and we have, accordingly, thought it appropriate to look at the relevant terms of the Preamble to the amending regulation which was Commission Regulation (EC) No.1974/2004. Paragraph (7) of the Preamble is in the following terms:
“Article 40 of Regulation (EC) No 1782/2003 provides for the application of a different reference period in hardship cases affecting production. It is therefore appropriate to specify that the application of that Article should be done on the basis of each direct payment referred to in Annex VI of that Regulation which corresponds to the different productions”.
 As we have seen, the Regulation provided that Para 3 be added to art 16 of the Regulation 795/2004 saying that Article 40 …”shall apply on the basis of each direct payment referred to in Annex VI of that Regulation”. Unsurprisingly, the language of art 16.3 is consistent with the Preamble. We do not think it is easy to read it as intended to do anything short of requiring the whole provisions of art 40 to be applied to each individual scheme for direct payment rather than requiring these provisions to be applied to all the schemes involved in a particular farmer’s operation whether adversely affected or not. The provision is expressed in terms which refer to the “application” of art 40 rather than simply to the method of assessment of adverse affects.
 As noted above, we accept that a purposive approach must be taken to European legislation and we understand that this may require a wider approach to construction than would be appropriate in relation to domestic material. If a primary meaning given to one article proved unworkable in the context of other provisions, we would be entitled to prefer any alternative construction open on the language used if it was necessary to avoid that consequence. We were also prepared to hear submissions that a test, lower than “unworkable” might have to be applied in furtherance of the aim of purposive construction. However, we did not hear positive submissions bearing on either of these potential lines of approach. In the circumstances, it is sufficient to say that we do not discount the possibility of approaching construction by means of an attempt to assess which approach appears to produce the more acceptable result. But we are satisfied that where an article is to apply by way of derogation, it is not sufficient reason for departing from the straightforward construction to be able to point to inconsistencies of results compared with the scheme as it might otherwise have applied. Nor do we think it sufficient to point to examples of possible situations where application to individual schemes would give a farmer a significantly higher reference amount than would have been possible if it was necessary to apply a single reference period to all schemes - although it may be said that that was the consideration which initially gave us pause.
 At the continued hearing we had the advantage of seeing the results of the parties’ different approaches set out clearly in tabular form. Broadly speaking, the calculations confirmed two main points. Where adverse circumstances affected one scheme and a farmer had responded to this by increasing production under another scheme, the use of separate reference periods had the potential to provide substantial over-compensation. It was agreed that, broadly speaking, production levels on British farms overall did not tend to fluctuate much from year to year. If a farmer was forced to adopt a single new reference period going back to a period when production was unaffected, it was likely that in most cases, production over the period 1997 to 1999 would not be significantly different from 2000 to 2002. In short, in Britain, figures for the earlier period would usually provide a perfectly reasonable basis for assessment of what the position would have been had there been no question of adverse circumstances.
 However, we took the view that the matter could not be resolved by having regard simply to typical British farming practices. Our concern was not with the way other countries might construe the Regulations but with the fact that the Regulations have to apply across Europe and have to cover a wide variety of circumstances. Farming in Great Britain is a well developed industry. Individual farms might change and diversify but broadly there would not be expected to be significant increases in production year on year. However, the legislators might have had in mind countries where agriculture was less developed. Where part of a farming enterprise was developing, increasing from year to year, adoption of a single reference period might well lead to prejudice. If one scheme was adversely affected by a relevant factor, it might well be prejudicial if allowance for that was made conditional on having the whole business, including the developing parts, rebased.
 Our simplistic example of such a situation was of a cattle enterprise growing steadily in numbers from an average of 3000 head in the years 1997 to 1999 to an average of 4000 in the years 2000 to 2002, with only a very small area, say 100 ha, in grain. We wanted to consider the effects of the different approaches to that situation. If an agri-environmental scheme took that grain area out of production for the whole of the primary reference period, how would parties approach the calculation? It was clear from the Scottish Ministers’ very clearly laid out tables that, in this situation, there was indeed a risk that the appellant’s approach would be likely to prevent the farmer being able to receive compensation for loss of his grain production. The reduction in relation to the unaffected scheme, if figures for it had to be based on the earlier period, would outweigh the compensation which might have been due in relation to the affected scheme if taken on its own. That risk was avoided by the Scottish Ministers’ approach. Neither party suggested any scenario under which that approach would be to the prejudice of the farmer.
 The various calculations provided in response to the examples suggested in our Note of 1 March 2011 confirm the potential for the respondents’ construction to lead, in some circumstances, to farmers whose production was affected by hardship, being significantly overpaid by comparison with what they might reasonably have expected to receive had there been no such hardship. We note the Ministers’ position that although it has been easy enough to figure situations where that might occur, it was not in fact a common occurrence. We do not know where the perceived balance of risk might have been thought to lie. But the legislators might have had reason to accept the risk of occasional apparent overpayment as justified by the need to avoid the risk of circumstances where the appellant’s construction would lead to a farmer failing to receive adequate compensation.
 We proceed on the broad basis that if the legislation produces an acceptable result in most circumstances, the fact that situations can be figured which might lead to apparently odd results would not in itself be a justification for departure from the straightforward construction. A purposive construction is not radically shaken by the chance that, occasionally, a particular farmer might turn out to get what might seem to be an unfair benefit. A major change such as the one decoupling subsidies from production would be expected to produce some inevitable unevenness in result. The plain intention of the Commission was that the calculation of reference amount across Europe should be based on the period 2000-2002. Only in the special circumstances set out in art 40 was this to be departed from. The Ministers’ construction is consistent with the aim of preserving the primary reference period as far as possible.
 As it turned out, our main concern came to be the question of how the scheme - as set out in art 43 - could apply in cases where the total of the different areas used for different schemes at different times exceeded the actual physical area of the farm. There would be no difficulty under the appellant’s construction. If a single reference period had to be used for all schemes, this problem simply did not arise. But the provisions of art 43 did not seem well adapted to the concept of separate reference periods and our concern was that it might be impossible to operate the scheme using separate periods.
 In the event, we have not been able to conclude that a practical approach cannot be found within the scope of a purposive construction of art 43 to deal with separate periods for different schemes in most, if not necessarily all, possible situations. As we have noted, Sir Crispin advanced a construction of art 43 which allowed this to be done on a proportionate basis. We did not find entirely persuasive the Scottish Ministers’ suggestion that the difficulty was addressed by reference to the provisions of arts 42(5), 58, and art 7 of Regulation 795/2004. We accept Sir Crispin’s contention that, on the face of it, these articles are dealing with a period after the initial calculation of entitlement. However, although we can see a difficulty in the way the Ministers are said to approach their calculations, we have not been persuaded that such difficulties justify a different construction of art 16.3. It may be that the apparent problems are so rare in practice that the question is academic. We are not satisfied that a difficulty in applying art 43 on a particular construction of art 16.3 would justify the conclusion that art 16 required a different construction. It might be that the possibility of a particular outcome had been overlooked by the legislators. If so, the risk of such an outcome could hardly be relied on as a guide to the legislative intention.
 Although there was some discussion of the proper approach to forage areas, we do not think that they give rise to any real difficulty in the present context. The approach taken does not have a bearing on the quantum of the reference amount. Grant was not paid on forage as such but by reference to the head of stock which happened to be kept on such forage. It might have been important if it led to a need to calculate entitlements by use of an area greater than the actual area of the farm. But, if the only schemes in issue are forage related the difficulty is resolved either by acceptance of the appellant’s contention that a proportion of forage be allocated to each scheme or by acceptance of the Minister’s contention that the forage figure should not be changed unless all stock schemes require to be rebased. Neither approach leads to a need to calculate entitlements by reference to an area greater than the actual area of the farm. It does not lead to a change in the total reference amount. It is unnecessary for us to say more.
 A potential problem remains in relation to the example of a switch from a stock, or headage, based scheme to a cereal, or area, based regime. Under the Scottish Ministers’ approach, an area which might be included in one stock based reference period as forage, could, potentially, reappear in a different reference period utilised for cereals. This could lead to a potential duplication in relation to the reference amount. There would not necessarily be a duplication of entitlement because the divisor used to calculate the value of each unit would reflect the element of duplication. However, the farmer might well find himself, or herself, with more entitlements than matching eligible hectares for the purposes of art 36(1).
 In Scotland the possibility exists of acquiring so-called “naked acres” and a farmer with such extra entitlements might be able to take them up by acquiring more ground.
 This may not be possible in all countries covered by the scheme and it is a possibility which may not have been envisaged by the draftsman. In any event, it must be kept in mind that the provisions operate by way of derogation from the scheme. They provide an option for a farmer affected by hardship. It might be that the draftsmen recognised that a broad approach was required and that they could not cover every eventuality. The option of taking “naked acres” is not entirely free from difficulty.
 As mentioned above, we do not think that a difficulty of application of a particular provision would necessarily force us to depart from the straightforward construction of the provision. In the present case we are not persuaded that any potential difficulties over calculation of entitlements under art 43 are sufficient to require us to reject the respondents’ construction. We can accordingly deal with the first issue by saying that we accept that the effect of art 16.3 and art 40 is to require separate reference periods to be used for individual schemes.
 As discussed in the Decision section below, we have decided not to issue a formal determination in respect of this issue at this stage. Had we required to do so, we might have thought it appropriate to invite parties to comment on a further point which, unfortunately, did not occur to us at an earlier stage. It appears to provide a further answer to the apparent problem of potential duplication of area. We note that the express provisions of art 40 relate to changes in the reference period for the purpose of calculation of the “reference amount”. There is nothing in art 40 which expressly authorises any derogation from provisions of the Regulation governing determination of the reference period for any other purpose. In other words art 40 only has a direct bearing on art 43 for the limited purpose of qualifying the reference to “reference amount” not for the purpose of qualifying the expression “reference period”. The substantive provisions of art 43 are contained in the first paragraph of 43(1). It provides that a farmer shall receive a payment entitlement calculated by dividing the reference amount in a specified way. In defining the divisor there is no reference to the “reference amount”. The divisor is defined by reference to the “three year average number of all hectares which in the reference period gave right to direct payments listed in Annex VI.”. The provisions of art 40 have no direct bearing on any part of that phrase. They are not incorporated by necessary implication because the divisor is defined in a way which avoids reliance on the “reference amount”. If this analysis is correct it would probably meet the supposed difficulty in relation to duplication of area. It might, in some circumstances, lead to a farmer having fewer entitlements than he would have had if there had been no adverse circumstances but the value of each entitlement would be higher. There might be other aspects we have overlooked and had we had any real doubt about the first issue we would have invited further submissions.
 We can deal shortly with the second issue. Although Mr Cameron concluded by restating his submission that the appellant’s acceptance that the commitment under ESA ended in 2001, meant that he could not rely on art 40.5, we are satisfied that his acceptance of a need to look at causation broadly, removes the foundations from that submission. In any event, we think it is inconsistent with the express provision in Article 40.1 allowing derogation in respect of events “occurring before or during that reference period”. If the appellant can show that absence of any production of cattle in 2002 was due to the adverse impact of the ESA commitments applying in 2001 or previous years, he might well come within the scope of the derogation. This may be illustrated by the decision reached in relation to the sheep. Although we have no material which would allow us to say whether or not that decision was sound in fact, we have no reason to doubt its soundness in principle. We are satisfied that the fact that there were no binding commitments under any scheme in 2002 does not make it impossible for the appellant to prove that cattle production was adversely affected during that year by the impact of the earlier ESA.
 Although that is sufficient to deal with the second point, it is right to note that in the Stage 2 panel report, in correspondence, and in the written pleadings, there were assertions couched in terms which suggested that the respondents clearly thought it necessary, for the purposes of art 40, to show that the adverse affect arose directly from an explicit contractual obligation imposed by the ESA. For example, in their answers to the grounds of appeal at 6(e) they said: “The ability of the respondent to adjust payment entitlements was tightly circumscribed by the relevant legislation. The appellant had to show there was a contractual obligation within the ESA which impacted on the claims that were submitted during the reference period. The appellant’s management agreement under the scheme did not require him to remove cattle from his holding or to reduce cattle numbers. Accordingly for these schemes, the respondent did not consider his production was affected due to being under an agri-environmental commitment”.
 Sir Crispin properly spent considerable time at debate dealing with the issue of contractual obligation but Mr Cameron, in his response, freely accepted that the question was one of causation rather than contract. He appeared to suggest that the respondents had always taken that approach and stressed that they had given the appellant ample opportunity to establish that the reduction in cattle had been due to the ESA. He said that the respondents had taken a reasonable decision on the material before them. However, he accepted that the Court would be entitled to consider the whole question afresh in light of any competent evidence before it. In accepting that a broad approach to causation was required, he made the point that this was only by contrast with a contention that the question was confined by reference to contractual obligations. Some difficult issues of causation might remain to be determined after a proof before answer.
 For the reasons set out above we are minded to make a determination that separate reference periods are appropriate for each different production based scheme and that the Scottish Ministers’ approach on this matter is correct. However, we have decided simply to appoint a further hearing without making a formal order. This will allow parties to consider further their positions in relation to the possibility of a reference to the European Court of Justice under art 267 of the Treaty on the Functioning of the European Union and also avoid any question arising at this stage as to time limits for appeal. We would be prepared to consider the issue of a formal decision if either party wished to take the question further at this stage.
 In relation to art 267, it was agreed that the observations of Lord Denning M.R. in Bulmer Ltd v Bollinger S.A. provided a sound guide to the approach we should take. Sir Crispin suggested that the short summary in Reg. v Stock Exchange Ex p. Else Ltd also provided useful guidance. Passing reference was also made to the Cilfit case. However, it was clear that neither party encouraged us to make a reference. Sir Crispin recognised that parties had the option of appeal to the Court of Session but resisted the idea of a case being stated by this Court at its own instance. He made the point that we would only decide to do so if we were in real doubt and, if so, we should simply make a reference under the terms of the Article rather than incur the delay and expense of involving the Inner House. But, in any event, he submitted that we could only make a reference if it was necessary for determination of the case before us that we reach a decision on the proper construction of a disputed provision. In the present case, the appellant contended that, on the facts, that issue had not yet been identified as necessary for determination of the case. Even if the appellants were wrong in their approach to the reference period they were, in any event, seeking to demonstrate that the beef production was adversely affected throughout the primary reference period.
 We accept that it is appropriate for us to determine the issues of fact in this case before seeking the assistance of the European Court. In the Stock Exchange casethe matter was expressed robustly. “I understand the correct approach in principle of a national court (other than a final court of appeal) to be quite clear: if the facts have been found and the Community issue is critical to the court’s final decision, the appropriate course is ordinarily to refer the issue to the Court of Justice unless the national court can with complete confidence resolve the issue itself.”: p 545D. That might suggest that the need to establish the facts was a condition precedent to any reference. However, the matter was not expressed in such terms in the Bulmer case – see at p 423C – and we do not need to go farther than say that we accept that, in the present case, it would be appropriate to determine all the facts before considering a reference.
 It has become clear that there will have to be a proof before answer to establish whether cattle production was adversely affected by the ESA Scheme in the years 2000 to 2002. It was agreed that if we reached such conclusion we should put the matter out for further consideration as to procedure. However, we are broadly satisfied that there has been such an exchange of assertion and counter-assertion over the years as to mean that there is little likelihood of either party being caught by surprise by evidence on any issue of fact. Accordingly, we see no clear need for further pleading.
 In the debate, various points were made about causation but, as the matter is to go to proof before answer, it is not appropriate to express any view on any of these points. However, it may be said that we understood Sir Crispin to contend that it would not be necessary to establish to what extent production was affected, provided some clear and not insignificant adverse impact could be proved. He referred to observations of the Court of Appeal in the T.A.Gwillim caseat para . We note that the passage cited might assist the appellant if concurrent causes of loss were being considered but that it did not address the question of a separate potential cause of the whole loss of production. We need express no view on this issue at this stage.
 We consider that the case can go to proof and we see no need for a preliminary hearing. However, we welcome suggestions from parties as to the most efficient further procedure and if it is felt that either a further hearing, or an order for more explicit pleadings, would lead to more efficient disposal of the case, parties should not hesitate to discuss such matters further with the Principal Clerk.
For the appellant: Sir Crispin Agnew of Lochnaw QC; Leyshons, Solicitors, Peebles
For the respondents: Donald Cameron, Advocate; Scottish Government Legal Directorate