Gaeilge

Title:	Word Perfect Translation -v- Minister for Public Expendituire & Reform
Neutral Citation:	[2019] IESC 38
Supreme Court Record Number:	111/2018
Court of Appeal Record Number:	203/2018
High Court Record Number:	N/A
Date of Delivery:	05/27/2019
Court:	Supreme Court
Composition of Court:	O'Donnell Donal J., McKechnie J., MacMenamin J., Dunne J., Finlay Geoghegan J.
Judgment by:	O'Donnell Donal J.
Status:	Approved
Result:	Appeal dismissed

---

O’Donnell J.
McKechnie J.
MacMenamin J.
Dunne J.
Finlay Geoghegan J.

BETWEEN:

Judgment of O’Donnell J. delivered the 27th day of May, 2019

Introduction
1 The Office of Government Procurement (“the OGP”) operates as an office of the Department of Public Expenditure and Reform, with responsibility for sourcing all goods and services on behalf of the public service. On 12 October 2015, the OGP published a request for tenders to establish a multi-supplier framework agreement (“the Framework Agreement”) for the provision of interpretation services (excluding Irish) in the public service. The Framework Agreement was established on 25 January 2016, and contained eight lots. These proceedings concern Lot 4, which was for the provision of interpretation services to the immigration services, including, inter alia, the Refugee Appeals Tribunal, and to the Legal Aid Board. Three entities were appointed to Lot 4, including Word Perfect Translation Services Limited (“Word Perfect”) and Forbidden City Limited, trading as Translation.ie (“Translation.ie”). Word Perfect is the applicant in these proceedings and respondent to the present appeal, and Translation.ie is the successful tenderer in the mini-competition which is the subject of the proceedings.

2 It appears that a first mini-competition under Lot 4 was cancelled after a dispute arose, and a second mini-competition was commenced by revised supplementary request for tender (“the SRFT”) issued on 7 December 2016, with an initial deadline for receipt of tenders of 6 January 2017. Following a process of clarification, amendments were made to the award criteria and the scoring scheme and the deadline was extended to 17 January 2017. The estimated value of the mini-competition was €300,000 to €600,000, and the term of the contract was to be for one year. It is a sobering thought that, while this case is being prosecuted by the parties and the courts involved with considerable speed and efficiency (the High Court and the Court of Appeal having produced comprehensive judgments on the substantive issues as well as a number of important interlocutory matters), the time period for the intended contract has now lapsed, and the costs involved in litigation must surely be out of all proportion to the value of the contract to the State parties involved and the respective tenderers. While the regime of public procurement and judicial review of awards is an important and beneficial scheme in ensuring fairness, objectivity and relative transparency in the award of public contracts, it is important that the process be capable of functioning in an effective manner, and ensures that the public service obtains the benefit of the services tendered for.

3 On 18 April 2017, Word Perfect was notified that Translation.ie was the preferred tenderer. The letter of 18 April 2017 also informed Word Perfect of the relative marks achieved by it and the preferred tenderer. Importantly, the letter showed that the result had been very close: out of a total of 1,000 marks available, Translation.ie had achieved 885 marks and Word Perfect 870, a margin of 15 marks from 1,000. While the marking scheme was clearly weighted 75 to 25 between quality of service and cost, it was nevertheless clear that Word Perfect had made the most financially attractive offer, and had received the full 250 marks available for cost, whereas Translation.ie had scored only 166 in that section.

4 Word Perfect then commenced these proceedings pursuant to S.I. No. 130/2010 - European Communities (Public Authorities’ Contracts) (Review Procedures) Regulations 2010 (“the 2010 Regulations”), seeking a review of the award on a number of grounds. In this court, the dispute has, however, narrowed to two issues. First, under the heading “Service Delivery Plan” in the SRFT, Translation.ie had scored full marks of 250, whereas Word Perfect had been awarded 197.5. The reason for Word Perfect’s lower score was explained. The tender submitted had exceeded the mandatory word limit by 37 per cent. Accordingly, only the first 2,000 words of Word Perfect’s tender under this heading had been taken into account. Furthermore, the letter gives the following additional explanation:-

“Translation.ie (250 marks)

An excellent response covering all the requirements set out in the SRFT, demonstrating a clear understanding of the business requirements of clients and providing convincing and comprehensive evidence of ability to provide robust and flexible provision of a sustainable, consistently high quality, reliable service on a 24/7/365 basis. The proposal contained clear information on methods employed to ensure interpreters retain their skills.

Your tender (197.5 marks)

A very good response in relation to reliability of service in the proposal to meet urgent requests, providing a detailed outline as to how bookings will be received, confirmed and activated on a 24/7/365 basis. However, your response did not include any proposal in respect of the methods employed to ensure that interpreters will retain their skills in the language and remain up to date with their practice and fluency to a sufficient standard to ensure effective delivery of the service in all four language groups.”

5 A second area of controversy related to the Quality Assurance Plan. Here, 200 marks were available, and Word Perfect, with full marks, had outperformed Translation.ie, which had achieved 170. Here, however, the reasoning offered was more limited. The obligation under Regulation 6(3)(a) of the 2010 Regulations was that the notice to the unsuccessful tenderer should include “the characteristics and relative advantages of the tender selected” (emphasis added). Accordingly, the letter of 18 April 2017 noted that the Word Perfect tender had been an “excellent response with excellent use of visualisation tools. A very comprehensive easy to understand narrative summary is also provided. Fully meets requirements”. The explanation in relation to Translation.ie was simply: “[a]s your tender scored higher than the successful bidder, there are no relative advantages”. The same text was repeated in the second portion of the Quality Assurance Plan relating to complaints management. While it was argued in the High Court (see [2018] IEHC 237) and the Court of Appeal (see [2018] IECA 156) that this failed to sufficiently comply with the obligation to set out reasons, this argument failed in both courts, and, in the event, was not pursued on this appeal. As Hogan J. pointed out in the Court of Appeal in the substantive judgment, and also in the interlocutory judgment delivered on the question of discovery, the directives and implementing regulations must find a balance between the fact that the tender process encourages parties to submit commercially sensitive information, and the necessity of nevertheless permitting a full challenge to a decision. This balance may necessarily curtail the amount of information disclosed in the notice to the unsuccessful tenderer.

6 While it will be necessary to return later to the issues arising in relation to the Service Delivery Plan and the Quality Assurance Plan in much greater detail, at this point it is sufficient to note that Word Perfect contended, and the Court of Appeal agreed, that the evaluation decision contained manifest errors in both awarding Translation.ie the full 250 marks in respect of the Service Delivery Plan, and in awarding 170 marks to Translation.ie on the Quality Assurance Plan. It is perhaps worthy of comment that both complaints deal not with the evaluation of the bid of the applicant, but rather with the way in which the successful bid was marked. This is permissible and perhaps unavoidable in any competitive and comparative bidding system: in theory, the complaint that the winner was marked too generously is as valid as a complaint that the loser was marked too harshly. The test of manifest error also means that the focus is inevitably on individual decisions rather than the overall process. But the combination of these features has the somewhat unfortunate result that there is a very close analysis of only limited areas of what is a complex and quite sophisticated evaluation process.

7 Public procurement challenges can be particularly difficult and give rise to more significant interlocutory proceedings than is the case in more straightforward litigation. The challenging party has only the limited amount of information available in its own tender, the clarification procedure, and the so-called “regret” letter. On the other hand, once proceedings are commenced, the award of a contract is required to be stayed automatically unless a court otherwise orders. The effect of such a stay, or its absence, can have a significant practical effect on the outcome of the dispute. On one hand, the inability to award a contract for the duration of the stay exerts significant pressure on the contracting party. On the other hand, if the stay is lifted, the unsuccessful tenderer is left to an uncertain and limited remedy in damages, with the consequence that the proceedings may no longer be worthwhile. In this case, the appellant applied to the High Court pursuant to Article 8A of the 2010 Regulations to lift the automatic stay. The High Court acceded to that application (see [2018] IEHC 1) but the Court of Appeal overturned that decision and reinstated the stay on terms that Word Perfect undertake to prosecute the proceedings with urgency (see [2018] IECA 35).

8 Word Perfect also made an application for discovery. The High Court granted the discovery sought in part, but refused the application for discovery of Translation.ie’s successful tender (see [2018] IEHC 158). On appeal, the Court of Appeal (Hogan J.; Peart and Gilligan JJ. concurring) (see [2018] IECA 87) delivered a comprehensive judgment reviewing the decision of the CJEU in Varec S.A. v. Belgium (Case C-450/06) [2008] E.C.R. I-581, in which the court emphasised the importance of confidentiality in the procurement process so that tenderers should be able to communicate relevant information to the contracting authority without fear that the authority will communicate the information to rivals. This obligation imposes a significant constraint on the extent to which documentation should be discovered in the course of any proceedings challenging the award, lest, as the CJEU observed, it should become desirable to challenge an award in part with a view to obtaining such information. Hogan J. concluded that as a result, a court should not simply apply the traditional test of relevance as outlined by Brett L.J. in Compagnie Financiere et Commerciale du Pacifique v. Peruvian Guano Co. (1882) 11 Q.B.D. 55 and quoted with approval by Kenny J. in Sterling-Winthrop Group Ltd. v. Farbenfabriken Bayer A.G. [1967] I.R. 97, but rather should only make discovery of documentation where it had been convincingly established as indispensable for the fair disposal of the procurement challenge. This in turn required a close and precise focus on the case as pleaded, which in this case led to a different outcome in the two areas of challenge which are now still live on this appeal.

9 Hogan J. ordered the appellant to make discovery of that portion of Translation.ie’s tender which dealt with the Quality Assurance Plan, because Word Perfect’s pleaded case was that it did not know why Translation.ie had been awarded 170 marks. This in turn was a consequence of the limited reason given by the contracting authority: i.e. that in this aspect of the tender, Word Perfect had achieved a higher score, and therefore there was no relative advantage to the Translation.ie tender. On the other hand, in relation to the Service Delivery Plan, Word Perfect had pleaded that it did not accept the evaluation that its tender did not contain any proposal in respect of the method to be employed to ensure that interpreters retained their skills. Hogan J. held that Word Perfect was not entitled to discovery of Translation.ie’s tender in this respect, because Word Perfect’s case focused upon the evaluation of its own tender. Nevertheless, he ordered that any discovery would be limited to named solicitors and counsel involved on behalf of Word Perfect and would not be disclosed to the client.

10 The decision on discovery was not the subject of appeal to this court. It is plain, however, that, as observed above, in any competitive marking system a loser may well complain either that the marks awarded to it were too low, or that the award of marks to the successful party was too high. Formidable problems are created with regard to the protection of commercially sensitive information when an unsuccessful tenderer seeks to challenge the assessment of the successful tender submitted on terms of confidentiality to the contracting authority. On the other hand, the limitation on discovery may make it particularly difficult to challenge the award. Here, Word Perfect obtained discovery of the preparatory documents, and as a result sought to amend its pleadings to include an allegation that the award of full marks to Translation.ie for the Service Delivery Plan was a manifest error – a challenge which succeeded in the Court of Appeal – but did not obtain further discovery of the Translation.ie tender in this respect. It is important, therefore, to keep in mind the limited information on which the High Court and Court of Appeal proceeded on the issue of the Service Delivery Plan, and the contrasting fact that both courts had available the full tender insomuch as it related to the Quality Assurance Plan.

11 It is apparent that there are real difficulties here. An unsuccessful party cannot be permitted to gain access to the full tender of the successful party simply by alleging that the marks awarded to the latter were too high and constituted a manifest error. But as long as that is a viable ground for challenge, it is difficult to see how it can be advanced without sight of the tender, particularly where the reasons given are limited. It may be that if more detail is given in the reasons letter, that would mean that discovery would not require to be ordered unless the challenger could make a convincing case from the reasons that a manifest error could be in principle identified from the existing material. Even then, however, the question of the correctness of the marks, or at least the absence of manifest error in that regard, may require some consideration of the marks and the tenders submitted. There is no perfect solution to this problem, and there may be no other course than to proceed carefully and rigorously, as the Court of Appeal did in this case, in attempting to balance the different public interests involved.

The marking scheme
12 An important part of this case, perhaps not focused upon before now, is the amended marking scheme upon which the evaluation proceeded. The version of the marking scheme as originally published to the tendering parties is that set out in the Court of Appeal judgment. This was amended (although not materially as far as these proceedings are concerned) in the course of the clarification process by a communication issued to the parties at 16.01 on 19 January 2017. This provided as follows:-

Weighting	Meaning
80 – 100 %	Excellent response that fully meets or exceeds requirements and provides comprehensive and convincing assurance that the tenderer will deliver to an excellent standard.
60 – 79 %	A satisfactory or very good response that demonstrates real understanding of the requirements and convincing assurance that the tenderer will deliver to a very good or high standard.
30 – 59 %	A response where reservations exist. Lacks full credibility/ convincing detail and does not provide confidence to the contracting authority that the required services will be successfully delivered.
1 – 29 %	A response where serious reservations exist. This may be because, for example, insufficient detail is provided, or the response has fundamental flaws, or is seriously inadequate or seriously lacks credibility with a high risk of non-delivery.
0	No response or partial response only and poor evidence provided in support of it: failure to meet the requirements.

13 The tenderers had previously been informed that there was a minimum pass mark required of 60 per cent for each section, below which tenderers would be eliminated from the competition.

14 Any marking scheme implies a form of rigorous scientific measurement. In fact, as any Leaving Certificate student knows, this is not necessarily the case where subjective evaluation is concerned. In such circumstances, the marks awarded are not necessarily objectively verifiable and repeatable measurements. They are instead subjective assessments reduced to marks, which are intended to provide some overall appraisal, and the capacity for relative comparison between candidates. In such a case, the purpose of a marking scheme is to provide as much objectivity and transparency to the process as possible, so as to promote objective assessment in the first place, and facilitate review thereafter.

15 In that context, it is apparent, albeit with the considerable benefit of hindsight, that the empathetic and encouraging management consultancy language of the methodology contains some ambiguities. In particular, the reference to “fully meeting and exceeding requirements” is difficult in itself, but is particularly so when paired with a range of marks running from 80 to 100 per cent. Clearly, there is a wide variation possible between tenderers which rank just above the 80 per cent threshold, and those which achieve full marks. This is particularly important, since, given the automatic failure for any tender falling below the 60 per cent mark in any requirement, the outcome of the tender is largely determined by the marks in the top two categories. It may have been clearer if the top category had simply been described as an excellent response providing comprehensive and convincing assurance that the tenderer will deliver to an excellent standard, with the individual marks allocated indicating their relative merit within that category.

The evaluation
16 As already observed, the competition in respect of this tender was particularly close. In fact, Lot 4 contained two closely related areas where translation services were required, that is, the immigration services and the Legal Aid Board. As it happened, Word Perfect had been the incumbent provider to the immigration services, and Translation.ie the incumbent provider to the Legal Aid Board. There were three formal evaluation meetings leading to the final award, and the appellant (via the OGP) has made discovery of the detailed evaluation documentation, including handwritten annotations of the respective meetings. A first evaluation meeting took place on 27 January 2017, shortly after the closing date for the receipt of tenders. The marks given were “holding marks”. On this preliminary evaluation, Work Perfect scored 878 out of 1,000 marks, and Translation.ie scored 833. The second evaluation meeting took place on 1 March 2017. On this evaluation, Word Perfect scored 874 and Translation.ie had moved ahead, having been scored at 885. A final evaluation took place on 27 March 2017, and in the final scoring the margin had increased slightly, in that Word Perfect’s score had been reduced to 870, and Translation.ie scored 885. As observed by the Word Perfect, at the close of the evaluation process there was only 15 marks in the difference out of a total of 1,000, in circumstances, moreover, where in the initial evaluation there had been a difference of 45 in Word Perfect’s favour. Furthermore, the key increases to the Translation.ie’s score between the first evaluation and the second and third evaluation were in the area of the Service Delivery Plan (where its scored increased by 25 marks) and in the Quality Assurance Plan (where its score increased by a cumulative 23.5 marks). These were the two areas in which it was alleged by Word Perfect that the assessment had contained manifest errors. Thus, it was asserted, correctly, that the scoring in these criteria made all the difference to the outcome of the competition.

Evaluation of the Quality Assurance Plan
17 The original SRFT issued on 7 December 2016 identified the framework clients as the Irish Naturalisation and Immigration Service, the Office of the Refugee Applications Commissioner, the Refugee Appeal Tribunal, the Reception and Integration Agency, the Irish Refugee Protection Programme, the Office for Promotion of Migrant Integration, and the Legal Aid Board. Section 3 of the SRFT set out the minimum requirements. The award criteria included an important note that tenderers should satisfy themselves that they “fully understand the specific requirements of the Immigration Services and of the Legal Aid Board and shall take this framework client’s specific requirements into account when providing responses to the SRFT”. It also emphasised that “tenderers must demonstrate how they propose to meet the specific service requirements that the Immigration Services and the Legal Aid [Board] have set out in this SRFT while adhering to the service levels set out in Schedule K of the agreement …” (emphasis in the original). The Quality Assurance Plan was set out at section 4.3. It stated that the immigration services and the Legal Aid Board required an effectively and efficiently managed interpretation service. Management information reporting was critical to this function: “[t]enderers must explain in their tender how they will provide high quality, well presented and accurate Management Information Reports (“MI Reports”)”. Accordingly, the tenderers were provided with monthly management information in a spreadsheet and were required to use the notional information and notional data to prepare three sample “bespoke reports”. The tenderers were required to provide three such sample reports for one calendar month. In the event, as part of the clarification process this was reduced to two bespoke sample reports, described as “Report 1: Contract Volume and Value (“CV Report”)”, for which 90 marks were available, and “Report 2: Complaints Management (“CM Report”)”, to which 110 marks were allocated. There were then a number of bullet points, as follows, which it is necessary to set out in full:-

• “In all cases the MI Report must clearly distinguish between telephone interpretation services and On Site Services;

• MI Reports must convey information to management in a clear and transparent manner and in a format that allows framework clients to understand the information. The detail provided must be sufficient for management to have no need to review source data other than for verification purposes;

• Sample monthly MI Reports must be displayed in an easy to read format using PDF, Open Office, or Microsoft Office readers. MI Reports must use relevant visualisation tools such as charts, graphs, etc. to display pertinent information (as set out in Appendix 2 of this SRFT) graphically, with clear labelling of titles, legends, values, axis, categories, trend lines, scales, data source, grid lines, data points, etc;

• Each report must include a narrative summarising the information provided, including how they will insure that these MI Reports will be provided to framework clients in a timely manner;

• The successful tenderer will be required to produce the MI Reports in the same format as that with their tender.”

18 Subsequently, in the same page it is stated that “responses will be evaluated on accuracy, clarity of presentation, ease of interpretation and relevance to the requirements of the framework clients as set out above”. Word Perfect submitted a request for clarification in strikingly aggressive terms:-

“Dear Sirs,

Please answer the questions in full and be specific with your answers. Do not be vague and leave your answers open to interpretation.

Please be transparent in your answers and don’t be ambiguous which leads to cancellation of tenders afterwards. We don’t want to see the same scenario at debrief meetings where your answers will be “you were very close but” and “the other company went the extra mile”.

Don’t expect and don’t envisage scoring bidders higher on criteria that wasn’t requested. Be transparent in order to avoid unnecessary delays which have had huge effect on the public purse and tax payers’ monies.

Respond with accuracy, clarity, ease of interpretation, and relevance to the questions asked.”

19 In relation to the Quality Assurance Plan, the request for clarification highlighted the sentence, “each report must include a narrative summarising the information provided, including how they insure that these MI Reports will be provided to framework clients in a timely manner”. The request for clarification stated:-

“on a plain reading this requirement appears to say that the management reports to be submitted periodically by the successful bidder over the course of the contract must include the narrative referred to. That does not seem to make sense. Please confirm that what is meant is that the tenders to be submitted by participating framework members must include a narrative summarising the information that will be provided in the management reports over the course of the contract, and explaining how the tender will insure that its management information reports will be provided to framework clients in a timely manner? If we are incorrect, please advise and clarify your comments in detail.”

20 This request gave rise to a clarification response which was issued to all tenderers on 22 December 2016. The response was:-

“You are correct in your reading of the SRFT that the requirement is that tenderers must include a narrative in each report summarising the Notional Information provided therein as this will be required from the successful tenderer going forward for the duration of the contract.

In addition, for the purposes of this SRFT, tenderers are also required to explain by way of a separate narrative how all monthly reports will be provided to Framework Clients in a timely manner for [sic] the duration of the contract.”

21 With the considerable benefit of hindsight, it is apparent that there was an ambiguity in the initial statement which did not distinguish between the narratives provided, and that this ambiguity was perhaps not entirely dispelled by the statement that “you are correct in your reading of the SRFT”. However, it is tolerably clear that a narrative was undoubtedly required, and, moreover, one which summarised the notional information which had been provided to tenderers and which was to be set out in the MI Reports.

22 As already set out, Word Perfect scored a full 200 marks under this heading. The final evaluation noted an “excellent use of visualisation tools. A very comprehensive easy to understand narrative summary is also provided. Fully meets requirements”. In the proceedings however, the focus has been upon Translation.ie’s marking under this heading. The final evaluation noted in respect of the contract volume and value aspect “good overview of the proposal to deliver high quality MI Reports based on clear metrics and centralised data. Visualisation tools were provided however these were not displayed in an easy to read format. Tender also failed to develop the narrative component of the report”. In this regard, it may be recalled that the SRFT stated that “each report must include a narrative summarising the information provided”. On the first evaluation, Translation.ie was marked at 58.5 marks out of 90, being just over the 60 per cent required. In the final evaluation, this was increased to 76.5. In respect of complaints management, the evaluation again noted that the tender “failed to develop the narrative component of the report”. Initially, on first evaluation, 88 marks out of a total of 110 was awarded. In the final evaluation this was increased to 93.5. Accordingly, Translation.ie ultimately received 170 marks out of a total of 200. This meant in turn that the Translation.ie tender had achieved 85 per cent of the marks available under this heading, and accordingly under the scoring methodology had been assessed to have provided “excellent response that fully meets or exceeds requirements, and provides comprehensive and convincing assurance that the tenderer would deliver to an excellent standard”.

23 The two tenders were discovered and considered by the High Court and Court of Appeal in turn. The Word Perfect tender set out a sample Complaints Management Report with graphic illustrations with the complaints received broken down by language, time, etc. It also contained a two page “narrative summary”. A similar structure was followed in the Contract Volume and Value Report. By contrast, the Translation.ie submission contained some introductory text which did not, however, summarise or refer to any of the information. The information itself was set out in graphic form, with certain tabular summaries included where the information was presented by service, language group, and agency. There was no doubt that the evaluators were fully entitled to consider that the Word Perfect tender was superior in this regard, but it was argued by Word Perfect that, in the absence of any text containing a narrative summary of the information, the Translation.ie tender could not be considered to “fully meet the requirements of the SRFT”, and accordingly it was a manifest error to accord Translation.ie more than 80 per cent of the marks under this heading.

24 In the High Court, Barrett J. concluded that this was not a manifest error. At para. 30 of his judgment, he said:-

“One question that arose at hearing was whether [T]ranslation.ie had in fact provided a narrative. Certainly the narrative it provided was not as good as that of Word Perfect. But that is not the test (and Word Perfect in any event scored higher); the test is whether [T]ranslation.ie provided a narrative. In this regard, the court asked at hearing ‘what is a narrative?’ In reply it was referred, inter alia, to certain dictionary definitions. Thus, the Collins Online Dictionary defines a ‘narrative’ as “an account, report, or story…”. The Oxford English Reference Dictionary (2nd ed.) describes a ‘narrative’ as “a spoken or written account of connected events in order of happening”. Counsel for the Minister noted that the verb to narrate comes from the Latin ‘narrat’ which means ‘to relate’ or ‘to tell’ and thus what one is looking for in this regard is something that tells a story, an approach which accords with the Collins and Oxford definitions.”

25 The High Court considered that it was undeniably the case that:-

“.. the coloured charts, bars, graphs and miscellaneous data supplied by SRFT in this regard, together with certain written, comparatively limited commentary do ‘tell a story’, do comprise a narrative and do otherwise conform with the requirements of the SRFT.”

26 This conclusion was reversed by the Court of Appeal. Hogan J. considered himself compelled to conclude that there was a manifest error in this regard on behalf of the evaluation team. The word narrative was a fairly ordinary English word which a reasonably diligent and well-informed tenderer would well understand as requiring a written statement. This was especially so in the context of the actual language of the tender document itself (“… a narrative summarising the information provided …”). Any doubts would be dispelled by the fact that the SRFT had stated just a few lines previously that the MI Reports “must use relevant visualisation tools” such as charts, graphs, etc. to display pertinent information graphically. He concluded that “[o]n any view, therefore, this relevant provision of the SRFT required the tenderer to supply both a visual display or depiction of the information by way of charts, graphs etc., and a narrative summary. Some might think that the narrative summary requirement was superfluous, but there is, I fear, no escaping the conclusion that, viewed objectively, this is what the SRFT actually required” (emphasis in the original). He concluded that there was no doubt that Translation.ie did not provide such a narrative summary in this case, and, accordingly, that the score of 85 per cent, which implied that Translation.ie had fully met or exceeded requirements, was therefore a manifest error.

The appeal to this court
27 By a determination issued on 23 October 2018, leave to appeal to this court was granted (subject to any refinement or argument at case management) on the following issues:-

“What is the test by which a tender award under public procurement legislation may properly be reviewed by a court and was that test met in respect of the two grounds as to narrative statement and professional development whereby the tender award was overturned by the Court of Appeal?”

28 In essence, therefore, the issue for determination in relation to the Quality Assurance Plan resolves itself into consideration of the meaning of the requirement in the SRFT for a narrative summarising the information when viewed against the marking methodology.

29 The question in relation to the Service Delivery Plan is equally net, although a little more difficult to isolate. In this regard, it will be recalled that the original complaint made by Word Perfect after receipt of the “regret” letter was that there had been a manifest error in the evaluator’s conclusion that Word Perfect had failed to satisfy the requirement in the SRFT to “ensure” that interpreters retained their skills. That decision of the evaluators was in turn related to the decision that Word Perfect had exceeded the word limit, and that the appropriate method of dealing with this was to limit the evaluation to the initial 2,000 words. This issue, and the consequent conclusion that Word Perfect had failed to satisfy the requirement in the SRFT, was not in issue in this appeal. However, Word Perfect had also argued that Translation.ie could not have been awarded full marks under this heading because, it was argued, it also had failed to provide sufficient information to ensure rather than simply encourage retention of skills by interpreters.

30 The SRFT provided that the Service Delivery Plan must address four bullet points, including that:-

“Tenderers must set out the methods employed to ensure that interpreters will retain their skills in the language and remain up to date with their practice and fluency to have sufficient standard to ensure effective delivery of the service in all four language groups.”

31 As already observed, Word Perfect did not have access to the Translation.ie tender, but sought to rely on the evaluators’ reports arising from the three evaluation meetings. It was originally proposed that Translation.ie would be awarded a 90 per cent mark as a minimum, subject to change depending on clarification. It was stated “it is noted that in relation to retention of language skills ‘Translation.ie encourages ongoing training by providing and recommending the participation in workshops, seminars, formal course work as well and informal learning opportunities in order to remain up to date with current developments in the field’. However, a formal policy in respect of this training would have given a higher level of assurance.”

32 Word Perfect invite the court to infer, not unreasonably, that the portion contained in the internal quotation, that is, starting with the words “Translation.ie encourages ongoing training…”, is drawn from the Translation.ie tender. However, formal clarification was not sought, because it appears the Chief State Solicitor’s Office advised that seeking clarification would or could be interpreted as constituting a material change and thus undermine the tender process. Again, it may be recalled that this was the second SRFT for Lot 4, the first having been abandoned as a result of a dispute raised by Word Perfect. Accordingly, the final evaluators’ report noted, “we discussed 90 per cent holding mark with the possibility of marking up once clarification is received in relation to encouraging and recommending. However, CSSO have advised that seeking clarification on this would be seeking a material change”. This was further elaborated on in an affidavit from Ms. Anne Lannon of the OGP, which recorded that, having received the advice from the CSSO, the evaluation panel then proceeded to re-examine Translation.ie’s response, and after discussion decided that clarification was not necessary in order to establish that the tender did meet all stated requirements, and that they were happy with the response. It was also argued that from an analysis of the hand written notes on the evaluation documents by the evaluation team, the management of interpreters, including such elements as shadowing of more skilled interpreters, was viewed as sufficient to ensure that skills would be maintained.

33 The High Court concluded that there was no manifest error in the manner in which the evaluation committee had approached its decision. The Court of Appeal, however, disagreed. It stated that the fundamental complaint was that the Translation.ie tender had merely stated that it would encourage skills retention on the part of interpreters, but the tender did not state how this would be ensured. At para. 30, the judgment continued:-

“So far as can be seen, much the same could – more or less – be said with equal measure in respect of the Word Perfect tender, because it too had itemised matters designed to encourage its translators to retain their skills, by, e.g., encouraging interpreters to attend continuous professional development training as distinct from requiring them to do so as a condition of their contracts of employment.”

34 Hogan J. concluded that the Translation.ie document simply referred to steps that would encourage rather than require interpreters to retain their language skills, and accordingly he found himself compelled to conclude that this too amounted to a manifest error, even allowing for the widest margin of appreciation which could fairly be permitted within the scope of the procurement rules. He concluded that both tenderers did not fully comply with the requirements of the SRFT, in that neither had demonstrated that they would ensure that their interpreters would retain their skills, rather than simply encourage them to. It was not that Translation.ie could not be considered to have outscored Word Perfect, but what could not be allowed to stand was the conclusion that Translation.ie had achieved a perfect score when – no less that their rival bidder – they submitted a tender which did not completely comply with the SRFT.

35 The issue in this aspect of the case also resolves itself to a question of analysis of what was meant by the requirement to “ensure that interpreters will retain their skills in the language” which was contained in the SRFT. The fundamental complaint appears to be relative. Word Perfect’s response was marked down on the basis that it did not include any proposal in respect of the methods employed to ensure that interpreters would retain their skills in the language and remain up to date with the practice and fluency. But it was suggested that, from the evaluators’ approach to the Translation.ie tender, the same, or something similar, could be said of it, albeit that the full tender was not the subject of analysis.

The law
36 There is little dispute but that the applicable law in this case was established in the leading case in this jurisdiction, SIAC Ltd. v. Mayo County Council [2002] 3 I.R. 148. In that case, the Supreme Court held that the standard of review required was that applied in European law: that is, a standard of manifest error. Furthermore, the word ‘manifest’ should not be equated with any exaggerated description of obviousness. The study of the case law undertaken by Fennelly J. showed that the Community courts were prepared to annul a decision when an error had clearly been made. It is important, however, to understand that judgment against the background of the national law of judicial review, and indeed the decision appealed from. The High Court in that case had applied the irrationality test deriving from The State (Keegan) v. Stardust Compensation Tribunal [1986] 1 I.R. 642: that is, a consideration of whether the decision could be said to plainly and unambiguously fly in the face of fundamental reason and common sense. Fennelly J. observed that it was apparent that the applicable test should be “rather less extreme”. Accordingly, the manifest error test may be understood as a departure from the irrationality standard applied in domestic law. However, the distance there is in practice between the two tests as applied is perhaps more difficult. In that regard, useful guidance is also to be obtained from the judgment of the Supreme Court in SIAC. There, Fennelly J. said, at p. 176, that the courts:-

“ …while recognising that awarding authorities have a wide margin of discretion, must recognise that this cannot be unlimited. The courts must exercise their function of judicial review so as to make the principles of the public procurement directives effective. In the case of clearly established error, they must exercise their powers. The application of these principles may not, in practice, lead to any real difference in result between the judicial review of purely national decisions and of those which require the application of Community law principles” (emphasis added).

37 In other words, the test as applied must still afford a “wide margin of discretion” to the awarding authority. It is accepted that the tender must be interpreted by the application of the so called RWIND test – an acronym derived from a key passage in the SIAC decision when it was referred to the CJEU. At para. 42 of that judgment (SIAC Construction Ltd. v. Mayo County Council (Case C-19/00) [2001] E.C.R. I-7725), the CJEU said:-

“More specifically, this means that the award criteria must be formulated, in the contract documents or the contract notice, in such a way as to allow all reasonably well-informed and normally diligent tenderers to interpret them in the same way.”

38 As Finlay Geoghegan J. pointed out in Gaswise Ltd. v. Dublin City Council [2014] IEHC 56, [2014] 3 I.R. 1, at p. 10, adopting the following passage from the judgment of the CJEU in Commission v. Netherlands (Case C-368/10) [2012] 3 C.M.L.R. 11, at para. 109:-

“The principle of transparency implies that all the conditions and detailed rules of the award of procedures must be drawn up in a clear, precise and unequivocal manner in the notice of the contract document so that, first, all reasonably informed tenderers exercising ordinary care, can understand their exact significance and interpret them in the same way, and, secondly, the contracting authority is able to ascertain whether the tenders submitted satisfy the criteria applying to the relevant contract”.

39 The question of interpretation of the request for tender is a matter of law for the court. Remembering however that this is not an exercise in statutory interpretation, and that a request for a tender must be understood in the context of the relevant industry involved. The court in Gaswise observed at p. 11 that the court “should attempt to put itself in the shoes of a reasonably well informed and normally diligent tenderer who would be responding to this particular [invitation to tender], i.e. a person providing the relevant gas services, and should not do so as a lawyer”. As Barniville J. emphasised in Transcore L.P. v. The National Roads Authority [2018] IEHC 569, (Unreported, High Court, Barniville J., 17 October 2018), at para. 191, “the court must focus on the ‘industry’ concerned in which the professionals and persons involved are not lawyers but participants in that industry”.

40 The fact that there is no significant disagreement about the principles involved does not mean that there was not a considerable dispute in this case as to the application of those principles, nor that the resolution of that dispute may not be important in clarifying the law. Indeed, this case illustrates the fact that the application of a legal test to a precise factual situation may often be the best illustration of what is sought to be achieved by the test in question. Furthermore, there is a point at which the clarity of the principle annunciated begins to blur at the edges. Thus, while it can be said that the RWIND test is a matter of law and is therefore to be determined by the court, and at the same time a considerable margin of appreciation must be accorded to expert evaluation of decisions on the facts, there is a point at which these two principles meet. For example, a decision to afford high, or perhaps full marks for what is deemed full compliance with the requirements of the SRFT necessarily implies an interpretation of what the SRFT requires. Similarly, while the different standards of review may be readily stated and distinguished, as Fennelly J. noted in SIAC Ltd. v. Mayo County Council [2002] 3 I.R. 148, the point at which it might said that a decision does not fly in the face of fundamental reason and common sense, but yet is a manifest error, is not easy to identify in fact. For my part, I would be inclined to see that standard of manifest error not so much in terms of the patent nature of any error, but rather as relating to the degree of confidence with which it can be said that the decision was wrong. An error may not be apparent on the surface of the decision, and may be difficult to identify and explain, but if, once understood, it is clear that it is indeed an error, then the test is satisfied. That, indeed, is consistent with the concept of a margin of appreciation being afforded to the decision-maker. Often, however, an error which is clear will be manifest from the decision itself.

41 It may appear that the two issues in this case are very similar, and can be resolved to the meaning to be ascribed to one word in each of the relevant portions of the SRFT: the meaning of the word ‘ensure’ (in contradistinction to ‘encourage’) in the context of the Service Delivery Plan required in section 4.1 of the SRFT, and the meaning of the word ‘narrative’ in the Quality Assurance Plan, as required by section 4.3. However, the question is more complex, and deserves more careful consideration.

Decision in relation to the Service Delivery Plan
42 Having considered the matter carefully, I do not think it is possible to conclude that the decision to award Translation.ie full marks on the Service Delivery Plan was a manifest error within the SIAC test.

43 Word Perfect’s case in this regard is dependent on the increase of marks which occurred during the evaluation process. Word Perfect is perhaps understandably suspicious of the increase, because it has subsequently transpired that this increase was critical in reversing the provisional result of the first round evaluation which had Word Perfect ahead. Word Perfect also argues that if 90 per cent is the most the evaluators could award without the clarification which they considered necessary, and if they subsequently decided against such clarification on legal grounds, then logically they could not increase the mark to 100 per cent. Further, Word Perfect point to, and the Court of Appeal accepted, the fact that the same issue (of ensuring that interpreters retained their skills) appeared to have been relied on in awarding a lower mark to Word Perfect under this heading.

44 I think, however, that there is considerable merit in the High Court judge’s observation that this approach involves reviewing the process leading up to the decision, rather than the decision itself. Furthermore, a review of that process reveals, in my view, only a well-structured and conscientious approach to the evaluators’ task. The lead evaluator, Ms. Anne Lannon, gave sworn evidence of the reasoning process that led to the mark, which emphasised that the original mark was a “holding mark”, and that the evaluators considered that it would be unfair to withhold top marks from what was an excellent tender in this regard on that basis. This evidence was not challenged in cross-examination, and was, moreover, entirely consistent with the documentation produced on discovery showing the development of the evaluation process. The Court of Appeal’s observations (albeit on a different issue which was not the subject of appeal to this court) are in my view apposite here:-

“34 […] the task of the evaluators is already difficult enough. If they were required to explain possible changes in thinking between evaluation meetings prior to the final decision it would add new layers of complexity – not least in terms of discovery and oral evidence – to an already complex system of public procurement litigation.

35 Such a requirement would, moreover, stifle the necessary freedom which evaluators must have to reflect on the respective merits of the bids. The evaluators must be prepared to stand or fall by a review of the final published evaluation for manifest error. But short of that they cannot be expected to have to defend what are, at best, tentative or provisional views expressed during the course of the evaluation process.”

45 The implicit comparison made by the Court of Appeal in suggesting that the Word Perfect and Translation.ie tenders were broadly similar in that they both could be described as encouraging rather than ensuring the maintenance of skills is, in my view, also misplaced. The Word Perfect tender was marked down in this respect because it exceeded the word count and thus was treated as having no proposal in this regard. Furthermore, it is not possible to make a fair comparison between the tenders, since the court did not have the full Translation.ie tender, or even the portion dealing with the Service Delivery Plan, but had at best the fragment which appears to have been quoted in the evaluation process. On the basis of the material available, it might be contended just as plausibly that the mark awarded to Word Perfect was generous in all the circumstances, given the difficulties created by the word count. This also illustrates the fact that the question of ensuring the maintenance of skills was only one component of the Service Delivery Plan requirement. In order to construe that the marking was erroneous, still less a manifest error, it would be necessary to review both tenders in their entirety, and, quite possibly, the approach taken by the evaluators to marking more generally. Finally, in my view, the maintenance of a skill is a human endeavor, and it is doubtful whether the SRFT, in requiring measures to ensure the maintenance of those skills, should be understood as requiring some form of absolute and humanly impossible guarantee, so that if a tender fell short of such a guarantee, it must necessarily also fall short of a 100 per cent mark, even if it was in every other respect excellent, and indeed exceeded expectations. That is perhaps to take too lawyerly an approach to the interpretation of words in the SRFT, to the exclusion of the industry context. I would, accordingly, respectfully disagree with the conclusion of the Court of Appeal in this regard, and set it aside.

Decision in relation to the Quality Assurance Plan
46 As observed above, this issue is dependent upon the interaction between the interpretation to be accorded to the word “narrative”, and, just as importantly, the operation of the marking scheme. The appellant argued strenuously that the interpretation adopted by the Court of Appeal was too verbal and text-bound, and that it was possible to see the charts presented as satisfying the requirement of a narrative. In other contexts, there might be considerable merit in this submission. Lawyers tend to prefer information presented in the form of text, whereas others may be much more comfortable with the same information provided in a tabular or numerical form. If the information is provided effectively it would be an error to lay emphasis on the form, particularly if that may reflect an unarticulated preference for the familiar. However, in the context of the SRFT, the conclusion of the Court of Appeal that a narrative summary in text form was required is in my view compelling. This is what I consider to be the natural meaning of the words, but particularly when taken in their context, which refers separately to the use of tables. It is the information in such tables which is required to be summarised in the narrative. The appellant criticised the Court of Appeal for not affording a degree of “curial deference” to the exercise of “judgment calls” by the evaluators, or “judicial restraint” in relation to the actual marks awarded. Instead, the appellant contends the Court of Appeal failed to appreciate the “holistic approach” taken by the evaluation committee to the consideration of the criteria in section 4.3 of the SRFT in its totality. However, these appeals to deference and a holistic approach only confirm, in my view, that, if a clear and natural view is taken of the requirement of the SRFT, the Translation.ie tender must be considered to have failed to provide what was required in this component of the Quality Assurance Plan.

47 However, the evaluation marks in this regard, if taken in isolation from the detail of the marking scheme, might not themselves be capable of being a manifest error. If the evaluators, as they did, had given the same marks, and had explained that the lower mark given to Translation.ie was because of the absence of a narrative, I do not think it would be possible to argue that the particular marks given constituted a manifest error. However, Word Perfect points to the terms of the marking scheme and, in particular, the explanation that marks between 80 and 100 per cent could only be given for an excellent response that “fully meets or exceeds requirements”. As touched on above, the requirement that a tender fully meet requirements to achieve 80 per cent was perhaps not a fully intended and thought-through consequence of the methodology. It appears to create something of a straightjacket, which may lead inevitably to the conclusion that a response which is excellent in nearly every respect and fully meets or exceeds the requirements in all but one area, cannot achieve more than 79 per cent of the marks, even though it may be manifestly superior to a competing bid that did, in this example, provide some narrative text. While I do not think that this is a desirable conclusion, and, indeed, it may not have been intended, I consider that it follows from the language used in the SRFT methodology for calculating scores.

48 Accordingly, I agree with the conclusion of the Court of Appeal in this regard that the award of marks in this band to Translation.ie was a manifest error, by reference not to the marks awarded or the quality of the submission, but rather because of the definition given of the submission which would attract marks in this band. It is necessary to consider finally what consequence follows from this finding. The fact that there was a margin of 15 marks out of 1000 between the parties, and that the error was in awarding 170 marks when 160 represented a submission that fully complied with the tender in this regard, means that, in theory at least, it is not inevitable that the error identified had a causal effect on the outcome of the competition. The evaluators could have awarded marks under 160 which would still have meant that Translation.ie was the successful party, although by a wafer-thin margin. I do not think this can influence the outcome, however, and the result of the competition must be set aside. It is indeed regrettable that a small contract for an important service has not been capable of being awarded in circumstances where the original SRFT was issued in early 2016. From even the limited material which has been reviewed on this appeal, it seems apparent that the approach to the evaluation of the bids was admirably thorough and detailed, and that, thereafter, the legal challenge was prosecuted, heard, and determined with commendable speed, efficiency and care in both the High Court and Court of Appeal. Given the narrowness of the margin, and the changes in the marks (and ranking) as the evaluation progressed, it was perhaps inevitable that these proceedings would be particularly contentious and difficult, and, of course, the requirement for this elaborate public procurement process is not a matter of domestic law. It would, however, be unfortunate if difficulties in the review process led to tenders which avoided issues of quality in favour of determination criteria that are less liable to challenge. There was, after all, a reason why the tenderer did not wish to select the service provider solely by reference to price. However, consideration of this case does emphasise the need for precision in the drafting of a tender and the construction of the tender process, care in the evaluation, and, not least, pragmatism and efficiency in any review.

Conclusion
49 Returning, therefore, to the question upon which leave to appeal was granted and set out at para. 27 above, it is now apparent that there is little dispute that the applicable test is that first set out by Fennelly J. in SIAC Ltd. v. Mayo County Council [2002] 3 I.R. 148: that is, a test of whether the decision is vitiated by manifest error. The main dispute in this case was whether the test had been correctly applied in the judgment under appeal. For the reasons set out above, I conclude that, in one respect, namely the evaluation of the successful tender in respect of the Quality Assurance Plan, there was a manifest error, and accordingly the decision must be set aside. I would therefore dismiss the Minister’s appeal against the decision of the Court of Appeal.