Karoshoek Solar One (RF) (Pty) Ltd v Dankocom (RF) (Pty) Ltd and Another (2022/13538) [2023] ZAGPJHC 766 (5 July 2023)

begin wrapper begin container begin header begin slogan-floater end slogan-floater - About SAFLII About SAFLII - Databases Databases - Search Search - Terms of Use Terms of Use - RSS Feeds RSS Feeds end header begin main begin center # South Africa: South Gauteng High Court, Johannesburg South Africa: South Gauteng High Court, Johannesburg You are here: SAFLII >> Databases >> South Africa: South Gauteng High Court, Johannesburg >> 2023 >> [2023] ZAGPJHC 766 | Noteup | LawCite sino index ## Karoshoek Solar One (RF) (Pty) Ltd v Dankocom (RF) (Pty) Ltd and Another (2022/13538) [2023] ZAGPJHC 766 (5 July 2023) Karoshoek Solar One (RF) (Pty) Ltd v Dankocom (RF) (Pty) Ltd and Another (2022/13538) [2023] ZAGPJHC 766 (5 July 2023) Download original files PDF format RTF format make_database: source=/home/saflii//raw/ZAGPJHC/Data/2023_766.html sino date 5 July 2023 REPUBLIC OF SOUTH AFRICA IN THE HIGH COURT OF SOUTH AFRICA GAUTENG DIVISION, JOHANNESBURG Case Number: 2022/13538 NOT REPORTABLE NOT OF INTEREST TO OTHER JUDGES REVISED 05/07/23 In the matter between: KAROSHOEK SOLAR ONE (RF) (PTY) LTD Applicant and DANKOCOM (RF) (PTY) LTD First Respondent MOHAMMED ASHRAF EBRAHIM CHOHAN SC Second Respondent JUDGMENT KEIGHTLEY, J: Introduction [1] The applicant in this matter, Karoshoek Solar One (RF) (Pty) Ltd applies to review the award of the second respondent, Mr Chohan, who was the arbitrator in arbitration proceedings between the applicant and the first respondent, Dankocom (RF) (Pty) Ltd.  The application is brought in terms of section 33(1) of the Arbitration Act [1] (the Act).  Mr Chohan abides the decision of the court.  To simplify matters, I refer to Dankocom as “the respondent” rather than the “first respondent”. [2] The dispute that was referred to arbitration arose out of an Engineering, Procurement and Construction contract (the contract) concluded on 12 December 2014 in terms of which the applicant engaged the respondent to design and build a solar power plant in the Northern Cape.  During the course of the project various disputes arose. The parties settled a number of those disputes in a settlement agreement dated 12 July 2019.  However, the parties were unable to settle one of the disputes, consequent upon which that dispute was referred to arbitration. [3] The nature of the dispute is central to the application for review.  As such, it requires particular analysis.  To this end, the contractual context to the dispute is important, as is the route followed in the arbitration . The contract and the arbitration [4] The parties are agreed that the broad nature of the dispute involved the interpretation of certain of the provisions in the contract, although they differ on the ambit of the interpretational dispute and on which provisions or schedules to the contract are relevant . [5] Under the contract the respondent guaranteed that the solar power plant (the Facility) would achieve a certain energy capacity, namely 100 MW net of concentrated solar power.  The contract, and in particular, Schedule 8 thereto, described the tests to be performed to determine whether the guaranteed capacity, the EEOU Performance Guarantee (the guarantee) had been met.  For purposes of the present dispute, the Long Term Performance Test (LTPT) is relevant.  If the LTPT indicated that the guarantee had not been met, the respondent became liable for performance liquidated damages (PLDs) to the applicant.  This was in terms of Schedule 9 to the contract . [6] The contract envisages that the outcome of the LTPT is determined by a computer model, or Facility Power Model (the FPM) as it is described in the contract.  The contract states that the FPM is attached asSschedule 30.  Much of the applicant’s case hinges on Schedule 30 and its relevance, as I discuss later.  For present purposes, it is sufficient to note that the FPM is essentially a computer software model that uses input data, much of which is contained in comma-separated value data computer files (.csv files).  Using this data, algorithms and so forth, the FPM models the energy capacity of the Facility.  The success or failure of the Facility to meet the guarantee is based on the outputs determined through this modelling process . [7] In terms of the settlement agreement, which ultimately led to the arbitration, paragraph 8.4 recorded that: “ In regard to the operating strategy of the Facility and its impact on the EEOU Performance Guarantee, the Parties acknowledge that there is a dispute between the Owner and the Contractor, which dispute involves whether the Contractor is entitled to adjust the Facility Power Model (Schedule 30 to the EPC Contract) when determining the amount of the EEQU Performance Guarantee to take into account the actual operating strategy of the Facility as opposed to the default strategy . The Parties acknowledge and agree that their respective rights and positions in regard to the interpretation of Schedules 8 and 9 concerning the operational strategy are fully reserved and agree to discuss the matter further as soon as practicable after the conclusion of the Agreement, in order to attempt to reach agreement on the interpretation. If the Parties are unable to resolve the dispute concerning the interpretation the Contractor shall be entitled to refer the dispute to arbitration for determination under the dispute resolution clauses in the EPC Contract.” (Emphasis added) [8] One of the categories of reference data measured for purposes of the LTPT is the operational strategy for the Facility.  The operational strategy is intended to maximise revenue for the applicant by ensuring that the thermal storage system is used to its full capacity each day so as to facilitate the maximum sale of energy during peak times.  By running the computer model on the relevant reference data, the parties can determine whether the operating strategy is achieving the performance that was guaranteed by the respondent . [9] As indicated in the underlined portion of the extract above, the main difference between the parties was whether the LTPT was to be conducted based on reference data drawn from the facility operating strategy actually adopted (post the Facility coming on line), or reference data based on the default operational strategy devised around the time the contract was entered into, and hence prior to the Facility coming on line. The default model (contained in the FPM), which was provided by a separate service provider, used reference data based on assumed conditions.  It thus predicted what the performance of the Facility would be in the future, based on default, and not actual inputs.  The respondent contended that the contract permitted an adjustment of the FPM (constituting the default reference data relevant to the operational strategy) so as to take into account the actual operating strategy. The applicant, on the other hand, took the view that the default reference data relevant to the operational strategy was what the parties had agreed was to be used to the exclusion of actual reference data inputs that had become available once the Facility started operating . [10] The parties were unable to reach agreement on the disputed issue and the respondent referred the dispute to arbitration.  The key to the dispute was the interpretation of the relevant provisions of the contract . [11] Clause 11.6 of Schedule 8 described the categories of reference data that was to be used for the LTPT as follows (in relevant part): “ 11.6.1 Annual weather file …. 11.6.2  Restrictions (for example grid and fossil fuel supply restrictions) … 11.6.3  Operation of the Facility 11.6.3.1 Operation of the Facility according to Operation and Maintenance Manual delivered by the Contractor; 11.6.3.2 Facility Operation strategy as defined in the Facility Power Model. This item specifically refers to the strategy followed to charge the storage and maximize production during peak hours. 11.6.3.3. Standard Time considered as included in the Facility Power Model. 11.6.4  Facility outage: scheduled outage 12 days. 11.6.5  Solar Field: 11.6.5.1 Solar Field annual average availability: 99%; 11.6.5.2 total annual average cleanliness factor: 97%, calculated as the product of mirror cleanliness factor (98%) and absorber tube cleanliness factor (99%); 11.6.5.3 annual percentage of broken/damaged mirrors: 0.25%; and 11.6.5.4 annual percentage of broken/damaged absorber tubes: 0.5%. 11.6.6 The Post COD Guarantees will be adjusted for degradation … .” [12] As is apparent from the above, clause 11.6.3 describes the reference data for the operating strategy as being that defined in the FPM.  The idea behind the operating strategy is, as stated, to charge the storage and maximise the production of electricity during peak hours . [13] If this was all that Schedule 8 had to say on the issue, there would have been no dispute between the parties.  However, clause 11.7 of Schedule 8 is critical to the dispute.  It provides for the “Correction of guaranteed values”’ and states: “ 11.7.1 If there is any deviation from the reference data referred to in paragraphs 11.5.1 to 11.5.6 (both inclusive) as stated in paragraph 11.5 , all the results of the Facility Power Model must be corrected according to the principles described in paragraphs 11.6.3.1 , 11.6.3.2 and 11.6.3.3 . 11.7.2 Paragraphs 11.5.3 , 11.5.4 , 11.5.5 and 11.5.6 apply irrespective of whether the Operator and the Contractor are Affiliated. 11.7.3 For the avoidance of doubt, regardless who the Operator is reference data referred to in paragraphs 11.6.1 and 11.6.2 must in all cases be corrected for the Long Term Performance Test according to the following principles: 11.7.3.1 for parameters differing from the reference data (paragraph 11.5 ), the actual value of the parameters must be measured during the Test period and used as input data for the Facility Power Model using the methodology described in the user manual of the Facility Power Model and in the rest of the technical documentation. In case a parameter change cannot be directly converted into a change of an input of the Facility Power Model, the Owner, the Contractor, the Lenders’ Representative and the Operator must agree on the appropriate correction of the results to be applied; 11.7.3.2; if during the Long Term Performance Test there is any deviation of an Operating parameter which is an input of the Facility Power Model other than the defined reference data (paragraph 11.5 ), the inputs into the Facility Power Model must be modified by agreement between the Owner, the Contractor, the Le nders ’ r epresentative and the Operator if a change in any of these parameters causes a decrease in Energy Output in respect of which the Contractor is entitled to relief under the Contract; 11.7.3.3. the Owner must provide and facilitate any and all Operating data requested by the Contractor in order to correct the Post COD Guarantees as specified in Schedule 9. This is especially important, but not limited to the data that may not be extracted directly from the DCS of the Facility, such as field cleanliness factor, mirrors and tubes status and scheduled outages.” (Emphasis added) [14] The paragraphs referenced in bold in the above extract are relevant to the question of rectification, which is raised in one of the applicant’s grounds of review.  At this stage, it should be noted that, save for the reference to “11.6.1 and 11.6.2” in clause 11.7.3, the parties are agreed that the cross-referencing is incorrect.  Further, that the references to 11.5 and its subparagraphs, should be read as a reference to 11.6 and its subparagraphs.  Also, the reference to 11.6 and its subparagraphs in clause 11.7.1 should be read as a reference to 11.7 and its subparagraphs.  I return to this issue later, but it is important to understand that clause 11.7 should be read accordingly. [15] The respondent relied on clause 11.7 to support its case that a deviation from the default reference data contained in the FPM was required.  Ultimately, the question was how these two clauses of Schedule 8 were properly to be interpreted . [16] However, before the arbitrator could proceed to the main dispute between the parties, he was tasked with resolving an interlocutory dispute, which I will refer to as the disclosure application.  The FPM (in Schedule 30) and its constituent .csv files comprised confidential information.  Although the applicant had access thereto, without agreement from the respondent, or without an award by the arbitrator, it could not disclose the FPM to its legal representatives and its witnesses.  The applicant wished to disclose the FPM to its lawyers and its expert witnesses and when the respondent refused, the applicant applied to the arbitrator for an interlocutory award permitting such disclosure. [17] The respondent opposed the disclosure application, contending that, save for the FPM manual (which was not confidential information), the FPM itself and its .csv files, contained in Schedule 30, were irrelevant to the main dispute between the parties.  The main dispute was a narrow one, relating simply to the interpretation of Schedules 8 and 9 of the contract, and not Schedule 30.  More particularly, contended the respondent, the dispute was whether the parties had intended that the operation strategy should be fixed, with reference to the default operation strategy, or whether it ought to be variable, with reference to the input data drawn from the actual operation strategy.  The FPM model, and its input reference data in the form of the .csv files, could not assist in determining this interpretational dispute. [18] The applicant maintained that Schedule 30, comprising the FPM and .csv files, was relevant to the dispute as pleaded.  This was so on a proper analysis of the respondent’s statement of claim in the main dispute and, in particular, on certain declaratory relief that had been sought. [19] During the course of the disclosure proceedings, the respondent amended its statement of claim so as to limit its declaratory relief in prayer 2.  It also abandoned an alternative prayer, prayer 3, which, according to the arbitrator in his award (the disclosure award), went beyond the mere interpretation of the contract.  It is not necessary to discuss this abandonment of prayer 3 any further, save to note that the arbitrator recognised that had the respondent not abandoned the alternative relief in prayer 3, it would have necessitated an analysis of the actual workings of the FPM and the .csv files.  According to the arbitrator: “The contractor, recognising this difficulty, consequently abandoned the relief sought in prayer 3”. [20] Once the respondent had scaled down the ambit of its claim for relief, the arbitrator agreed with the respondent that: “ In my view, the now limited prayer 2 must be read in context and against the articulated dispute pleaded by the contractor in its statement of claim. Doing so reveals that it does no more than seek to achieve a principled view on whether or not the operation strategy in the FPM should comprise the actual operation strategy as it alleges it should, or whether it should include the default operation strategy. It does not seek to achieve an award on the correctness or otherwise of the data to be used in the FPM .” (emphasis added) [21] The arbitrator found that: “ Having regard to the pleaded issues, the most important of which I have referred to above, it is quite clear that what has been referred to arbitration is an interpretational dispute . Whilst the outcome of the interpretational dispute of the EPC contract may have subsequent consequences on other disputes between the parties, including whether or not any input data to the FPM was correct or not, or whether there is any liability for liquidated damages, those disputes have not been referred to arbitration. Neither the FPM nor the .csv data files are relevant to the interpretation of the EPC contract and in particular, whether the input data to the FPM ought to refer to the actual operation strategy or the default strategy. ” (emphasis added) [22] He also found that: “… with the abandonment of prayer 3, any issues relating to whether or not the input data was intended to maximise revenue or whether they were not intended to conceal operational errors or whether they were in accordance with international standard practices, no longer arises for consideration or determination. If that is so, an evaluation or assessment of the FPM and any .csv data files would not only be unnecessary but would moreover be irrelevant .” (emphasis added) [23] The arbitrator dismissed the disclosure application.  The arbitration proceeded without the applicant’s legal representatives and its expert witness having access to the FMP and its .csv files.  The information was also not placed before the arbitrator. [24] As we shall see, this state of affairs (the exclusion of Schedule 30 comprising, the FPM and .csv files, from the arbitration) constitutes a key aspect of the applicant’s review application.  It is important to note, however, that the applicant did not seek to review the disclosure award, either in its immediate aftermath, or in its present review application.  I will return to this observation later in my judgment. [25] The arbitration proceeded on the main disputed issue.  In terms of the plea for a declarator, after the amendment referred to earlier, this now read: “ The input data for the operation strategy in the Facility Power Model shall comprise of the actual operation strategy followed at the Facility on a daily basis.” The question of whether in addition the respondent persisted with its plea for the rectification of clause 11.7 of Schedule 8, as discussed above, is contentious and forms the basis for one of the grounds for review.  I deal with this in detail later. [26] The parties exchanged witness statements and led the evidence of their witnesses before the arbitrator.  The applicant led the evidence of an expert witness, Mr Pine, who, as I indicated earlier, did not have access to the FPM and .csv files.  Much is made of this in the applicant’s grounds for review, which I address later. [27] The arbitrator’s award was handed down on 28 February 2022.  It was in the following terms: “ 72.1. paragraph 11.7 of Schedule 8 to the EPC contract is rectified to correct the cross-references as follows: ‘ 11.7.1 lf there is any deviation from the reference data referred to in paragraphs 11.6.1 to 11.6.6 (both inclusive) as stated in paragraph 11.6, all the results of the Facility Power Model must be corrected according to the principles described in paragraphs 11.7.3.1, 11.7.3.2 and 11.7.3.3; 11.7.2 Paragraphs 11.6.3, 11.6,4, 11.6.5 and 11.6.6 apply irrespective of whether the operator and the contractor are affiliated; 11.7.3 For the avoidance of doubt, regardless of who the operator is, reference data referred to in paragraphs 11.71 and 11.7 2, must in all cases be corrected for the long-term performance test according to the following principles: 11.7.3.1 for parameters differing from the reference data (paragraph 11.6) ...; 11.7.3.2 if during the long-term performance test there is any deviation of an operating parameter which has an input of the Facility Power Model other than the defined reference data (paragraph 11.6) …’ 72.2. the declarator sought by the claimant in prayer 2 as amended, is dismissed; 72.3. each party is to pay their own costs of the arbitration.” [28] The respondent was the claimant referred to in paragraph 72 of the award.  Its claim for substantive relief in the form of the declarator set out earlier was dismissed.  Despite having successfully opposed the respondent’s case for declaratory relief, the applicant seeks to review and set aside the arbitrator’s award, for reasons I discuss shortly.  To understand the motivation behind the applicant’s review, it is necessary to examine the body of the award, and the arbitrator’s reasoning more closely. [29] The arbitrator noted that the parties were agreed that the dispute, “at this stage, is one relating to the proper interpretation of the EPC contract.”  He set out the respective views of the parties: the respondent contended that paragraph 11.7 of Schedule 8 envisaged that the reference data reflected in paragraph 11.6 ought to be adjusted for actual values, while the applicant’s position was that the input data was fixed and the contract did not permit inputs comprising the actual operation strategy employed . [30] In paragraph 31 of the award, the arbitrator reasoned that: “ Thus what paragraph 11.7 contemplates (certainly in respect of sub paragraphs 11.6.4 and 11.6.5) is an adjustment to the parameters of the reference data when there is a deviation thereof by using the actual data. Although the parameters in respect of the reference data, "operation of the Facility", are not identified, there is no reason in principle, why the same approach cannot and should not be applied when there is a deviation from any of the parameters in respect of that reference data.” (Emphasis added) [31] However, paragraph 11.7 was “not as open ended as the contractor (the respondent) may contend”.  The arbitrator pointed to the second sentence of paragraph 11.7.3.1 and to paragraph 11.7.3.2.  Both incorporated qualifications to the permissibility of adjustments for the use of actual, as opposed to default, input data.  In the case of the former, if a parameter change cannot be directly converted into a change of an input, then the parties would have to agree on the appropriate correction of results to be applied.  As to the latter, any deviation of an input other than that defined in paragraph 11.6, can only be modified by agreement, if the change causes a decrease in energy output.  According to the arbitrator, these qualifications were imposed to protect the owner/applicant.  This was the more commercially sensible interpretation . [32] It is useful to set out the ‘Final Analysis’ section of the award in full, as it forms the heart of much of the applicant’s review: “ 61. The purpose of the Model was to enable the contractor in particular to simulate or project the required capacity that the Facility would have to meet for the purposes of the guarantees that the contractor had given to the owner. In order to do so, the Model of necessity, had to employ certain default input data that predicted for example the weather during a typical meteorological year as well as the operation strategy that would be employed by the operator during that period. 62. But the EPC contract recognised that the data utilised by the Model for the purposes of these projections or simulations, may actually be different during the test period and that consequently, it would be inappropriate to hold the contractor liable to certain guarantees based for example on a weather prediction or forecast that turned out to be inaccurate. For that reason, paragraph 11 of Schedule 8 contemplated the use of the actual data insofar as there had been a deviation from the parameters of the reference or default data. 63. The owner's witnesses accepted that the operation strategy that would be employed by the operator was dependent on the weather. Once that concession was made and once the owner acknowledged that the actual weather data is to be used as an input in the Model, it would follow that the actual operating strategy (save for that strategy which was not in accordance with the agreed upon strategy and which resulted in an improper operation or maintenance of the Facility), should be used as an input data. 64. That is precisely what sub paragraph 11.7.3.1 contemplates when it refers to the actual value of the parameters to be measured during the test period and to be used as input data for the Model. 65. However neither the default operation strategy file nor the actual operating strategy file utilised by the contractor, have been disclosed by the contractor. The owner points out that this impedes the proper interpretation of the EPC contract and in particular, paragraph 11 of Schedule 8 thereto. This arises because I had earlier made an award of the insistence of the contractor, that neither the Model nor the .csv files were relevant to the proper interpretation of the EPC contract and consequently need not have been disclosed by the contractor. In retrospect, it may have been useful if not preferable to have sight of the .csv files in order to better understand the parameters that were used as input data. This does not however make it impossible to interpret the paragraph, just more difficult. But counsel for the owner submitted that because of this difficulty, I should decline the invitation to make a declaratory order as prayed for by the contractor. 66. There is considerable force in that submission. 67. The award which the contractor seeks is a declaration that: "The input data for the operation strategy in the Facility Power Model shall comprise of the actual operation strategy followed at the Facility on a daily basis." (my emphasis) 68. But as I have pointed out above, the actual input data to be used in the Model is subject to the qualification reflected in sub paragraph 11.7,3.'I of Schedule 8. In the case of a parameter change which cannot be directly converted into a change of an input of the Model, the owner, the contractor and lender's representatives as well as the operator have to agree on the appropriate correction of the results to be applied. 69. I do not know whether the actual data sought to be used by the contractor can be directly converted into a change of an input in the Model. The declarator sought by the contractor does not take this into account and is consequently overly broad. It may include all actual data irrespective of whether such change may be directly converted as an input to the Model. 70. Thus, whilst I am of the view that on a proper interpretation of paragraph 11 of Schedule 8 to the EPC contract, the parties intended that the actual operating data may be used as an input in the Model when determining the LTPT this is only in circumstances where there has been a deviation from the reference data listed in paragraph 11.6(or that contained in the default .csv files) and moreover when the parameter change can be directly converted into a change of an input to the Model. 71. In light of the fact that the declarator sought by the contractor potentially includes those circumstances that may preclude the contractor from using the actual operating data as an input into the Model without obtaining the consent of various other parties including the owner, it would be inappropriate to grant the declarator in the form sought by the contractor and I accordingly exercise my discretion not to grant it.” (all emphases added) [33] Apart from the issue of whether the rectification award was properly made, the applicant’s main cause of complaint is not the refusal of the declarator sought by the respondent, but rather what the arbitrator finds in paragraph 70.  The applicant says that in subsequent communications, the respondent has asserted that based on this paragraph, the LTPT was in fact passed in June 2020.  Accordingly, the respondent is claiming a refund of substantial PLDs that were paid, it says in error.  The applicant accordingly asserts that paragraph 70 constituted a finding, which the respondent is treating as if it has operative effect.  It is this finding, which forms the main reason for the review, although I should make it clear that the applicant also challenges the rectification award. Grounds of review [34] It is trite, although worth repeating in the context of this case, that the question in a review under s 33(1) of the Act is not whether the arbitrator erred in his or her award.  As the Supreme Court of Appeal put it in Telcordia : [2] “ The fact that the arbitrator may have either misinterpreted the agreement, failed to apply South African law correctly, or had regard to inadmissible evidence does not mean that he misconceived the nature of the inquiry or his duties in connection therewith. It only means that he erred in the performance of his duties. An arbitrator ‘has the right to be wrong’ on the merits of the case… .” [35] Section 33(1) provides that: “ Where- (a) any member of an arbitration tribunal has misconducted himself in relation to his duties as arbitrator or umpire; or (b) an arbitration tribunal has committed any gross irregularity in the conduct of the arbitration proceedings or has exceeded its powers; or (c) an award has been improperly obtained, the court may, on the application of any party to the reference after due notice to the other party or parties, make an order setting the award aside.” [36] By agreeing to arbitration the parties limit interference by courts to the ground of procedural irregularities set out in this section.  They waive their right to any further ground of review, under common law or otherwise. [3] [37] The applicant relies on both facets of s 33(1) (b) as bases for its review, namely gross irregularity and excess of power (absence of jurisdiction).  In summary, the applicant’s grounds of review are the following: 37.1 As to excess of power, the applicant attacks both the rectification relief granted in paragraph 72.1 of the award, and the finding in paragraph 70. 37.2 It contends that in awarding a rectification of paragraph 11.7 of the contract, the arbitrator exceeded his jurisdiction because: (1) the parties had agreed that the respondent would abandon its claim for rectification; and (2) the rectification granted was not in line with the rectification as pleaded in the respondent’s statement of claim. 37.3 Regarding the finding in paragraph 70, the applicant contends that this finding was not a prayer that either party claimed.  Nor was the arbitrator asked to make his own or any alternative substantive “finding” other than the terms set out in the declarator.  The applicant asserts that having dismissed the declarator, the arbitrator had no power to grant any other declarator or make any “finding” of any similar force or effect.  In making his finding in paragraph 70, the arbitrator exceeded his power. 37.4 On the gross irregularity ground, the applicant again attacks paragraph 70 of the award on three interrelated bases. 37.5 In the first instance, the applicant contends that the arbitrator committed a fundamental misdirection and proceeded in a grossly unfair and irregular manner by making the paragraph 70 finding when he did not have the whole contract before him.  According to the applicant, this arose from the arbitrator’s “misdirected” disclosure ruling to the effect that Schedule 30, comprising the FPM, with its constituent .csv files, was irrelevant and thus excluded from the arbitration proceedings. 37.6 The applicant sought also to argue, in its written heads of argument, that this court should apply the test of gross irregularity laid down in the unfair labour context in Sidumo . [4] The submission in this regard was that in disregarding material facts and evidence, the arbitrator had reached a decision that no reasonable decision-maker could have reached and thus had committed a gross irregularity, rendering the finding in paragraph 70 reviewable. 37.7 The final contention under the gross irregularity also goes to paragraph 70.  The contention here was that the arbitrator breached the applicant’s right to a fair trial by, among other things, ruling that the FPM was excluded from the proceedings. [38] I should point out that the applicant did not press the Sidumo point at the hearing, a wise election, given the jurisprudence, including that of the Constitutional Court, holding that the Sidumo principles do not apply to the review of a private arbitration award under s 33(1) of the Act. [5] [39] In considering the applicant’s case for review, I start first with the gross irregularity attack against the finding in paragraph 70 of the award.  Thereafter, I deal with the excess of power ground of review in respect of that finding.  Finally, I turn to the gross irregularity and excess of power attacks against the rectification award in paragraph 72.1 . Gross irregularity: the finding in paragraph 70 [40] The applicant accepts, as it must, that gross irregularity goes to the arbitrator’s methodology, and not to the merits of his decision. [6] [41] It is the applicant’s case that the origins of the asserted gross irregularity lie in the arbitrator’s “misdirected” ruling that the FPM, contained in Schedule 30, was irrelevant to the issues in dispute.  As a consequence of this ruling, says the applicant, the arbitrator did not have the whole of the contract before him.  This is because Schedule 30 and the FPM, were an integral part of the contract.  Accordingly, so the argument goes, it was grossly unfair and irregular for the arbitrator to proceed to interpret the contract and reach the interpretive conclusion he reached in paragraph 70.  The applicant calls in aid the frequently cited dictum of Wallis JA in Endumeni [7] that the interpretation of a document requires: “reading the particular provision or provisions in the light of the document as a whole …” [8] .  It relies also on Capitec [9] and particularly on the underlined sentence below, in which the SCA held that: “ [Endumeni] and its progeny emphasise that the meaning of a contested term of a contract (or provision in a statute) is properly understood not simply by selecting standard definitions of particular words, often taken from dictionaries, but by understanding the words and sentences that comprise the contested term as they fit into the larger structure of the agreement, its context and purpose. Meaning is ultimately the most compelling and coherent account the interpreter can provide, making use of these sources of interpretation. It is not a partial selection of interpretational materials directed at a predetermined result .” [10] [42] The applicant submits that it was a grave misdirection for the arbitrator to make findings on the meaning of clause 11.7 without having before him the entire contract and, in particular, Schedule 30 and the FPM.  According to the applicant, the FPM was the focal point of the declarator sought by the respondent.  The arbitrator acted irregularly in making any finding on the meaning of the contract in the absence of this evidence.  It was also grossly unfair of the arbitrator to deprive the applicant’s lawyers and expert access to the same parts of the contract to which the respondent had access.  For these reasons, the applicant contends that the finding in paragraph 70 should be reviewed and set aside. [43] In support of its case that the FPM was a vital component of the contract and the dispute, the applicant referred to the definition of the Facility Power Model in the contract.  It is defined as meaning “the Facility power model attached hereto as Schedule 30”.  It is not disputed that the .csv files form part of the FPM.  The applicant also pointed out that the dispute in the settlement agreement was described as “involving whether the Contractor [respondent] is entitled to adjust the Facility Power Model (Schedule 30 to the EPC Contract) when determining the amount of the EEOU Performance Guarantee to take into account the actual operating strategy of the Facility as opposed to the default strategy.”  Further, that the Notice of Arbitration described the dispute in similar terms.  These are all indications, according to the applicant, of the FPM’s central role in the dispute and support its ground of review that the .csv files that formed part of the FPM ought not to have been excluded . [44] One of the difficulties for the applicant is that it never sought to review the disclosure award which excluded the FPM.  The applicant accepts that the review must proceed on the basis that that award cannot be upset on review at this stage.  This does not matter, says the applicant, because the irregularity is to be found not in the exclusion consequent on the disclosure award, but rather on the effect that the award had on the proceedings in the main arbitration.  The applicant says that having excluded the FPM and associated .csv files from the arbitration, the only appropriate course open to the arbitrator was that he was fundamentally precluded from making any findings on the proper interpretation of the contract.  This is what makes his finding in paragraph 70 reviewable, rather than a non-reviewable error of law. [45] Taking this line of submission further, counsel for the applicant referred to paragraphs 65 and 66 of the award, which are set out in full above.  Here, the arbitrator referred to the disclosure award and how, in retrospect, it may have been “useful if not preferable” to have had sight of the .csv files “in order better to understand the parameters”.  The arbitrator noted that this made it more difficult, albeit not impossible, to interpret the contract.  He referred to the applicant’s submission that “because of this difficulty, I should decline the invitation to make a declaratory order as prayed by the contractor”, and expressed the view that “[t]here is considerable force in that submission.”  The arbitrator then dismissed the respondent’s application for declaratory relief. [46] The applicant submitted in this regard that it is arguable that the reason for the dismissal of the relief was because, as counsel put it, the penny dropped for the arbitrator only in the main hearing that the absence of the FPM and .csv files evidence created an insurmountable obstacle for the declaratory relief sought by the respondent.  It was arguable, contended counsel for the applicant, that it was for this reason that the arbitrator declined to award the declaratory relief sought.  This being so, the arbitrator took a grossly irregular step by nonetheless reaching an interpretational finding in paragraph 70. [47] I find these submissions fundamentally problematic.  For one thing, once it is accepted, as the applicant does, that the exclusion award must stand, the applicant’s attack on the finding made in paragraph 70 begins to assume very much the nature of an appeal, based on an error of law on the part of the arbitrator by excluding the relevant evidence, rather than a legitimate gross irregularity review.  The real complaint seems to be that the arbitrator erred in excluding the evidence in the disclosure award, but because that award is unassailable, the complaint is refashioned as a review.  However, even if the arbitrator had second thoughts about his disclosure award (and I am not persuaded that he had any material second thoughts), this can only have led to a further error of law on his part in making the finding that he did in paragraph 70. [48] A related problem for the applicant on this score is that on a proper reading of the award, the dismissal of the declaratory relief was not because of any second thoughts on the part of the arbitrator or because the arbitrator was ultimately persuaded by the applicant’s argument.  It is patently clear from, in particular, paragraph 69, immediately preceding the impugned paragraph 70, and the paragraph following, that the reason the declaratory relief was refused was because it was “overly broad”.  The arbitrator could not grant the relief because he did not know “whether the actual data sought to be used by the contractor can be directly converted into a change of input” and thus, whether the qualification in paragraph 11.7.3.1, discussed earlier, applied.  It is for this reason that he concluded, in paragraph 70, that the actual operating data “may” be used, this was only in certain circumstances and, in para 71, because the declarator sought potentially included circumstances where the rider may apply, it would be inappropriate to grant it. [49] Whether the arbitrator was right or wrong in this interpretation is irrelevant.  The point is that the applicant’s submissions in this regard do not support a case for review. [50] What is more, there is a fundamental flaw in the premise on which the gross irregularity ground of review stands.  It proceeds on the assumption that the FPM and its .csv files were an integral part of the contract and central to the interpretational dispute that was referred to arbitration.  It is for this reason that the applicant contends that a gross irregularity was committed by the arbitrator’s failure to consider the whole contract when he embarked on the interpretive exercise and made his finding in paragraph 70.  Once again, it is difficult to understand the complaint as falling properly within the ambit of a gross irregularity rather than a non-reviewable error of law.  However, be that as it may, and assuming that it is a valid review attack, it is misdirected. [51] It is so that the FPM is defined as Schedule 30 in the contract.  It is also so that the FPM is referred to in the dispute as formulated.  However, this does not mean that the dispute was about the inner workings of the FPM, and in particular the .csv files, algorithms and other formulae of which it is composed.  The respondent’s stance throughout the arbitration was that the dispute referred to the arbitrator did not concern an evaluation of the workings and algorithms of the FPM, but was an in-principle dispute concerning the proper interpretation of Schedules 8 and 9, and not Schedule 30 of the contract. The inner workings of the FPM were not in dispute.  The arbitrator agreed in his disclosure award.  He found: “ [17] …In my view, the now limited prayer 2 [the declaratory relief] must be read in context and against the articulated dispute pleaded by the contractor in its statement of claim. Doing so reveals that it does no more than seek to achieve a principled view on whether or not the operation strategy in the FPM should comprise the actual operation strategy as it alleges it should, or whether it should include the default operation strategy. It does not seek to achieve an award on the correctness or otherwise of the data to be used in the FPM .” (emphasis added) And: “ [27] Having regard to the pleaded issues, the most important of which I have referred to above, it is quite clear that what has been referred to arbitration is an interpretational dispute. Whilst the outcome of the interpretational dispute of the EPC contract may have subsequent consequences on other disputes between the parties, including whether or not any input data to the FPM was correct or not, or whether there is any liability for liquidated damages, those disputes have not been referred to arbitration . Neither the FPM nor the .csv data files are relevant to the interpretation of the EPC contract and in particular, whether the input data to the FPM ought to refer to the actual operation strategy or the default strategy.” (emphasis added) [52] Whether or not the arbitrator’s conclusions were correct is not relevant to this review. The interpretational dispute was ruled by the arbitrator to be a narrow one. The arbitration proceeded on the basis that the dispute was not about the inner workings of the FPM, and the FPM and .csv files were irrelevant to the arbitration.  The parties are bound by those conclusions.  The dicta relied on by the applicant in support of its contentions do not require that irrelevant contractual provisions must be taken into account.  This would be absurd.  A decision-maker may err in her interpretation of a document by not appreciating the relevance and significance of certain provisions, but that is a matter for appeal.  In circumstances where, as here, an arbitrator has made an unassailable finding on the ambit of the dispute and the irrelevance of certain evidence, it cannot be said that he committed a gross irregularity by interpreting the contract in the absence of that evidence. [53] As to the issues of fairness and the alleged breach of the applicant’s fair trial rights, these too must be seen in the context of the nature and ambit of the dispute.  It is trite that procedural fairness is a contextual measure.  What is unfair in one context may be fair in another.  In this case, the relevant context is provided by the arbitrator’s binding ruling that the dispute was a narrow one, and that the FPM and .csv files were irrelevant to the dispute.  Seen in this context, what the applicant contends for is a finding that the arbitrator acted unfairly in excluding irrelevant evidence.  This simply cannot be a valid basis for review.  The fact that the applicant’s expert witness bemoaned his inability to view the FPM and .csv files, and that the applicant’s counsel continued to assert his client’s stance on the issue throughout the arbitration proceedings cannot create a case for unfairness where no ground exists. [54] At the end of the day, the arbitrator ruled on the irrelevance of the FPM, he approached the dispute as a narrow one, and in line with this approach, his interpretation of the contract was one of principle.  He made no finding on the correctness or not of the FPM and .csv files.  It is not surprising, then, that he expressed, in paragraph 60 of his award, that: “[s]ave for one aspect which I deal with under the final analysis section of this award, I did not find the evidence of any of these witnesses particularly helpful on the interpretation of paragraph 11 of Schedule 8 to the EPC contract.”  The arbitrator’s expressed view underscores the point that given the narrow ambit of the dispute, the exclusion of the FPM and .csv files from the arbitration did not give rise to reviewable unfairness. [55] I conclude, for all of these reasons, that there is no merit in the applicant’s contention that paragraph 70 of the award should be reviewed and set aside on the grounds of gross irregularity and unfairness. Excess of power: the finding in paragraph 70 [56] As I noted earlier, and as is clear from my discussion of the issues thus far, the applicant’s real complaint is not the dismissal of the claim for declaratory relief, but the finding of the arbitrator in paragraph 70.  The applicant does not accept this finding because it has implications for the respondent’s liability for PLDs which are adverse to the applicant’s interests.  If the dispute had been determined by a court, one would imagine that the applicant would have appealed this finding.  This avenue not being open to the applicant in the arbitration context, it has attempted to fit its complaint into the category of a review, based on an excess of power ground. [57] Given the obvious difficulty presented by needing to avoid the complaint being seen as an appeal against the merits of the finding in paragraph 70, the case for the applicant is nuanced.  The applicant accepts that the arbitrator had the power to interpret paragraph 11 of the contract.  However, according to the applicant, the arbitrator’s power extended only so far as he could either grant the declarator in the precise terms stated by the respondent in its amended statement of claim, or dismiss it.  What the arbitrator did not have, says the applicant, was the power to make any other operative award or conclusive finding on the meaning of paragraph 11.  This is what the arbitrator did in paragraph 70 and his findings there fell outside of his jurisdiction. [58] In its founding affidavit the applicant states its case thus: “ The Owner [applicant] accepts that the arbitrator was empowered to consider and interpret the contract. That was indeed his primary duty, according to the pleaded dispute. But the dispute over the proper interpretation of the contract that was pleaded and referred to him for final determination was confined to the Contractor's [respondent’s] claim for a declarator. The Owner [applicant] therefore does not question the arbitrator's power to examine and interpret the contract, provided it was ancillary to and directed at determining the central issue on interpretation that was before him, namely to determine whether or not the declarator sought accorded with a proper interpretation of the contract.” [59] The applicant goes on to assert that in paragraph 70 the arbitrator purported to make a definitive finding on the “proper interpretation” of the contract although this was in terms “other than those claimed in the declarator”.  It says that in so doing the arbitrator “most regrettably and unnecessarily” strayed beyond the narrow issue and beyond the power conferred on him. [60] I must confess to finding the applicant’s case peculiar.  It accepts that the arbitrator’s primary duty was to interpret the contract.  It accepts that the arbitrator had the power to interpret the contract in a manner “ancillary to and directed at determining the central issue” before him.  Despite this, it seeks to straight-jacket the arbitrator’s interpretive powers to extending no further than a “yay” or a “nay” to the interpretation preferred by the respondent in its declaratory relief. [61] How was the arbitrator to reach the point of a “yay” or a “nay” on that declaratory relief without reasoning his way towards that end?  How was he to do so without embarking on a reasoned interpretive exercise and recording his findings along that path? How can it possibly be said that those findings were not ancillary to determining the central issue before him?  Had paragraph 70 not been there (and indeed its accompanying paragraphs, particularly 67 to 69 and 71) the parties would have been left wondering on what basis the declarator was dismissed.  We know, from my earlier analysis of the award, that these paragraphs went to the heart of the dismissal.  They explain that the declarator was dismissed because it was too broad, and why it was too broad.  It seems to me to be patently clear that these paragraphs and the findings in them were quintessentially ancillary to the arbitrator’s accepted primary power, namely to interpret the contract and to determine whether, on the basis of that interpretation, the declarator should be granted or dismissed. [62] For these reasons I find that there is no merit in this ground of review. The rectification issue [63] The applicant challenges the rectification awarded in paragraph 72.1 on the ground of excess of power.  There are two legs to this review ground.  In the first instance, the applicant says that the parties had agreed that the rectification claimed by the respondent was to be abandoned.  In the second instance, the applicant says that the rectification in its terms did not accord with those of the rectification claimed. [64] Regarding the first aspect of this ground of review, the respondent disputed that the parties had agreed that it would no longer seek the rectification it claimed.  The applicant relied on an email it had sent to the arbitrator regarding the issue of rectification.  Its case is that the email demonstrates the agreement between the parties that the rectification would be abandoned and thus establishes that the arbitrator no longer had any power to rectify the contract. [65] The email reads: “ Dear Sir We write in connection with the plea for rectification. As you will have noted from the Defendant's special plea on jurisdiction, although the Defendant acknowledges that there is an error with the cross-referencing, the Defendant does not have the power to consent to the rectification in terms of the project documents and furthermore, the Defendant disputed your jurisdiction to deal with the issue of rectification. We do, however, have instructions from our client that it is prepared to proceed on the basis that you may interpret schedule 8 on the basis of the corrected cross-referencing indicated in paragraph 18 of the Statement of Claim. This has been communicated to the Claimant and it is agreed that there will be no order for costs against the owner as a result of this approach. The Claimant has indicated that this does not exclude such costs from an overall costs order that may be issued by you. The point is that there will be no special order for costs against the Defendant based on the fact that the Defendant is not objecting to your dealing with the interpretation of schedule 8 on the basis of the corrected cross-referencing. Thank you. Regards”. [66] The defendant referred to in the email is, of course the applicant.  The respondent contends that the email does no more than reflect an agreement that there would be no order for costs against the applicant.  It says that there was no agreement that the respondent would abandon its plea for rectification, nor does the letter reflect such agreement.  There is merit in the respondent’s contention.  The first substantive paragraph in the email records the applicant’s stance, namely, that it is prepared to proceed on the basis the basis of the corrected cross-references contained in the plea for rectification.  It does not record any agreement that the plea for rectification would be abandoned.  In any event, insofar as there is any dispute on this point, the respondent’s version must prevail. [67] There is thus no merit in this aspect of the review. [68] The remaining aspect of this ground of review is that the rectification awarded in paragraph 72.1 does not accord with the pleading for rectification in the statement of claim.  The complaint relates not to the entire paragraph 72.1, but rather to one portion of it.  More specifically, the rectification awarded in relation to paragraph 11.7.3.  In its original form, as noted earlier, this paragraph read (in the original): “ For the avoidance of doubt, regardless of who the operator is, reference data referred to in paragraphs 11.6.1 and 11.6.2 must in all cases be corrected for the Long Term Performance Test according to the following principles … .” (emphasis added) [69] Elsewhere in the original, paragraph 11.7 included cross-references to paragraphs 11.5, and 11.6.  The parties agreed that these were erroneous and that these references were to be to read as cross-references to 11.6 and 11.7 respectively.  In other words, it appears that an additional subparagraph to paragraph 11 was added at some point in the drafting of the contract, such that what had originally been 11.5 became 11.6.  However, the cross-references clearly were not amended to follow suit in paragraph 11.7.  Hence the erroneous cross-references according to the original paragraph numbering.  This is simply a matter of common-sense if one reads the paragraphs carefully. [70] The applicant’s complaint is that as regards 11.7.3, in its plea for rectification in the statement of claim the respondent had left the reference to 11. 6 .1 and 11. 6 .2 unchanged, but that the arbitrator had nonetheless rectified 11.7.3 such that the references were now to 11. 7 .1 and 11. 7 .2.  This, says the applicant was beyond his power as he was only empowered to rectify specifically on the terms pleaded by the respondent. [underlined for emphasis] [71] If one reads paragraph 11.7 in its entirety, it is patent that the arbitrator’s rectification accurately reflected what the parties had intended.  Paragraph 11.7.1 (post-rectification) refers to deviations from the reference data referred to in paragraphs 11.6.1 to 11.6.6.  What is more, it says that the results must be corrected according to the principles described in paragraphs 11.7.3.1 to 11.7.3.2.  Paragraph 11.7.3 then says (as rectified by the arbitrator) that for the avoidance of doubt, the reference data referred to in 11.7.1 and 11.7.2 must be corrected according to the following principles. [72] There is a clear link between this paragraph and paragraph 11.7.1.  The only way in which these paragraphs make any sense is if the sub-paragraphs are read so that the “principles” referred to in 11.7.3 apply to all the reference data identified in paragraph 11.7.1.  To leave 11.7.3 in its original form (as the applicants insist the arbitrator should have done), would lead to an irrational result.  It would not lead to the expressed “avoidance of doubt” as to which reference data are subject to the 11.7.3 principles.  On the contrary, it would create doubt that could never have been intended by the parties.  What the parties obviously intended, and what the arbitrator recognised, was that all the data references were to be subject to the principles in paragraph 11.7.3, and not just those referred to in paragraphs 11.6.1 and 11.6.2, as recorded.  Without the rectification there would be a clash between paragraph 11.7.1 and 11.7.3.  The rectification pleaded simply overlooked that an additional change had to be made. [73] The arbitrator’s rectification was effected to correct a patent error in the contract.  His powers as arbitrator were wide enough to do so.  Under paragraph 11.2.11 of the applicable arbitration rules, he was afforded the power to order rectification of any contract.  This is ancillary to his overall power to exercise the “widest discretion and powers allowed by law to ensure the just, expeditious, economical and final determination of all the disputes raised in the proceedings”.  Rectification was expressly raised as an issue.  All that the arbitrator did was to exercise his overall power to order further rectification to correct the patent error and to ensure that the contract accorded with the obvious intention of the parties. [74] For these reasons, I find that there is no merit in this ground of review either. Conclusion and order [75] The applicant has been unsuccessful on all of the grounds of review advanced.  The application must be dismissed. [76] I make the following order: “ The application is dismissed with costs, including those of two counsel, one of whom is senior counsel.” R M Keightley JUDGE OF THE HIGH COURT JOHANNESBURG Delivered:  This judgment was prepared and authored by the Judge whose name is reflected and is handed down electronically by circulation to the Parties/their legal representatives by email and by uploading it to the electronic file of this matter on Court Online/Case Lines.  The date for hand-down is deemed to be 05 July 2023 . Date of Hearing:   18 April 2023 Date of Judgment: 05 July 2023 APPEARANCES For the Applicant: Adv. M Kriegler SC and Adv. M Schafer Instructed by: Norton Rose Fulbright South Africa For the First Respondent: Adv. W La Grange SC and Adv. A Russell Instructed by: Pinsent Masons South Africa For the Second Respondent: Notice to Abide Instructed by: Dockrat Attorneys Inc. [1] 42 of 1965. [2] Telcordia Technologies Inc v Telkom SA Ltd [2006] ZASCA 112 ; 2007 (3) SA 266 (SCA) at para 85. [3] Id at para 51. [4] Sidumo & Another v Rustenburg Platinum Mines Ltd and Others [2007] ZACC 22; [2007] 12 BLLR 1097 (CC); 2008 (2) SA 24 (CC). [5] See Lufuno Mphaphuli & Associates (Pty) Ltd v Andrews [2009] ZACC 6 ; 2009 (6) BCLR 527 (CC); 2009 (4) SA 529 (CC) par 232 to 234 ; National Union of Mineworkers obo 35 Employees v Grogan NO & Another [2010] ZALAC 3 ; (2010) 31 ILJ 1618 (LAC) at para 33. [6] Ellis v Morgan; Ellis v Dessai 1909 TS 576 at 581. [7] Natal Joint Municipal Pension Fund v Endumeni Municipality [2012] ZASCA 13; 2012 (4) SA 593 (SCA). [8] Id at para 18. [9] Capitec Bank Holdings Ltd and Another v Coral Lagoon Investments 194 (Pty) Ltd and Others [2021] ZASCA 99; 2022 (1) SA 100 (SCA). [10] Id at para 50. sino noindex make_database footer start