African Distribution Company S.a.R.L v Aastar Trading Pte Ltd

This is a claim by African Distribution Company S.a.R.L. (‘ADC’), a company from Cote d’Ivoire, to set aside an arbitration award by an English Arbitrator, Mr Noyce (‘the Arbitrator’) dated 21^st February 2024 (‘the Award’) in favour of a Singaporean company, AASTAR Trading (‘AASTAR’). ADC took no part in that arbitration, as it says it was never served with notice of the proceedings, nor knew anything about them, until it received the Award on 8^th July 2024 in Ivorean enforcement proceedings brought by AASTAR. However, ADC contends that it was not formally served with the Award in those proceedings on 30^th July 2024, which prompted ADC to issue this claim on 27^th August 2024.

As a result, ADC is outside the 28-day time limit under s.70(3) Arbitration Act 1996 (‘AA’) to bring its challenges to the Award (which expired on 21^st March 2024) for lack of substantive jurisdiction under s.67 AA and serious procedural irregularity under s.68 AA. For both grounds ADC alleges non-service of the arbitral proceedings by AASTAR and so that the Arbitrator was not properly constituted. ADC also alleges under s.68 breach of arbitration rules and no reasonable opportunity to participate. ADC seeks an extension of time for these two challenges under s.80(5) AA and CPR 62.9. This not a matter for ‘relief from sanctions’ under CPR 3.9 (but that is an issue for the admission of the late statement from ADC’s Managing Director, M Niangdou). AASTAR contests both applications contending it served ADC with notice of arbitration by email to ADC’s generic email addresses (‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’ the ‘generic emails’) ADC had used to communicate with AASTAR previously. AASTAR also says ADC’s challenge on 27^th August 2024 was more than 28 days after it admits receiving the Award on 8^th July, so the extension should be refused. ADC was represented by Mr Wassouf and AASTAR by Mr Lamming.

I pay tribute in particular to Counsels’ skill in addressing an issue of wider relevance which only came up in argument. That is whether, as a matter of law, ADC’s third ground of challenge to the Award under s.72(1) AA can be pursued even though issued after an award and outside the 28-day time limit. This is a subject on which there is a divergence of judicial opinion, all of which is obiter:

In Bernuth, this ‘s.72(1) point’ only emerged on the morning of the hearing and it also came up rather obliquely during this hearing. Just before skeletons were due, ADC conceded it needed an extension of time for all its challenges. At the hearing, Mr Wassouf did not advance interpretation (b) until I queried (a); and even then, he did so cautiously, as the potential availability of a remedy under s.72(1) AA without time-limit might weaken ADC’s argument for an extension of time. Due to that and to pressure of time, I told Counsel I would only decide whether ADC’s s.72(1) claim was arguable. But whilst I am only deciding s.72(1)’s arguability in this case, I can (especially in a reserved judgment) reach a concluded view on the effect of s.72(1) as a matter of statutory interpretation. Mr Lamming did in fact research and argue the point and relies on Merkin and Flannery where ‘interpretation (a)’ is fully articulated. If it is right, it might strengthen ADC’s extension application, so I should grasp the nettle and say so. But if I prefer ‘interpretation (b)’, then it may be useful to have a reasoned judicial response to Merkin and Flannery’s argument (which has not yet been considered judicially so far as I know), especially given the new changes to the Arbitration Act 1996 in the Arbitration Act 2025. However, since I would then be going no further than saying that s.72(1) is arguable in this case, my conclusion on it would not bind the parties: who can fully-argue the law, whether s.72(1) is open to ADC given its concession; and if so, whether it can obtain any relief under s.72(1), before another judge. But I could take into account the arguability of the s.72(1) point on the extension of time now.

This s.72(1) point is relevant to ADC’s application to extend time, as whether or not it arguably has a s.72(1) claim if time is not extended is relevant to a key factor on extensions of time under s.80(5) AA: ‘whether in the broadest sense it would be unfair to the applicant to be denied the opportunity of having the application determined’. That is one factor Colman J in AOOT Kalmneft v Glencore [2001] 2 All ER (Comm) [59]:

“Accordingly, although each case turns on its own facts, the following considerations are…likely to be material: (i) the length of the delay; (ii)  whether, in permitting the time limit to expire and the subsequent delay to occur, the party was acting reasonably in all the circumstances; (iii)  whether the respondent to the application or the arbitrator caused or contributed to the delay; (iv)  whether the respondent to the application would by reason of the delay suffer irremediable prejudice in addition to the mere loss of time if the application were permitted to proceed; (v)  whether the arbitration has continued during the period of delay and, if so, what impact on the progress of the arbitration or the costs incurred in respect of the determination of the application by the court might now have; (vi) the strength of the application; and (vii)  whether in the broadest sense it would be unfair to the applicant for him to be denied the opportunity of having the application determined.”

I will consider these ‘Kalmneft Factors’ in my conclusions on the extension at the end, after addressing two pure legal issues. Firstly, the ‘s.72(1) point’ of statutory interpretation relevant to factor (vii). Secondly, whether sending a notice of arbitration to a generic email address can be good service of arbitral proceedings, discussed in Bernuth and other cases and relevant to factor (vi). But I will start with the factual background.

The evidence before me on the application consists of a documentary bundle running to 600 pages (with some recent additional documents). This includes detailed statements from ADC’s solicitor Mr Abu-Manneh dated 27^th August 2024 (the date of issue of the challenge) and AASTAR’s solicitor Mr Ferrigno dated 1^st April 2025 (shortly after AASTAR was served with the challenge on 14^th March 2025, in accordance with a Court-approved extension).

However, on 7^th July 2025, less than two weeks before the hearing, ADC applied to rely on an updating statement from M Niangadou, which added significant evidence to that provided by Mr Abu-Manneh’s statement almost a year earlier, including: (i) the use of the generic email addresses to which AASTAR sent the notice of arbitration; (ii) M Naingadou’s evidence that he had checked those generic email addresses and could find the emails relating to the arbitration; and (iii) that he was given legal advice when he first saw the Award on 8^th July 2024 that he could ignore it in the Ivorian enforcement proceedings. This last point raises questions of waiver of privilege. Mr Abu-Manneh also said that he was advised by Ivorian Counsel ADC was entitled to ignore the Award in Ivorian law until formal enforcement proceedings started. As Mr Lamming says, as evidence by a lawyer of what another lawyer said, this may be evidence of Ivorian Law which requires permission. But Mr Wassouf confirmed that ADC only relied upon evidence of the fact an Ivorian lawyer advised M Naingadou to ignore the Award on or around 8^th July 2024, not as evidence of Ivorian Law.

In Mr Lamming’s skeleton argument, he set out a detailed objection to the admission in principle of M Niangadou’s statement (which as a late statement without permission would require relief from sanction under CPR 3.9); but also contended it was of limited force as raising more questions than answers. At the start of the hearing, I indicated I would admit it de bene esse as Mr Wassouf suggested. Indeed, having heard argument, I consider there is significant force in Mr Lamming’s criticisms of M Niangadou’s statement:

The following factual background is derived from Mr Abu-Manneh’s statement, supplemented by the statements of M Niangadou and Mr Ferrigno, the Award and various contemporary documents (especially emails) which Mr Wassouf and Mr Lamming took me through in submissions. As Mr Wassouf observed, in the absence of oral evidence, the evidential exercise on Kalmneft Factor (vi) – the strength of the challenge – resembles that in other types of interlocutory application concerning merits of a claim, such as summary judgment. So, whilst the Court may be able to resolve points of interpretation of a contract or statute, it should be cautious of resolving conflicts of evidence (except perhaps evidence supported or contradicted by undisputed contemporary documents). Indeed, Cooke J observed in Thyssen Canada v Mariana [2005] EWHC 219 (Comm) at [56] (cited in Brown & Sons v Crosby Homes [2008] EWHC 817 (TCC) at [31]):

“...[T]here are extensive conflicts of evidence between the parties ......I am not in a position to evaluate the evidence as it appears on the statements in a way which enables me to take this into account as a factor one way or the other in the context of an extension of time."

ADC is an Ivorian company specialising in the import-export and distribution of fresh or industrial natural food products. AASTAR is a Singaporean trading company specialising in supply of various products including food products. From May 2021 to March 2022, they entered into numerous contracts (16 were considered in the Award) for the supply by AASTAR to ADC of Indian rice (the one contract for Vietnamese rice was not considered in the Award) with a clause for arbitration by the Grain and Feed Trade Association (‘GAFTA’).

The various contracts were all on AASTAR’s standard form and contained two sections. The first section was headed ‘Main Terms’ and set out the Parties’ contact details, the amount of rice being sold and bought; the price, the payment dates etc. The ‘Main Terms’ differed from contract to contract. But ADC’s contact details in each of contract was given as an address in Ivory Coast: “BP 3354 Abidjan, Ivory Coast”; and a telephone number: “Tel: +225 21241245”. In each case, AASTAR’s standard contract required ADC to sign, stamp and return a scanned copy of the contract by email to AASTAR’s email address ‘aastaragriops@kpn-corp.com’, with the original copy to be couriered to the AASTAR’s physical address in Singapore. It added that if AASTAR did not receive ADC’s acknowledgement or any objection to the contract term within 2 business days of receipt, the contract would be deemed binding and conclusive. It also provided for a mechanism for contract variations and amendments in writing (which was occasionally exercised). The second section of each contract was headed ‘General Terms and Conditions’ and was substantially identical across all the contracts. Clause 3 was headed ‘Arbitration’ and provided that “if any dispute in connection with this contract occurs, both parties agree to submit to GAFTA Arbitration in Singapore in accordance with Arbitration Rule 125 of GAFTA’. Notwithstanding this, as I shall explain, the GAFTA rules provide for arbitration in London, which is where the Arbitrator was appointed and issued the Award. Clause 5 ‘Special Terms’ included that “All disputes related to this Contract shall be settled in Singapore Court of Law. All terms and conditions when not in contradiction with the above will be as per GAFTA 122”. (I believe ‘GAFTA 122’ was not an older version of the GAFTA Rules, but its standard rice contract No.122).

The first use of ADC’s generic emails (‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’) which Counsel took me to was in January 2022, when as Mr Wassouf pointed out, most of the contracts had already been negotiated and implemented. Whilst AASTAR must have got those email addresses from somewhere, this was an email to ADC – specifically M Niangadou - from AASTAR, attaching three sales contracts. I have not found ADC’s emails returning those – so as Mr Wassouf says, it is possible this was from a different email address. On 15^th March 2022, AASTAR emailed to both ADC’s generic emails the letter of indemnity for a shipment comprising various contracts said to have arrived in Abidjan in Cote D’Ivoire on 8^th March 2022. AASTAR chased this to the ADC generic emails on 16^th March and the same day, someone at ADC responded from ‘adc_ci@yahoo.fr’ attaching the ‘signed and hidden LOI’ (seemingly ‘hidden’ a mistranslation of ‘cachee’, rather than ‘cachet’ meaning ‘stamped’). ADC used the same generic email ‘adc_ci@yahoo.fr’ on 27^th June 2022. M Niangadou returned to AASTAR a scanned version of an amendment dated 21^st June 2022 to the parties’ last contract AAS22-044 originally dated 8^th March 2022 and in similar terms to the others I described, changing the contract price from $2,548,000 to €2,415,700. He did this by responding on the ‘adc_ci@yahoo.fr’ address to an earlier chain of emails from AASTAR to both ADC’s generic emails about one of the January 2022 contracts. Therefore, I accept Mr Wassouf’s point that there is no evidence of use by either party of ADC’s generic emails for the contracts in 2021 and only use from time to time in 2022, when the evidence suggests regular use of the telephone. But on the other hand, as Mr Lamming said, email use in 2022 included contract amendments and letters of indemnity, where ADC both received emails to and sent them from its ‘adc_ci@yahoo.fr’ generic email.

However, ADC also used its ‘adc@adc.ci’ generic email: on 15^th December 2022, an unnamed person used the ADC generic email ‘adc@adc.ci’, marked ‘urgent’, to email AASTAR and others. They reported serious problems with rice transported, for which ADC’s customers were refusing to pay because the quality of the rice had been affected by the time the ship took to arrive. ADC rejected AASTAR’s representative’s proposed reduction and demanded a solution before he left Abidjan. Further emails were sent on ‘adc@adc.ci’ from someone at ADC on 30^th December 2022 and 5^th January 2023. Indeed, on 5^th May 2023, there was a chasing email from AASTAR to both ADC’s generic emails about outstanding payments; and a reply from someone at ADC using ‘adc@adc.ci’ querying the bank details AASTAR used. They did not suggest a different ADC email address should be used – indeed ADC never seems to have told AASTAR not to use either of the generic emails.

Whilst it is not pleaded in ADC’s challenge to the Award, Mr Wassouf had instructions that by the end of 2022, a dispute had broken out between the parties, with AASTAR demanding payment and ADC insisting on discounts because of various issues with some of the shipments. Certainly, those emails suggest some such disputes. Mr Abu-Manneh and M Niangadou also make clear that the latter never admitted liability to AASTAR for the outstanding payments, which across the sixteen disputed contracts amounted to €1,895,099.61 and $299,000 (on one contract in dollars). As I understand it, it is ADC’s case that those figures are the discounts from the original contract.

Less than a month after the last use of the ‘adc@adc.ci’ generic email in May 2023, on 1^st June 2023, AASTAR’s shipping law firm in Singapore ‘Dennis Matthew’ emailed a letter before claim to both of ADC’s generic emails ‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’. This claimed outstanding payment on sixteen of the contracts of €1,895,099.61 and $299,000 and warned that unless these sums were paid by 15^th June 2023, AASTAR would start proceedings. ADC did not respond to this – whether to suggest a different email or at all.

The 20^th July 2023 is perhaps one of the most important dates in this case, because it was the day AASTAR initiated the arbitration and says that it served the Notice of Arbitration on ADC by emailing it to its two generic emails ‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’. Indeed, the Notice of Arbitration sent by ‘Dennis Matthew’ but drafted by another Singaporean lawyer Prem Gurbani was marked ‘by email only’, even though it included ADC’s correct address from the contracts. The Notice relied on the same sixteen contracts between December 2021 and May 2022 and claimed the outstanding prices of €1,895,099.61 and $299,000. The Notice referred to the same arbitration clause in each of the contracts but proposed an arbitrator not in Singapore but in the UK, Mr Davies. The Notice asked ADC to confirm within 9 days whether it agreed Mr Davies as arbitrator, or to nominate an alternative. It is not disputed that on 20^th July 2023, that Notice of Arbitration was emailed to GAFTA by ‘Dennis Matthew’ to initiate arbitration, notifying GAFTA of its proposal of Mr Davies as arbitrator and requesting arbitration for all sixteen contracts be consolidated. The email had the subject line ‘Commencement of GAFTA Arbitration proceedings by Aastar Trading Pte Ltd against African Distribution Company’ and was ‘cc-ed’ to Prem Gurbani and ADC’s generic emails ‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’. This would have been a different subject line – and from a different email address (AASTAR’s lawyers not AASTAR itself) – from previous emails by AASTAR to ADC.

Shortly before this hearing, another email was provided to the Court by AASTAR’s English lawyers MFB. This was again dated 20^th July 2023 and again from ‘Dennis Matthew’ cc-ing in Prem Gurbani with the same subject line: notifying the arbitration; attaching the Notice of Arbitration; and suggesting Mr Davies as sole arbitrator, unless ADC objected or proposed an alternative within 9 days i.e. by 29^th July 2023. This email on its face states it was sent directly to ADC’s two generic emails ‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’ – it was ‘flagged’ by ‘Dennis Matthew’, but there is no suggestion they requested a read receipt. I admitted this email for AASTAR (as I did de bene esse M Niangadou’s statement for ADC). It is important because it has been ADC’s case that the Arbitrator appointed by GAFTA Mr Noyce said that all correspondence with ADC had been sent through GAFTA that was invalid service under its rules. But that ‘cherry-picks’ what he said:

“2.2.

All communications from the Tribunal to the parties (and from Claimants to Respondents during the proceedings were through GAFTA, as secretariat of the procedure under GAFTA Arbitration Rules No.125 and thus were correctly served upon both parties.”

However, in the very next paragraph, the Arbitrator added that:

“2.3.

On 20 July 2023, [the] Claimants served Notice of Arbitration on [the] Respondents.”

I discuss below the key factual issue whether this email was received by ADC at its generic emails; and if so, whether it was opened and read; and likewise, all the arbitral emails sent by GAFTA to ADC’s generic emails. I also return to whether ‘Dennis Matthew’ sending the Notice of Arbitration to ADC’s generic emails was in law valid service. But the point is that it is AASTAR’s case – which the Arbitrator accepted – that AASTAR’s lawyers emailed ADC the Notice of Arbitration and from then on emails were through GAFTA.

In any event, ADC did not respond to the Notice of Arbitration or object to AASTAR’s proposed arbitrator Mr Davies (e.g. because he was based in the UK, not Singapore as the contracts provided in the arbitration clause, even though clause 5 gave the Singapore Courts exclusive jurisdiction). On that point, the Arbitrator stated in the Award at paras.1.4 and 7.2-7.3 that whilst the GAFTA Rules changed in January 2022 (between some of the disputed 16 contracts and others) under each version, the seat of an arbitration by GAFTA was London, which took precedence over the reference to Singapore in the arbitration clause. Right or wrong, that decision is not subject to challenge.

According to Mr Abu-Manneh’s statement and the Award, on 30th August 2023, the Defendant filed its‘Claim Submissions’ seeking €1,895,099.61 and $299,000 under the 16 contested contracts. On 6 September 2023, GAFTA appointed Mr P Noyce as sole arbitrator and directed ADC to file submissions within 28 days’ of service of AASTAR’s submissions. Whilst there was no more direct communication between AASTAR (or their lawyers) and ADC, this and all subsequent correspondence from GAFTA was to ADC’s generic email accounts (i.e. ‘adc@adc.ci’ and ‘adc_ci@yahoo.fr’) which GAFTA had been given by AASTAR. However, as M Niangadou says, since these GAFTA emails too came from an unfamiliar email account, even assuming they were ‘received’, once again it is entirely plausible that they (and GAFTA’s future emails) were automatically filtered as ‘spam’ in ADC’s generic emails.

On 4th October 2023, the 28 day limit expired and on the 9^th October, the Tribunal gave the Claimant a further 7 days to file any submissions, with a warning from the Arbitrator if they failed to do so, his intention was to give a peremptory order to respond within a short time frame and if they did not, to proceed to determine the Award based on the evidence before him. ADC still did not respond, so the Arbitrator in a peremptory order gave them one final week to do so on 17^th October, to which again ADC did not respond. On 26^th October 2023, the Arbitrator considered ADC had enjoyed a reasonable opportunity to respond and that he would proceed to determine the award unless he had further questions. But on 6^th December, the Arbitrator (through GAFTA emails) did ask for further representations within 14 days, including on the Singapore exclusive jurisdiction clause. On 20^th December, AASTAR’s lawyers provided their further representations, but once again, ADC did not respond. So, the Arbitrator gave ADC a further 21 days as it was Christmas; and in the absence of a response, on 16^th January 2024, he made another peremptory order giving ADC a final 3days to respond, which ADC did not do. So on 23^rd January 2024, the Arbitrator closed submissions and proceeded to determine the Award on the evidence he had, under his power to do so under s.41(7)(c) AA if a party had failed to comply with a peremptory order.

On 21^st February 2024, the Arbitrator issued the Award. AASTAR’s claim succeeded on 13 of the 16 disputed contracts. It failed in proving the debt claimed on AAS22-009 from 15^th December 2021; AAS21-161 from 28^th December 2021, and AAS22-026 from 8^th February 2022. The total award was $299,000 plus not the €1,895,099.61 as claimed in the Notice of Arbitration - but €2,264,648.33 (even missing off the inconsequential €0.40 award on AAS22-003); with costs of £17,994. The Award was emailed to the parties (once again to ADC’s generic emails) on 21^st February 2024.

This brings me to the key factual dispute, as Mr Ferrigno for AASTAR points out in total from 2022-2024, 26 emails from itself, its lawyers or GAFTA were sent to ADC’s generic email addresses and says: ‘it was not a case of a single email slipping through the net’. M Niangadou’s statement retorts that:

“Mr Ferrigno[‘s statement] expresses…disbelief 26 emails related to the GAFTA arbitration and sent to the generic email addresses would have been ‘missed’. I confirm I have not seen these emails at any time; and to the best of my knowledge, none of my ADC colleagues have seen them either. In fact, during the preparation of this testimony, I did a new search in the inboxes of generic email addresses and did not find the emails regarding the arbitration contained there….I don’t know why these emails weren’t received - it could be that because they contained similar subject lines and content (referring to GAFTA) and came from people we didn’t receive emails from, they were filtered through ‘spam’ filters and over time were deleted. This is speculation …and I am not certain.”

I tease apart two factual issues here: (i) whether the Notice of Arbitration and arbitral emails were in fact received in ADC’s generic email addresses; and (ii) if so whether anyone at ADC actually read them. AASTAR’s Singaporean lawyers (unlike in Bernuth) did not request a ‘read receipt’ with the email attaching the Notice of Arbitration, nor did GAFTA with later arbitral emails.

On the first factual issue, M Niangadou says ‘he does not know why these emails were never received’, which is consistent with what Mr Abu-Manneh earlier said his client (i.e. M Niangadou) believed. So, whilst the burden of proving non-receipt is on ADC, as Mr Wassouf said, it has evidence of non-receipt, bearing in mind the difficulty of proving a negative. But Mr Lamming picked holes in M Niangadou’s statement on when and which email inboxes had been checked and the absence of forensic computer evidence almost a year into this litigation. He also pointed to the emails (including the one sending the Notice of Arbitration on 20^th July 2023) as apparently correctly addressed, with no evidence of ‘bounce back’ to AASTAR or GAFTA. So, I accept ADC’s evidence of non-receipt of the emails (including the Notice of Arbitration) is quite limited at this stage for the purposes of Kalmneft factor (vi). But as with summary judgment, I can also consider the possibility at any trial of stronger (e.g. forensic) evidence; and without an ‘opened’ read receipt, I am simply not in a position now to find as a fact those emails were even received. But equally, as ADC’s actual evidence of ‘non-receipt’ is currently quite limited, all I can realistically say at this stage is that ADC’s case that it did not receive relevant emails, particularly the Notice of Arbitration, is – as I will explain, ‘arguable’, but not ‘strong’ under Kalmneft factor (vi).

However, I can go further towards ADC on the issue whether the Notice of Arbitration and arbitral emails were opened and read by anyone at ADC. M Niangadou accepts he speculates in suggesting the emails, even if received, may have been (automatically) filtered as spam and deleted over time, But in the absence of such read receipts or other evidence (as opposed to AASTAR’s assertion) of receipt, that speculation is plausible. After all, the emails were sent from new and unfamiliar email addresses to ADC with a new subject line to ADC’s generic emails. Indeed, it is just as plausible with 26 emails as with one. Email has come on a long way in the 20 years since Bernuth, where arbitration emails to a generic email address used for other purposes were mistakenly deleted as ‘spam’ by employees in unrelated roles who had no idea of their significance. Nowadays, this ‘spam filtering’ is commonly done automatically, as all of us could testify who have at times been relieved to find an important email has gone mistakenly but automatically into our ‘junk folder’. This is one reason why the CPR is cautious of e-mail service, as Lord Sumption explained in Barton v Wright Hassall [2018] 1 WLR 1119 (SC) at [17], which though said in the context of solicitors, illustrates general practical drawbacks with the use of email for crucial legal communications:

“There are moreover particular problems associated with electronic service, especially where it is sought to be effected on a solicitor. A…solicitor’s office must be properly set up to receive formal electronic communications such as claim forms. As the Law Society’s Practice Guidance on electronic mail (May 2000) points out, ‘e-mail presents new problems, because it can arrive unperceived by other members of staff’. The volume of e-mails and other electronic communications received by even a small firm may be very great. They will be of unequal importance. There must be arrangements in place to ensure that the arrival of electronic communications is monitored, that communications constituting formal steps in current litigation are identified, and their contents distributed to appropriate people within the firm, including those standing in for the person primarily responsible for the matter when he is unable to attend to such communications as they arrive.”

Therefore, for all AASTAR’s complaints about what is and is not in M Niangadou’s statement, all this evidential doubt could have been removed by the simple step of AASTAR’s Singaporean lawyers on 20^th July 2023 requesting a read receipt for the Notice of Arbitration email (assuming it was received); and GAFTA staff doing likewise with arbitral emails later. For AASTAR, their Singaporean lawyers’ initial omission creates an evidential gap about whether the Notice of Arbitration was read. An email read receipt would have shown (at least without contrary forensic evidence) whether the email attaching the Notice of Arbitration had been opened. If it had been, the obvious inference would also be that the email and its contents had been read. But I do not have such evidence and indeed I do have positive evidence from M Niangadou that the email attaching the Notice of Arbitration and the relevant emails from GAFTA were never seen. Indeed, this gives rise to a separate point about the GAFTA emails I touch on later. Therefore, I consider ADC’s case relevant emails were not opened and read as reasonably (but without forensic evidence not very) strong. However, I will consider later whether service requires emails to be received or to be opened and read.

ADC however does accept M Niangadou himself saw the Arbitration Award on 8^th July 2024. As Mr Abu-Manneh gives a different account but was not personally involved, I will stick with the account of M Niangadou himself, subject to Mr Wassouf’s proviso mentioned above reflecting Mid-East Sales - that M Niangadou on this point is only giving evidence of fact that he was advised by an Ivorian lawyer, not ‘indirect evidence’ of Ivorian law. M Niangadou says that on 8^th July 2024, AASTAR’s Ivorian Counsel involved a bailiff to notify ADC of its proceedings to enforce the Award in the Ivorian Courts. The bailiff attended ADC’s office and provided M Niangadou with some arbitration documents, including the Award. M Naingadou says he did not examine the Award in detail and so was unaware of the purpose of the arbitration, the basis of the proceedings on which the Award had been decided and the conclusions of the Arbitrator. M Niangadou sought advice from ADC’s Ivorian Counsel about how to respond; and says the Counsel advised him that as the enforcement proceedings were ‘ex parte’ and ‘irregular’, ADC:

“…should simply ignore them, pending the outcome of the enforcement proceedings. [If] it were granted, we would have the opportunity to challenge the decision in a timely manner, but that no action was possible at this stage. We therefore chose to ignore the documents in question and take no action, in accordance with what I was told was our right under Ivorian law’.

However, on 30^th July 2024, ADC was served, again by physical delivery by a bailiff at ADC’s premises, with the Ivorian enforcement proceedings and at that stage M Niangadou was advised to examine the documents, which he did. ADC first applied in the Ivorian proceedings on 9^th August 2024 to set aside the enforcement order; and then on 27^th August 2024 (just within 28 days of ‘formal service’ on 30^th July in the Ivorian enforcement proceedings), ADC issued the present proceedings in London to challenge the Award.

Three grounds of challenge are set out in the Claim Form (as later amended after Mr Ferrigno’s statement in March 2025, by the consent of the parties, because of the omission of the extension of time application under s.80(5) AA, which later still, was conceded to be required, as I have explained above):

“The grounds on which these applications are made are set out in detail in the first witness statement of Mr Abu-Manneh. In summary:

GROUND 1 - Sections 67 and 72: Lack of jurisdiction

6.1

the Tribunal in the arbitration did not have jurisdiction over the Claimant in these proceedings;

6.2

that is because the Tribunal was not properly constituted in that the procedure provided for in the GAFTA Rules was not followed;

6.3

the Defendant never properly served such a notice on the Claimant, in accordance with Rule 3.1(a) of the GAFTA Rules;

6.4

as such, GAFTA’s appointment of a sole arbitrator pursuant to Rule 3.1(c) of the GAFTA rules was improper, as GAFTA’s

obligation to appoint an arbitrator under Rule 3.1(c) was never triggered; 6.5 further or alternatively, the Claimant seeks a declaration from the Court pursuant to s.72(1)(b) of the AA 1996 that the Tribunal was not properly constituted.

GROUND 2 - Section 68: Procedural irregularities

7.1

the Tribunal's and GAFTA's conduct in the proceedings constitutes a serious procedural irregularity within the meaning of sections 68(2)(a), (c) and (e). This is because:

7.2

the Tribunal’s use of two generic email addresses for purposes of “notifying” the Claimant was a failure to comply with its obligation to act fairly and impartially as between the parties;

7.3

it was additionally a failure to give each party a reasonable opportunity of putting its case and dealing with that of its opponent in accordance with its duty under s.33(1)(a) of the AA 1996;

7.4

the Tribunal’s finding that mere service on GAFTA should mean that documents “were correctly served upon both parties was a failure to conduct the proceedings in accordance with the procedure agreed by the parties, because service did not occur in accordance with the GAFTA Rules or the underlying contracts; and

7.5

GAFTA’s exercise of its power to appoint an arbitrator under Rule 3.1(c) of the GAFTA Rules exceeded its powers for the reasons given above in GROUND 1.

GROUND 3 - Section 70(3) and CPR 62.9(1): Extension of time…

8.1

the Claimant was notified of the Award on 30 July 2024 and filed the arbitration claim form on 27 August 2024, within the 28-day time limit set out in section 70(3) of the AA 1996. If the Court takes the view that the time limit is properly measured from an earlier date, the Claimant should be granted an extension of time to challenge the Award pursuant to CPR 62.9(3). That is because: (a) the Claimant did not have notice of the Award until long after 21 February 2024; (b) when it received proper notice of the Award it acted promptly to challenge it both in Cote d’Ivoire and in England (within 28 days of receipt); and (c) there is a real risk of substantial and irremediable injustice if the Award is permitted to go unchallenged by the Claimant in circumstances where this arbitration has concluded with a substantial damages award in the absence of the Claimant, and in circumstances in which the Claimant has substantive defences.”

Following the service of AASTAR with these proceedings in March 2025, in May 2025, the current hearing of ADC’s extension of time application under s.80(5) AA was listed for 18^th July 2025. M Niangadou’s statement was served on 7^th July, which I admitted de bene esse. At the hearing, as I have said, I queried the correctness of the concession (albeit only in relation to s.72(1) AA), which ADC had made only days beforehand, that an extension of time was required for all its claims, notwithstanding its pleaded case set out above that it was not. However, as I have said, for this case, I will only decide whether ADC’s s.72(1) claim is arguable, as it is relevant to Kalmneft Factor (vii) – fairness – of the extension of time, given the possibility of a (albeit limited) alternative remedy. This does not prejudice AASTAR in resisting the extension, since the existence of that remedy may assist it in doing so (as opposed to potentially prejudicing AASTAR on the underlying claim, which it can address on another occasion if it arises). Indeed, it is also a point of statutory interpretation and law which Mr Lamming had already fully researched and briefly addressed in his skeleton and which is fully-articulated in Merkin and Flannery. I therefore turn to that point first.

The s.72(1) point is an issue of statutory interpretation. The modern approach following R(O) v SSHD [2023] AC 255 (SC) at [29]-[31] and [41] is to ‘seek the meaning of the words which Parliament used’ in the provision and the ‘internal aids’ of the legislation itself; but tested against its ‘external aids’, like pre-legislative reports; and interpretative presumptions. I will consider s.72(1) and its ‘statutory setting’ in the AA; then Bernuth and The Prestige (No.1) that considered its pre-legislative context, before reaching my own conclusion.

s.72 AA is the key section, although as will become apparent, s.72(2) is a supplementary provision to its surrounding sections – what I will call its ‘statutory setting’ - but first s.72 AA itself states (my italics):

“72 Saving for rights of person who takes no part in proceedings.

(1)

A person alleged to be a party to arbitral proceedings but who takes no part in the proceedingsmay question (a) whether there is a valid arbitration agreement,(b) whether the tribunal is properly constituted, or (c) what matters have been submitted to arbitration in accordance with the arbitrationagreement, by proceedings in the court for a declaration or injunction or other appropriate relief.

(2)

He also has the same right as a party to the arbitral proceedings to challenge an award (a) by an application under section 67 on the ground of lack of substantive jurisdiction inrelation to him, or(b) by an application under section 68 on the ground of serious irregularity within the meaning of that section affecting him; and section 70(2) (duty to exhaust arbitral procedures) does not apply in his case.”

I have used those italics to highlight three preliminary observations about s.72:

ss.67-68 are part of a group of sections from ss.66-71 headed ‘Powers of the Court in relation to an award’. Most are also qualified by s.73 (along with s.72 in the next group of sections headed ‘Miscellaneous’), preventing substantive jurisdiction and some other challenges by participants who could have but did not object about it to the arbitral tribunal, or did not challenge its jurisdiction ruling within time. But s.73 does not apply to non-participants, as is made clear by the opening words of each of its subsections, limiting their scope to ‘parties to arbitral proceedings’. ss.66-71AA materially state as follows:

“66.— Enforcement of the award.

(1)

An award made by the tribunal pursuant to an arbitration agreement may, by leave of the court, be enforced in the same manner as a judgment or order of the court to the same effect….

(3)

Leave to enforce an award shall not be given where, or to the extent that, the person against whom it is sought to be enforced shows that the tribunal lacked substantive jurisdiction to make the award. The right to raise such an objection may have been lost (see section 73)…..

[N.B. s.71 explains consequential effects of variations or setting aside of awards].

“70.— Challenge or appeal: supplementary provisions

(1)

The following provisions apply to an application or appeal under section 67, 68 or 69.

(2)

An application or appeal may not be brought if the applicant or appellant has not first exhausted (a) any available arbitral process of appeal or review, and (b) any available recourse under section 57 (correction of award or additional award).

(3)

Any application or appeal must be brought within 28 days of the date of the award or, if there has been any arbitral process of appeal or review, of the date when the applicant or appellant was notified of the result of that process….”

That last provision is obviously central to this case and s.80(5) AA adds that:

“Where any provision of this Part requires an application or appeal to be made to the court within a specified time, the rules of court relating to the reckoning of periods, the extending or abridging of periods, and the consequences of not taking a step within the period prescribed by the rules, apply in relation to that requirement.”

The relevant ‘rule of court’ to s.70(3) and s.80(5) AA is CPR 62.9, stating:

“(1)

The court may vary the period of 28 days fixed by section 70(3) of the 1996 Act for – (a) challenging the award under section 67 or 68 of the Act; and (b) appealing against an award under s.69 of the Act.

(2)

An application for an order under paragraph (1) may be made without notice being served on any other party before the period of 28 days expires.

(3)

After the period of 28 days has expired (a) an application for an order extending time under paragraph (1) must (i) be made in the arbitration claim form; and (ii) state the grounds on which the application is made; (b) any defendant may file written evidence opposing the extension of time within 7 days after service of the arbitration claim form; and (c) if the court extends the period of 28 days, each defendant’s time for acknowledging service and serving evidence shall start to run as if the arbitration claim form had been served on the date when the court’s order is served on that defendant.”

(I discuss later some of the key case-law on ss.70(3), 80(5) and CPR 62.9).

Whilst it is reasonably settled that there is no specific time-limit for s.72(1) AA (because it does not rely on s.67-68, so s.70-71 do not apply to it either), it is hotly debated whether s.72(1) is limited to pre-award applications. On one hand, the learned authors of Russell on Arbitration at paras.7-152 and 8-063 suggest s.72(1) can and has been used to challenge arbitral awards for lack of substantive jurisdiction without time-limit, like the award procured by fraud set aside 2½ years later in Arab National Bank v El-Abdali [2005] 1 Lloyds Rep 541. On the other hand, the learned authors of Merkin and Flannery on the Arbitration Act 1996 at para 72.2.6 (who admit to having changed their minds), conclude s.72(1) is limited to pre-award applications. They point to the ‘present tense’ of s.72(1) - whether there ‘is’ (not ‘was’) a valid arbitration agreement, whether the tribunal ‘is’ properly constituted and whether matters ‘have been submitted’ to arbitration in accordance with the agreement all ‘in fact suggest relief ‘being granted during the currency of the arbitration’. They contrast what they call the ‘past tense’ of s.72(2): ‘he also has the same right as a party…to challenge an award (a) by an application under s.67’. They say:

“In summary, s.72(1) applies pre-award and any post-award relief seeking to attack an award has to be made under s.72(2).”

Mr Lamming not only adopted, but skilfully (if briefly) developed this point to argue s.72 was intentionally ‘chronological’: (1) is before an award (e.g. an anti-arbitration injunction); and (2) is after the award with a time-limit via s.70(3) but with an extension of time under s.80(5), so that there is no need for s.72(1) to apply to awards, as it would then overlap entirely with s.72(2).

The learned authors of Merkin and Flannery and Mr Lamming also rely on the observations of Christopher Clarke J (as he then was) in Bernuth. Like this case, there was email service of a Notice of Arbitration which, as I explain later, he held was valid. But he also noted at [57] a fallback argument by the arbitral non-participant only made at the hearing that it could apply to set aside the award under s.72(1). At [58], Clarke J rejected that argument (my italics):

“It seems to me that [Counsel for the participant] is on strong ground in saying that section 72(1) must have a more limited scope than section 67(3) at least to the extent that, if an applicant seeks, as Bernuth does, to set aside the award, he must proceed under section 67. The distinction between section 72(1) and section 72(2) appears to reflect the distinction between an application for a declaration under section 67(1)(b) and a challenge to an award under section 67(1)(a) with a consequent order under section 67(3). But the distinction between declaring an award to be of no effect because the tribunal did not have substantive jurisdiction and setting it aside would not appear to be major and it is not immediately apparent why an application under section 72(1) should be subject to no time limit, whereas an application under sections 67 and 72(2)(a) is subject to the time limit for an application under section 67 specified in section 70(3). Section 72(1) seems to be primarily intended to deal with the position at an interlocutory stage, when the court may be prepared to declare that an applicant is not bound by the arbitration agreement and to restrain the respondent from further continuance of the arbitration …[Counsel for the non-participant]submitted that s. 72(1) was the principal provision to be invoked if substantive jurisdiction was challenged and that s.72(2)(a) was inserted as an abundance of caution.”

I italicise those two passages, because neither go quite as far as Mr Lamming and the learned authors of Merkin and Flannery suggest. Clarke J in Bernuth at [58] said s.72(1) was ‘primarily intended’ to deal with interlocutory applications, but they go further and say it is limited to such applications. Clarke J did not say that s.72(1) could not in principle apply to awards at all, but rather (albeit ‘at least’) that it could not be used to set aside awards.

As I shall explain, Clarke J’s analysis in Bernuth is reconcilable with Walker J’s analysis in The Prestige (No.1) that arose out of an oil spill on the Spanish coast. Spain initiated criminal proceedings against the shipowners and their insurance ‘Club’. The Club refused to participate and initiated an arbitration in London in which Spain refused to participate and the award found the Club not liable to Spain. The Club sought to enforce in England, but Spain disputed substantive jurisdiction under s.66(3) under both s.72(2)/s.67 (though needed a short extension of time) and s.72(1). Walker J extended time for the s.67 challenge, largely based on the principle articulated by the Supreme Court in Dallah v Pakistan [2010] 3 WLR 1472 articulated by Lord Collins at [98] that a party objecting to arbitral jurisdiction can either challenge it in the courts of the arbitral seat, or in the courts where the award is sought to be enforced.

However, in The Prestige (No.1), Walker J also concluded obiter s.72(1) was also available to Spain, notwithstanding Bernuth. He gave five reasons at [82]:

I respectfully consider The Prestige (No.1) can be reconciled with Bernuth. Firstly, whilst s.72(1) may be ‘primarily intended to deal with the position at an interlocutory stage’ (Bernuth), ‘it is not necessary to confine s.72(1) to the position before an award’ (Prestige). Secondly, neither case suggested s.72(1) was the subject of a time-limit; and since s.72(1) is discretionary, this could include refusal on grounds of delay if it has caused substantial prejudice etc. Thirdly, whilst Clarke J said in Bernuth at [58] that ‘if an applicant seeks to set aside the award he must proceed under s.67’, Walker J in The Prestige (No.1) did not say that s.72(1) could be used to set aside an award (as opposed to injunct it or declare it invalid). However, as Merkin and Flannery point out, in Sino Channel Asia v Dana [2016] Bus LR 778, Eder J not only agreed with Walker J’s analysis in The Prestige (No.1), he did use s.72(1) to set aside an award (despite refusing an extension of time), which is more difficult to square with Bernuth. As noted later, the Court of Appeal in Sino ([2018] Bus LR 532) allowed the appeal on the jurisdiction issue, but made no comment about the scope of s.72(1). As recently as June 2022, in a full review of case-law on extensions of time in Hays v Bloomfield [2023] 1 Lloyds Rep 519, Henshaw J at [104-[105] referred to The Prestige and Sino, but made clear he was not asked to resolve the scope of s.72(1), implying that it is not yet well-settled.

I turn to my own conclusion. With real respect and deference to the far greater expertise and authority of the learned authors of Merkin and Flannery, I consider s.72(1) can be used to ‘question’ and give declaratory and injunctive relief about arbitral awards, not just pre-award applications; and potentially even set them aside, but only with caution. I reach that view for five reasons.

Firstly and most simply, whilst the views of Clarke J in Bernuth and Walker J in The Prestige No.1 were obiter, as the learned authors of Merkin and Flannery point out, the view of Eder J in Sino was strictly ratio and the Court of Appeal in reversing him did not criticise his view on s.72(1). Whilst none of the decisions bind me, I should follow a later decision which considers earlier conflicting ones unless I am convinced it is wrong: Changtel v G4S [2023] BCC 143 (HC) at [118]. Whilst I have doubts whether s.72(1) extends to setting-aside awards for the reasons Clarke J gave in Bernuth, given the uncertainty in judicial and academic views, I am not convinced Eder J was wrong in Sino, so will follow his decision. In all other respects, I respectfully entirely agree with Eder J’s and Walker J’s views, for the following reasons.

Secondly, I focus on the wording of s.72(1) itself which I repeat underlined:

“A person alleged to be a party to arbitral proceedings but who takes no part in the proceedingsmay question (a) whether there is a valid arbitration agreement,(b) whether the tribunal is properly constituted, or(c) what matters have been submitted to arbitration…by proceedings in the court for a declaration or injunction or other appropriate relief.”

I accept the present tense of (b) ‘whether the tribunal is properly constituted’ does hint at current arbitral proceedings. However, whether there ‘is’ a valid arbitration agreement and what matters ‘have been’ submitted to arbitration could equally be said after an award, but no-one says their temporal scope is different. Three cases in Merkin v Flannery not cited to me shed a little more light on the language of s.72(1). Broda Agro v Toefler [2010] 2 CLC 621 (CA) held a party ‘takes part’ in arbitration – so is excluded from s.72 - even if they only partly participate. So, logically, there may remain doubt whether they have ‘taken no part in proceedings’ until afterwards (see Tomlinson J in Peterson Farms v C&M [2004] Lloyds Rep 614 at [23]). But equally, it is clear from Fiona Trust v Privalov [2006] Bus LR 686 (CA) that s.72(1) is available pre-award. This suggests Parliament did not intend (see R(O) at [31]) s.72(1)’s present tense to limit it to pre-award (e.g. by saying: ‘a person who is taking no part’). But surely it would not hint with ambiguous tenses, but simply use express words e.g. ‘may question before the arbitral award’. Or as in s.72(2), Parliament could have adapted other AA rights for non-participants e.g. the power under s.32 AA for a party during arbitration to apply to court on ‘any question as to the substantive jurisdiction of the tribunal’. (If s.72(1) AA is only pre-award, one might have expected to find it adjacent to s.32, not adjacent to ss.66-71 related to awards). In my view, the language of s.72(1) is not only available for pre-award applications, but is also open to applications after awards, or is at the very least ambiguous about that.

This leads on to my third reason, which interprets the neutral (or ambiguous) language of s.72(1) in the light of s.72(2), which I partly repeat and underline:

“He also has the same right as a party to the arbitral proceedings to challenge an award (a) by an application under s.67 on the ground of lack of substantive jurisdiction inrelation to him…”

As Walker J said in The Prestige (No.1) at [82(3)], ‘he’ at the start of s.72(2) is the same ‘person’ referred to in s.72(1). Whilst the learned authors of Merkin and Flannery suggest s.72(2) is in the past tense, to my mind it too is in the present tense, in speaking of the rights that non-participant person ‘has’. Moreover, s.72(2) does not say ‘he can only challenge an award’ under ss.67 or 68, but rather he also has the same right as a participant to challenge an award under ss.67-68, implying those are additional rights to challenge awards – i.e. additional to the right in s.72(1). So, s.72(2) actually implies that s.72(1) is open to post-award applications questioning substantive jurisdiction.

Fourthly, there is how s.72(1) AA ‘fits’ its statutory setting in ss.66-73 AA, especially relief under s.67 via s.72(2)(a). Clarke J in Bernuth at [58] thought s.72(1) was primarily intended for pre-award applications, since if it applied to awards that would overlap with s.72(2)(a) but without its time-limit. But this overlap in relief but not in time-limit may be explicable on three bases:

Finally, this ‘flexible’ and broad interpretation of s.72(1) AA I have just summarised is in my view consistent with: (i) its pre-legislative context; (ii) recent amendments to the AA; and (iii) the presumption of Access to Justice. (I was not addressed about (ii) and (iii), but they simply confirm my view):

ADC’s argument that the Arbitrator lacked substantive jurisdiction under s.67 AA and its s.72(1) claim both depend on whether AASTAR validly served the Notice of Arbitration on ADC on 20^th July 2023 at ADC’s two generic email addresses. GAFTA’s later emails are more relevant to whether there was a serious procedural irregularity under s.68 AA. Indeed, in Bernuth, Clarke J at [54] and [59] suggested invalid service of a notice of arbitration meant the arbitral tribunal was not properly-constituted under s.72(1)(b)/s.67, while s.68 was for cases where the tribunal had substantive jurisdiction, but ‘failed properly to comply with its obligations’ causing ‘substantial injustice’. But ss.67-68 do overlap on the issue of service as Clarke J said in Bernuth at [63]:

“[I]f the proceedings were not validly served so…Bernuth had not had a reasonable opportunity of putting their case, they would have suffered a substantial injustice [So, it would not] be just to allow the award to stand.”

The AA statutory framework set out above needs a little elaboration about email service. The relevant provisions were set out by Popplewell J (as he then was) in Glencore v Conqueror [2017] Bus LR 2090 at [19-[20] (which I slightly adjust for the circumstances in this case and italicise):

“14 Commencement of arbitral proceedings

…(4) Where the arbitrator or arbitrators are to be appointed by the parties, arbitral proceedings are commenced in respect of a matter when one party serves on the other party or parties notice in writing requiring him or them to appoint an arbitrator or to agree to the appointment of an arbitrator in respect of that matter.

16 Procedure for appointment of arbitrators

…(3) If the tribunal is to consist of a sole arbitrator, the parties shall jointly appoint the arbitrator not later than 28 days after service of a request in writing by either party to do so.…

(4)

If the tribunal is to consist of two arbitrators, each party shall appoint one arbitrator not later than 14 days after service of a request in writing...

17 Power in case of default to appoint sole arbitrator

(1)

Unless the parties otherwise agree, where each of two parties to an arbitration agreement is to appoint an arbitrator and one party (‘the party in default’) refuses to do so, or fails to do so within the time specified, the other party, having duly appointed his arbitrator, may give notice in writing to the party in default that he proposes to appoint his arbitrator to act as sole arbitrator….

76 Service of notices etc

(1)

The parties are free to agree on the manner of service of any notice or other document required to be given or served in pursuance of the arbitration agreement or for the purposes of the arbitral proceedings.

(2)

If or to the extent that there is no such agreement the following provisions apply.

(3)

A notice or other document may be served on a person by any effective means.

(4)

If a notice or other document is addressed, pre-paid and delivered by post….(b) where the addressee is a body corporate, to the body’s registered or principal office, it shall be treated as effectively served.”

AASTAR says the Notice of Arbitration emailed on 20^th July 2023 was both a notice of commencement under s.14(4) AA and a nomination of arbitrator under s.16(3), but that as ADC failed to respond, a sole arbitrator was appointed under s.17(1). There is no dispute in this case about any of that. The dispute is whether the use of ADC’s generic email was effective service under s.76 AA.

It is ADC’s case under s.67 and s.72(1) AA that there was no effective service of the Notice of Arbitration on 20^th July 2023. As Mr Wassouf says (on authority discussed below), it is only necessary in a time-limit extension application to have a high-level consideration of the merits (Kalmneft Factor (vi)), but he contends the merits of ADC’s case on non-service are strong. But as I have said, I cannot at this stage find as a fact whether the Notice of Arbitration (or the GAFTA emails) here were actually received by ADC, though its evidence of non-receipt is quite limited but arguable under factor (vi). By contrast, ADC’s evidence that the emails were not opened or read is reasonably strong, which I can take into account for ADC. However, before considering the Kalmneft factors, it is convenient to address three submissions of principle by Mr Wassouf on factor (vi) (albeit in reverse order): (i) The parties agreed in the contracts under s.76(1) that there would be postal service of arbitration (which was effective under s.76(4)); (ii) email service was ineffective under the GAFTA Rules which the parties agreed under s.76(1); or (iii) even if the default provision s.76(3) applied, email service to ADC’s generic emails was not ‘effective means’ because ADC never agreed to the use of the generic emails for service; and the case of Bernuth that held that was unnecessary under s.76(3) is distinguishable.

In Bernuth, Clarke J at [22]-[25] noted s.76(3) AA was different than the then-provisions of CPR 6 which prohibited service by email without prior written consent of the recipient. In Bernuth, the notice of arbitration was emailed by the shipowner to the charterer’s generic email address on its website and in the Lloyds Directory (but only practically used for general business matters). It was admitted the notice was received but misunderstood as ‘spam’. Clarke J said:

“28.

I do not regard the provisions of CPR Part 6 as an appropriate benchmark by which to judge whether or not service by e-mail is effective in the context of an arbitration. The [CPR] cater for litigants of all kinds from major corporations represented by the most accomplished..solicitors to [those] represented by more modest firms and those…not represented at all. By contrast arbitrations are usually conducted by businessmen represented by, or with ready access to, lawyers. s.76(4), when providing that a notice could be served on a person by any effective means, was in my judgment purposely wide. It contemplates that any means of service will suffice provided that it is a recognised means of communication effective to deliver the document to the party to whom it is sent at his address for the purpose of that means of communication (e.g. post, fax or e-mail). There is no reason why, in this context, delivery of a document by e-mail – a method habitually used by businessmen, lawyers and civil servants – should be regarded as essentially different from communication by post, fax or telex.

29.

That is not to say that clicking on the ‘send’ icon automatically amounts to good service. The e-mail must, of course, be despatched to what is, in fact, the e-mail address of the intended recipient. It must not be rejected by the system. If the sender does not require confirmation of receipt, he may not be able to show that receipt has occurred. There may be circumstances where, for instance, there are several e-mail addresses for a number of different divisions of the same company, possibly in different countries, where despatch to a particular e-mail address is not effective service.

30.

But in the present case none of those difficulties arise. The e-mail[s]… were received at an e-mail address that was held out to the world as, and so far as the evidence shows, the only e-mail address of Bernuth. Someone looked at the e-mails on receipt and, apparently, decided that they could be ignored, without making any contact with the sender. The position is, to my mind, no different to the receipt at a company’s office of a letter or telex which, for whatever reason, someone at the company decides to discard. In both cases service has effectively been made, and the document received will, in the first instance, be dealt with by a clerical officer….

34.

In short, I do not regard service of the…notice [of arbitration] by e-mail as ineffective because, when it was received, a particular employee did not think that a serious legal matter would be sent to that address. That e-mail and those that followed it, are plain and straightforward in their terms. They bear none of the hallmarks of ‘spam’. On the contrary they called for serious attention….[The first email] purported to initiate arbitration proceedings by calling for agreement as to an arbitrator. I should be surprised if much junk e-mail purports to do that or to emanate, as later e-mails did, from an LMAA arbitrator. If the e-mails never reached the relevant managerial and legal staff, that is an internal failing which does not affect validity of service and for which Bernuth has only itself to blame. Having put info@bernuth.com into the current Lloyd’s Maritime Directory as their only e-mail address, they can scarcely be surprised to find an e-mail inviting them to agree to appointment of an arbitrator in a maritime matter was sent to that address.

35.

I do not accept that, in an arbitration context, in order for service to be effective, it is essential that the e-mail address at which service is purportedly made has been notified to the serving party as an address to be used in the context of the relevant dispute. Section 76 does not say as much and there is no basis upon which that can be implied.”

Of course, in the last 20 years since Bernuth, business communications have changed dramatically, but notably relevant provisions of the CPR for service of proceedings and documents have not substantially changed. CPR 6.3 for Claim Forms, CPR 6.20 for other documents and in each case PD6A para 4 still only permit service by email if (i) the party to be served or their solicitors have agreed in advance in writing to accept service by email and (ii) nominated a specific email address. This continuing caution was explained by Lord Sumption in Barton at [17] (quoted above) by reference to the risk of miscommunication. In Barton a litigant-in-person emailed a claim form on the last day before its expiry to solicitors who had not accepted service by email. The Court held this was invalid service and there was no ‘good reason’ to validate it retrospectively under CPR 6.15 which did not violate Art.6 ECHR, as the CPR was clear and justified. Lord Sumption explained the purpose of having strict rules of service at [16]:

“Although the purpose of service is to bring the contents of the claim form to the attention of the defendant, the manner in which this is done is also important. Rules of court must identify some formal step which can be treated as making him aware of it. This is because a bright line rule is necessary in order to determine the exact point from which time runs ……. [I]t has never been enough that the defendant should be aware of the contents of an originating document such as a claim form. Otherwise, any unauthorised mode of service would be acceptable, notwithstanding that it fulfilled none of the other purposes of serving originating process.”

Clarke J’s differentiation of the default service rules as including email under s.76(3) AA for arbitration from email service under CPR PD6A para 4.1 is entirely consistent with Lord Sumption’s analysis in Barton. It is not that arbitral claims to court are exempt from the CPR: The Maria K [2003] 2 CLC 1 (CA) held that faxed service of a claim form challenging an arbitral award was invalid for similar breach of the predecessor of CPR PD6A para 4.1. Rather it is that in Bernuth at [28], Clarke J simply made the point that the ‘bright-line rule’ on service of a notice of arbitration under s.76(3) AA in the commercial context of arbitration did not have to be the same as for service of a claim form in the ‘one size fits all’ CPR. That is (very respectfully) plainly right and no more violates Art.6 ECHR than the stricter rule under the CPR violated it in Barton.

However, the analysis of service in Bernuth has been qualified by a different principle - agency, considered in Sino and Glencore. As noted, Sino was unusual (Gross LJ dryly remarked at [5] it showed ‘there was no shortage of variety in commercial transactions’). The claimant charterer contracted with the defendant shipowner via a third party ‘front company’ run by an agent (‘C’) purporting to be employed by the claimant but who ran the contracts via his front company. When a dispute arose, the defendant emailed C the notice of arbitration and the claimant argued that was invalid service, so the arbitrator lacked substantive jurisdiction. As I said, Eder J set aside the arbitral award under s.72(1), as he held emailing the notice of arbitration to C was invalid service. He said at [47]:

“[E]ven where an employee or agent has a wide general authority to act on behalf of his employer/principal, [it] does not (without more) generally include an authority to accept service of a notice of arbitration.”

Whilst the Court of Appeal did not disagree with that as a matter of principle, it disagreed on what Gross LJ also called at [47] a ‘most unusual case’ on the facts. More widely, Gross LJ re-iterated his own earlier observations at first instance in Lantic Sugar v Baffin [2010] 2 All ER (Comm) 1170 at [39]:

“s.76 was intended to operate more flexibly than the regime for service provided in respect of court proceedings by the CPR. So, by way of examples, notices commencing arbitration (ss.14 and 76 AA) should be interpreted broadly and flexibly rather than strictly or technically………. So too, s.76 is more flexible than the CPR as to mode or means of service …. But…even assuming there are differences between the issuing of claim forms and the service of notices [of arbitration]…[s]eparate corporate personality cannot simply be ignored…[N]othing in the Act, authority or principle exempts an arbitration claimant from serving a notice commencing arbitral proceedings on the correct party.”

Consequently, in Glencore, Popplewell J held that emailing a notice of arbitration to the email address of a specific employee of the respondent who had conducted the contract (but who happened to have just left its employment before the notice) was not valid service under s.76 AA. Unlike the unusual agent in Sino, in Glencore that employee (even if he had remained) did not have actual or ostensible authority to accept service of it, by email or otherwise. But Popplewell J distinguished service by generic email address as in Bernuth:

“25.

There is a distinction…between an e-mail address which is a personal business address of an individual, and one which is generic. Individual e-mail addresses are readily recognisable as in the form name@domainname. By contrast businesses often have generic e-mail addresses which do not identify the name of the person who will receive the e-mail sent to that address [e.g.] Info@glencore.com…Where an individual e-mail address is used, the sender will reasonably expect the e-mail to be opened and read by the named individual. With a generic e-mail address, the sender will not, at least usually, know the identity of the person who will open and read the e-mail. [But] if the generic address has been promulgated by the organisation, whether on its website or otherwise, the sender can reasonably expect the person who opens the e-mail to be authorised internally to deal with its contents if the subject matter falls within the scope of the business activity for the purpose of which the generic e-mail address has been promulgated. A promulgated generic chartering department e-mail address gives rise to the legitimate expectation that communications relating to chartering operations sent to that address will come to the attention of a person …internally authorised to deal with them. This is true also of those who open post addressed to a company address. The company can be expected to ensure the letter or e-mail is opened by someone with internal responsibility for putting it in the hands of whoever needs to deal with it on behalf of the company, and if he fails to do so, that risk falls on the company.

26 An e-mail sent to an individual employee’s e-mail address is different. It is being sent specifically to a named individual within the organisation. ….Whether it constitutes good service if directed to an individual’s e-mail address must depend upon the particular role which the named individual plays or is held out as playing within the organisation.”

Therefore, the principle in Bernuth that a respondent’s generic email address can be ‘effective means’ of service under s.76(3) AA, even without prior written agreement to its use, is consistent with the principles of service in Barton and agency in Sino. However, on agency, Popplewell J in Glencore clarified Bernuth by saying the generic email address must have been ‘promulgated’ by the party to be served, ‘on its website or otherwise’ so ‘the sender can reasonably expect the person who opens the email to be authorised internally to deal with it’. In Bernuth, the generic email had been ‘promulgated’, so service using it was valid.

Mr Wassouf sought to distinguish Bernuth on two bases. Firstly, the generic email in Bernuth was held out on Bernuth’s website (and in the Lloyds Maritime Directory) as its only email address, whereas that is not the case here. Secondly in Bernuth, there were read receipts showing relevant emails and notices had been read (which indeed was undisputed) which there were not here. I accept these are relevant points of factual distinction, but not of legal distinction:

The GAFTA Rules No.125 relevantly stated at the time that (my italics):

“1.2: The juridical seat of the arbitration shall be and is hereby designated pursuant to section 4 of the Arbitration Act 1996 as, England…

3.1(a) If he requires the appointment of a sole arbitrator, the claimant shall, before expiry of the time limit… serve a notice on the respondent, seeking his agreement to the appointment of a sole arbitrator by Gafta.

3.1(c)…Where…the respondent has not responded to the claimant’s notice under 3.1(a) above… Gafta shall appoint an arbitrator…

4.4: All submissions and evidence shall be served by sending them to the other party, with copies to Gafta…

21.1: All notices to be served on the parties pursuant to these Rules shall be served by letter, fax, or E-mail or other electronic means. For the purposes of time limits…date of despatch shall, unless otherwise stated, be deemed to be the date of service. Service on…agents named in the contract shall be deemed proper service under these Rules. So far as concerns such notices, this Rule over-rides any other provisions of the contract.”

In LFT Agroneftiprodukt v Ameropa [2021] EWHC 3474 (Comm) (to which I was not referred asit related to a point not presently in issue – the form of a notice of arbitration), Sir William Blair described para 3.1(a) GAFTA Rules No.125 as ‘reflecting the language’ of s.14(4) AA on notices of arbitration. Just as such notices should be interpreted commercially, the GAFTA Rules themselves should be interpreted in the context of the AA, especially as the GAFTA Rules specify London as the arbitral seat. (I return to this issue at the end of this judgment, given the arbitration clauses in the contracts in the present case assumed the seat would be Singapore – also specified by the exclusive jurisdiction clause - but the Arbitrator’s jurisdictional decision that the arbitral seat was London is not presently challenged by ADC). As with Rule 3.1(a) and s.14(4) AA in Ameriopa, Rule 21.1 is consistent with s.76(3) AA in making it clear that email service is valid under GAFTA Rules. Indeed Rule 21.1 does not apparently even require the same ‘promulgation’ of an email address for service as s.76(3) AA does in the sense in Bernuth and Glencore. That said, the GAFTA Rules do not entirely mirror the AA: GAFTA Rule 3.1(c) displaces s.17(1) AA and dispensed with the need for AASTAR to nominate its own arbitrator given ADC’s failure to respond; as that is done by GAFTA itself, as it did on 6^th September 2023. Moreover, GAFTA Rule 4.4 stated from that point onwards, all communications were via GAFTA. Whilst ADC initially suggested there was invalid service of the Notice of Arbitration because it was served by GAFTA not AASTAR as s.14(4) AA required, as I have discussed, this was ADC’s mistake: GAFTA did not email the Notice of Arbitration, AASTAR’s lawyers did and Mr Wassouf did not really pursue this. Whilst I cannot at this stage dismiss it, ADC’s argument that email service breached GAFTA Rules is weak.

Indeed, finally under ‘service’, the concluding sentence of GAFTA Rule 21.1 states that it overrides any contractual provision as to service in the contract. So, even if the contracts in the present case had specifically required service of notices of arbitration by post as Mr Wassouf submitted, that would be over-ridden by GAFTA Rule 21.1 which permits email service. Whilst Mr Wassouf pointed to s.76(4) AA stating post service is valid, it does not say only postal service is valid – s.76(3) AA says service may be by ‘any effective means’. In any event, as Mr Lamming said, the contracts did not specify postal service, they merely gave ADC and AASTAR’s postal addresses by way of identification of the parties and their contact details. Indeed, email was built-in to the contractual process because ADC had to sign, stamp and email a scanned copy of each contract to AASTAR. This was entirely sensible – one contractual party (who drafted the terms) was in South-East Asia, the other in West Africa. Requiring arbitral service by post would have been a surprising commercial decision; and as it was not required by GAFTA Rule 21.1, one would expect clear contractual words to that effect, which are simply not there. So, neither s.76(3) AA, nor the GAFTA Rules, nor the contracts, make ADC’s legal or evidential case under s.67 or s.72(1) AA that there was not valid service more than simply ‘arguable’.

Finally, I turn to ADC’s extension application and my conclusions on it. For convenience, I repeat the material parts of s.70(3) and s.80(5) AA and CPR 62.9, as well as the well-established Kalmneft factors for such extensions of time:

“70(3) Any application…must be brought within 28 days of…the award.

80(5) Where any provision of this Part requires an application or appeal to be made to the court within a specified time, the rules of court relating to the extending or abridging of periods… apply….”

The relevant ‘rule of court’ for s.70(3) and s.80(5) AA is CPR 62.9(1), stating:

“The court may vary the period of 28 days fixed by section 70(3) of the 1996 Act for – (a) challenging the award under section 67 or 68 of the Act; and (b) appealing against an award under section 69 of the Act.”

I also note the Commercial Court Guide (2022) para. O9.2 (which applies in arbitration claims in the Circuit Commercial Court under its Guide at para L1.5) states any application for extension of time will require ‘cogent reasons’.

Whilst both Counsel have touched on the well-known principles of relief from sanction under CPR 3.9, an application for an extension of time to issue an arbitral challenge under s.80(5) AA is governed by a different rule of court: namely CPR 69.2. But there is a resemblance between the principles and some of the language of the three factors in Denton v White [2014] 1 WLR 3926 (CA) and the seven Kalmneft factors stated by Colman J in Kalmneft at [59]:

“Accordingly, although each case turns on its own facts, the following considerations are…likely to be material: (i) the length of the delay; (ii)  whether, in permitting the time limit to expire and the subsequent delay to occur, the party was acting reasonably in all the circumstances; (iii)  whether the respondent to the application or the arbitrator caused or contributed to the delay; (iv)  whether the respondent to the application would by reason of the delay suffer irremediable prejudice in addition to the mere loss of time if the application were permitted to proceed; (v)  whether the arbitration has continued during the period of delay and, if so, what impact on the progress of the arbitration or the costs incurred in respect of the determination of the application by the court might now have; (vi) the strength of the application; and (vii) whether in the broadest sense it would be unfair to the applicant for him to be denied the opportunity of having the application determined.”

However, prior to dismissing the application, at [60] of Kalmneft, Colman J did stress that the relative weight to be given to the factors in a given case was influenced by the international arbitration context he had discussed at [57]-[58]:

“57.

In approaching the identification of the applicable criteria…at least in international arbitrations, English arbitration is probably the most widely chosen jurisdiction of all. It is chosen because of the ready availability of highly skilled and experienced arbitrators operating under a well-defined regime of legal and procedural principles in what is often a neutral forum. Supervisory intervention by the courts is minimal and well defined and the opportunities for a respondent with a weak case to delay the making of an award or to interfere with its status of finality are very restricted. Accordingly, much weight has to be attached to the avoidance of delay at all stages of an arbitration, both before and after an interim or final award. If the English courts were seen by foreign commercial institutions to be over-indulgent in the face of unjustifiable non-compliance with time limits, those institutions might well be deterred from using references to English arbitration in their contracts. This is a distinct public policy factor which has to be given due weight in the discretionary balance.

58.

On the other hand, it has to be recognised that because of the extremely wide international nature of the market for English arbitration many of the parties may be located in remote jurisdictions and may have little or no previous experience of international or English arbitration. When these relatively unsophisticated parties find themselves involved in..an arbitration it is only to be expected they move somewhat more tentatively than would an international trading house well experienced in this field. It would therefore be wrong to fail to make at least some allowance for this factor in evaluating the element of fault in failing to comply with time limits.”

As noted in later cases e.g. Hays, the Kalmneft factors were not criticised, indeed effectively endorsed, by the Court of Appeal in The Maria K at [38]-[42] (and later Broda Agro at [52]-[54]). However, whilst Mance LJ (as he then was) in The Maria K at [39] referred to Kalmneft factors (i)-(iii) as ‘primary factors’, that simply related to how the first instance judge had applied them, not as a general comment (as Henshaw J explained in Hays at [63]). Nevertheless, Mance LJ’s comment at [42] of The Maria K was clearly of general application:

“[A]s to factor (vii), general considerations of fairness…they must ….always be viewed in the particular context that Parliament and the courts have repeatedly emphasised the importance of finality and time limits for any court intervention in the arbitration process.”

Moreover, as Colman J said in Kalmneft at [59], his seven factors are not necessarily exhaustive. One additional factor is the ‘Dallah principle’ that a party objecting to arbitral jurisdiction is entitled to refuse to participate until there is an application to enforce. In The Prestige (No.1) at [79]-[81], Walker J said it was so fundamental that it justified delay between promulgation of an award and enforcement (and delay after enforcement was down to Spanish law).

In addition, Counsel cited various first instance cases on arbitration extensions of time, including Kalmneft, Brown, Terna Bahrain v Al-Shansi [2013] Lloyds Rep 86, S v A&B [2016] 1 CLC 638, P&ID v Nigeria [2020] EWHC 2379, Minister of Finance (Malaysia) v International Petroleum [2021] EWHC 2949 (Comm) and Hays. Most were reviewed in Hays and I need not consider them extensively. Instead, I shall refer to those cases as relevant when addressing the various Kalmneft factors. Given that Hays has clarified factors (i)-(iii) are not ‘primary factors’ as such; but in this case (ii) and (iii) on one hand and (iv) and (vii) on the other are inter-related, I will start with factor (vi) - merits - as I have already discussed it, then factor (i), then (ii) and (iii) together with the Dallah principle, then factors (iv), (v) and (vii) together, before my overall conclusion.

In the context of an application to extend time for an arbitral challenge under s.80(5) AA and CPR 62.9, the strength of the challenge may not always be a particularly crucial factor since, as Mance LJ observed in The Maria K at [41], it may not be so clearly strong or so weak when assessed at an interlocutory stage as to affect the discretion to extend time. As Mr Wassouf observed, this resembles the Court’s caution in assessing the merits in summary judgment applications (e.g. by consideration of other evidence that might realistically - as opposed to ‘Micawberistically’ – emerge). This caution was explained by Popplewell J (as he was) in Terna at [31]-[33] where he differentiated between a ‘rolled-up hearing’ as there, from the interlocutory approach described at [31]:

“…On an application for an extension of time, the court will not normally conduct a substantial investigation into the merits of the challenge...since to do so would defeat the purposes of the Act. However, if the court can see on the material before it that the challenge involves an intrinsically weak case, it will count against the application for an extension, whilst an apparently strong case will assist the application. Unless the challenge can be seen to be either strong or intrinsically weak on a brief perusal of the grounds, this will not be a factor which is treated as of weight in either direction on the application for an extension of time. If it can readily be seen to be either strong or weak, that is a relevant factor; but it is not a primary factor, because the court is only able to form a provisional view of the merits, a view which might not be confirmed by a full investigation of the challenge, with the benefit of the argument which would take place at the hearing of the application itself if an extension of time were granted.”

As I noted above, a similar point was made about evidential conflicts in Brown.

I acknowledge Popplewell J in Terna spoke of essentially three options on Kalmneft factor (vi): an ‘intrinsically strong’ case pointing towards extension; an ‘intrinsically weak one’ pointing against; and neutrality like Thyssen/Brown. But those cases did not concern service, which Mr Wassouf rightly emphasised as fundamental: not just to the fairness or even jurisdiction of legal proceedings (arbitral or court), but their essential integrity. Therefore, as Kalmneft factor (vi) focusses on similar considerations as summary judgment, if ADC’s service challenges are neither ‘strong’ nor ‘weak’, I will still consider whether they are ‘arguable’ so as to count neither against nor strongly in favour of extension, but modestly in favour, which at least reflects that merits are not a primary factor.

Hays illustrates similar interlocutory caution – and distinctions in the strength of individual challenges. Hays was an application to extend time to challenge arbitral awards in New York law enforced in England, in which the extension applicant had not participated. It challenged them under s.72(2) AA: both for lack of substantive jurisdiction under s.67 AA (a ‘corporate veil’ party issue in New York law); and serious irregularity under s.68 AA as the arbitrator had calculated compound interest on an unrequested basis. Henshaw J held at [123]-[134] the s.67 jurisdictional challenge was not sufficiently strong to take it into account in the extension application. By contrast, he held at [108]-[122] that the s.68 challenge for procedural unfairness in interest calculations was strong. So, he extended time for the s.68 challenge, but not the s.67 challenge.

Mr Wassouf focussed on ADC’s s.67 AA jurisdiction challenge alleging non-service, which not only overlaps with the s.72(1) claim, but also part of the s.68 challenge at paras 7.4 and 7.5 of the Claim Form: alleging inadequate ‘service’ under GAFTA Rules or the contract; and so no power to appoint the Arbitrator. I will call these the ‘jurisdiction/service challenges’. Whilst there are subtle differences between paras 6.1-6.5 and 7.4-7.5, on a ‘high level consideration’, they all boil down to the same two issues of email service I have discussed:

However, tucked away in the Claim Form there is a distinct point on s.68 AA:

“7.2

the Tribunal’s use of two generic email addresses for purposes of ‘notifying’ the Claimant was a failure to comply with its obligation to act fairly and impartially as between the parties’; 7.3 it was additionally a failure to give each party a reasonable opportunity of putting its case and dealing with that of its opponent in accordance with s.33(1)(a) AA 1996.”

This is not a pleaded complaint about jurisdiction or service, but under s.68(2)(a) AA whether GAFTA failed to comply with its duty under s.33(1) to:

“(a)

act fairly and impartially as between the parties, giving each party a reasonable opportunity of putting his case and dealing with that of his opponent, and(b) adopt procedures suitable to the circumstances of the particular case, avoiding unnecessary delay or expense, so as to provide a fair means for the resolution of the matters falling to be determined.”

In Hays, Henshaw J said at [79]-[80] fairness under s.33 AA meant arbitrators must give parties a fair opportunity to deal with an issue, though Popplewell J in Terna at [85(4)-(5)] distinguished between not giving a party that opportunity and a party failing to take that opportunity. (I bore that elemental principle of fairness in mind on the s.72(1) point, which I raised and gave Counsel the opportunity to argue, but I have not finally decided it: only that it is ‘arguable’).

Nevertheless, whilst this ‘GAFTA fairness point’ (as I shall call it) is pleaded, paras.7.2-7.3 Claim Form do not explain how GAFTA acted unfairly by use of ADC’s generic emails, at least as distinct from ADC’s ‘jurisdictional/service challenges’ which focus on AASTAR(‘s lawyers): the Notice of Arbitration and its effect on the validity of the appointment of the Arbitrator etc. Mr Wassouf argued the Arbitrator should have also directed postal service of the Notice of Arbitration, but as I said, that was not required under GAFTA Rules or contracts and is not arguably ‘unfair’. Another possibility is that GAFTA should have checked during the arbitral proceedings whether ADC was receiving its arbitral emails (e.g. by requesting read receipts). But that would mean even if the Arbitrator had jurisdiction, he may be unfair in not proactively checking receipt of his own orders. That goes beyond not only the contracts and GAFTA Rules, but even the CPR which provides for default judgment: whereas the Arbitrator here made a reasoned award not accepting the claim in full. In any event, that is not how the GAFTA fairness point is pleaded or was argued. Were this a summary judgment application, this lack of clarity might lead to an order for particularisation. But in the context of Kalmneft factor (vi), the fair approach is not to ignore the ‘GAFTA fairness point’, but simply to say, echoing Terna and Brown, it is not an issue that points either in favour of or against extension. Indeed, I cannot even say it is ‘arguable’ when it has not been argued in detail.

In Hays at [62] and [64] Henshaw J summarised principles on length of delay:

“62.

Popplewell J in Terna…[at [27(i) stated]: ‘s. 70(3) of the Act requires challenges to an award under sections 67 and 68 to be brought within 28 days. This relatively short period of time reflects the principle of speedy finality which underpins the Act, and which is enshrined in s.1(a) ……… The [applicant needs to] show the interests of justice require an exceptional departure from the timetable laid down by the Act. Any significant delay beyond 28 days is to be regarded as inimical to the policy of the Act’….

64…(i) The facts of the individual case must be ‘considered with care. There is no principle of law that any particular length of delay either cannot ever be unjustified, at one extreme, or will always be unjustified at the other extreme’ (Minister of Finance [127])….. (ii) The length of delay must be judged against the yardstick of the 28 days provided for in the Act. Therefore, a delay measured even in days is significant, and a delay measured in many weeks or in months is substantial: (Terna [28]).”

Henshaw J also noted that whilst delays of even less than 28 days had been treated as significant and substantial, extensions of 6 weeks had been granted in some cases; and in Minister of Finance itself, one of 5½ months’ culpable delay was granted after 11 months of non-culpable delay. Indeed, in P&ID, an extension was granted of 4½ years from the initial award and almost 3 years from the final award, though in that extraordinary case, there was a strong prima facie of fraudulent conduct of the arbitration, even bribery of the innocent party’s own legal team. The real point is simply that there is no maximum extension.

Since the Award in this case was dated 21^st February 2024, but this claim was not issued until 27^th August 2024, it is 159 days (over 5 months) beyond the 28-day deadline of 21^st March 2024. On any view, 159 days is a substantial period. However, it is helpful to look at the delay in this case in three sub-periods:

On the reasonableness of applicants’ delay, in Hays at [66], Henshaw J stressed:

“…the importance of adducing evidence [explaining delay] rather than relying only on submissions….Whilst a delay occasioned by lack of awareness of the time limit, or by ‘laxity, incompetence or honest mistake’ can hardly be described as ‘good’ reasons for delay, they will weigh less against an applicant than ‘a deliberate informed choice’.”

However, there are different types of ‘deliberate informed choice’. Henshaw J in Hays at [66] there was referring to what Popplewell J had said in Terna at [29]-[30] about intentional non-compliance as in Terna, which raises the public interest in respect for legal rules in a way inadvertent non-compliance does not. But sometimes the law respects such a ‘deliberate informed choice’: for example, as the Dallah principle respects a decision not to participate in arbitration on jurisdictional grounds, but then to contest jurisdiction on enforcement, which I was not addressed about, but will consider briefly in a moment. Then there is a deliberate informed choice not to get English legal advice on English arbitration. I was addressed by both Counsel about ADC in sub-period 2 getting Ivorian advice on resisting enforcement of the Award (albeit not waiving privilege on it), but not getting English advice on challenging the Award. This can be relevant to an applicant’s reasonablenessy under Kalmneft factor (ii), as it was in Kalmneft itself where Colman J held a failure to get English legal advice on an English award was unreasonable given the arbitrator’s advice to do so and so did not simply reflect a lack of familiarity with international arbitration. This was discussed in Hays at [69(i)] and Henshaw J then at [69(ii)] discussed Broda Agro at [52]-[56], where the Court of Appeal endorsed Teare J’s refusal to extend time for a Russian party experienced in arbitration who relied on Russian lawyers’ advice about a GAFTA arbitral award in London which disagreed with a Russian court judgment, although that party clearly knew about the London arbitration.

Rather than considering sequentially reasonableness of ADC (and its lawyers) under Kalmneft factor (ii), then contribution of AASTAR/lawyers and GAFTA / the Arbitrator under Kalmneft factor (iii), I will consider those factors together under sub-period 1 then 2. I start with sub-period 1 and Kalmneft factor (iii).

I turn to sub-period 2 - the 22 days between 8^th July 2024 when ADC admits first seeing the Award and 30^th July 2024 when it was first formally served with it in the Ivorian enforcement proceedings. Mr Wassouf does not suggest AASTAR or GAFTA made any contribution under Kalmneft factor (iii) for this. Counsels’ debate was on ADC’s reasonableness under Kalmneft factor (ii) on three points:

That one point is the Dallah principle, on which I was not addressed on the extension of time (as opposed to briefly on the s.72(1) point). However, in The Prestige (No.1), Walker J found that it justified deliberate delay in contesting an award by a non-participant until enforcement. As he said at [55]-[57], in Dallah Pakistan refused to take part in an arbitration in Paris contending it was not a party to the arbitral agreement. The Supreme Court held a party who declined to participate in an arbitration on jurisdictional grounds was not bound by an arbitrator’s jurisdiction decision. Lord Collins stated the key principle at [98]:

“[A] party which objects to the jurisdiction of the tribunal has two options. It can challenge the tribunal's jurisdiction in the courts of the arbitral seat; [or] resist enforcement in the court [of] recognition and enforcement.”

In The Prestige (No.1), Walker J said at [79] and [81]:

“The Dallah principle is….so fundamental that it should not be whittled down unless the interests of justice so require. The contention [that] inaction following the…Award, but prior to notice of the application to enforce, must count against [Spain] when seeking extensions of time…[in circumstances] of the present case…seems to me an impermissible whittling down of [it].. [But] I do not rule out the possibility in other cases the Dallah principle may not for some particular reason carry as much weight as..in the present case.”

However, the Dallah principle respects deliberate choice not to participate in arbitration on jurisdictional grounds, not delay due to claimed ignorance of it. Whilst the Dallah principle must not be whittled down, nor should it be extended to different scenarios. It was not relevant to a party who did not participate due to cost but then contested jurisdiction in Hays, despite Dallah being cited (on s.67 being a rehearing not a review of the arbitral decision, now adjusted slightly by the 2025 Act). Therefore, in my view the Dallah principle is simply irrelevant to sub-period 1 (or only of very minor weight). Whilst it may be relevant to sub-period 2 as ADC deliberately chose to await enforcement, that was after discovering the Award, not a choice to not participate in arbitration in disputing jurisdiction or being a party to arbitral agreement as in Dallah and Prestige. So even if the Dallah principle applies to sub-period 2, it is only of minor weight.

I would cautiously say ADC’s inadvertent delay of 109 days in period 1 weighs modestly against extension, but its considered delay for a further 22 days in sub-period 2 weighs more strongly, albeit tempered to an extent by the Dallah principle. So factor (ii) weighs significantly against extension and factor (iii) is neutral as AASTAR is not responsible for either sub-period (or if 1, modestly).

I deal with these factors together, because factor (v) – the impact on arbitration - is not significant as the arbitration is over, so it adds little to factor (iv) – the prejudice to AASTAR beyond loss of time if the application were refused. That in turn here is the corollary of factor (vii) – whether it would be unfair to ADC to deny it the opportunity to challenge the award by refusing the application. In relation to factor (iv), in Hays at [70]-[71], Henshaw J observed that:

“70.

As indicated in Kalmneft, it is relevant to consider whether the respondent to the application would by reason of the delay suffer irremediable prejudice in addition to the mere loss of time if the application were permitted to proceed.

71.

On the other hand, as the Court of Appeal confirmed in [The Maria K] at [39] citing Euston, an extension of time may be refused even in the absence of such prejudice. As Steyn LJ said in Euston, by reference to a predecessor to the Act…It is not only the private interests of the parties that are relevant. There are wider interests at stake, notably the proper functioning of our arbitration system ... One of the aims of the Act of 1979 was to promote speedy finality in the enforcement of arbitration awards… Since nobody can prevent the losing party in an arbitration from applying for leave to appeal even in the most unmeritorious cases, it is of supreme importance to the proper working of our arbitration system there must be an effective procedure to ensure applications for leave are promptly made…”

It follows from that citation (and the similar observations of Andrew Baker J in Malaysia at [124(d)] that it is also the policy of the 1996 (and the 2025) Acts.

In relation to factor (vii), I have already quoted Mance LJ’s observation in the Maria K at [42] that fairness to the applicant has to be viewed in the context that: ‘Parliament and the courts have repeatedly emphasised the importance of finality and time limits for any court intervention in the arbitration process’. Indeed, as is clear from Euston cited in the Maria K, this was so even before the 1996 Act. Moreover, I also note this observation by Teare J at first instance in Broda Agro at [57] cited and approved by the Court of Appeal in that case and quoted in Hays at [69] on the issue of English advice but of wider scope (my italics):

“Broda says that it would be unfair because it faces an award of $6m in circumstances where it has not had its evidence on the question as to whether there was a contract considered by either the arbitration tribunal or the court. If one assumes that its claim that there was no contract is arguable this is an undoubted hardship. But the question is whether that hardship is unfair.”

As I started this judgment by saying – and the reason I spent so long on obiter comments on s.72(1) which are not decisive of it, factor (vii) (and to an extent, factor (iv) as well) are affected by whether ADC has an alternative remedy under s.72(1) AA. However, this is only true of ADC’s ‘jurisdiction/service challenges’ under s.67/68 AA. By contrast, any s.72(1) claim for ADC is effectively irrelevant to ADC’s s.68 ‘GAFTA fairness’ point. Therefore, as with Kalmneft factor (vi), it is necessary to consider these two challenges separately, but on each one, to consider the inter-related questions of prejudice to AASTAR of extension and fairness to ADC of its refusal together.

On ADC’s jurisdictional/service challenge that it was never served with the arbitral proceedings - principally under s.67 AA but overlapping with s.68 AA – in my judgement, it has in principle an alternative remedy under s.72(1) AA even if I do not extend time. Whilst my decision does not bind the parties, if it is right, that is relevant both to the fairness to ADC of refusing an extension (Kalmneft factor (vii)) and to the prejudice to AASTAR of granting it (Kalmneft factor (iv)). However, ADC’s jurisdictional/service challenge is no more than ‘arguable’ on the facts and if s.72(1) cannot apply to an award, there is no alternative remedy. Even if I am right about s.72(1) AA on the law, ADC’s claim under s.72(1) would also have additional hurdles to its s.67 challenge in AASTAR’s procedural objections such as the concession or delay. But those are down to ADC’s own conduct and are just factors to be assessed in the round, especially as its pleaded declaration that the Arbitrator was not properly constituted is not strictly ‘s.67 relief’ engaging the concern in Bernuth. So, whilst I understand Mr Wassouf’s caution whether s.72(1) will assist ADC, if it can actually assemble cogent expert evidence (as opposed to M Niangadou’s limited belated evidence) of non-receipt of the Notice of Arbitration etc in its generic email addresses; and/or persuade a judge it did not ‘promulgate’ them, in my view, s.72(1) may mitigate the risk of unfairness to ADC in refusing an extension if it turns out that it never was served (which even after almost a year of this litigation, it has still not evidenced as any more than arguable). However, I stress this is only a risk of unfairness, not cogent evidence of it in the context of the importance of time-limits: Maria K. Whilst AASTAR facing s.72(1) anyway reduces its prejudice if an extension is granted, that would still cause it prejudice of costs thrown away on arbitration and enforcement: and respondent prejudice is not required: Hays at [70]-[71]. So, on the jurisdiction/service challenge, if I am right about s.72(1), Kalmneft factors (iv) and (vii) do not support extension. If I am wrong about s.72(1), unfairness to ADC remains only an unproven risk that only modestly supports extension given the importance of time-limits.

But ADC’s s.68 ‘GAFTA fairness point’ is not a jurisdictional challenge so cannot fall within s.72(1) AA. Consequently, refusing an extension means the argument cannot be run. However, its strength is neutral and it has not been properly articulated as distinct from the jurisdiction/service challenges, so that result would not be unfair. Moreover, unlike those challenges, the GAFTA fairness point is not AASTAR’s responsibility. Therefore, granting an extension would be hard on AASTAR who would suffer real prejudice. So, Kalmneft factor (iv) weighs against extension and more than balances Kalmneft factor (vii).

Finally, I will consider all the Kalmneft factors in the round on the ‘GAFTA fairness point’ and the ‘jurisdictional/service’ challenges respectively:

So, I dismiss ADC’s s.67 and s.68 challenges, but there remains its s.72(1) claim. I am grateful to Mr Wassouf and Mr Lamming not only for the very high quality of their submissions (especially on the s.72 point which I raised at the hearing), but also to them and those that instruct them that after the circulation of my draft judgment, directions were agreed (save as to costs, where I ruled ADC should pay AASTAR’s costs of and occasioned by the extension application hearing). In short, the parties have permission to instruct a single joint forensic IT expert and to file further lay evidence afterwards; and the case can then be listed for three days in the New Year to include any application by ADC to amend its claim.

That said, I hope that the parties will instead resolve their differences in the meantime by a sensible commercial agreement, so both can get on with running their businesses and put this dispute behind them.