Visa Inc & Ors v Luxottica Retail UK Limited [2026] EWHC 615 (Comm)

[2026] EWHC 615 (Comm)Case No CL-2023-000455
IN THE HIGH COURT OF JUSTICE
KING'S BENCH DIVISION
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
COMMERCIAL COURT
Venue Royal Courts of Justice, Rolls Building, Fetter Lane, London, EC4A 1NLDate 17 th March 2026
Before
Paul Stanley KC(sitting as a Deputy High Court Judge)
Between
VISA INCClaimantsVISA EUROPE LIMITEDClaimantVISA EUROPE SERVICES LLCClaimantVISA UK LIMITEDClaimantLUXOTTICA RETAIL UK LIMITEDDefendant
Daniel Piccinin KC and Ava Mayer (instructed by Linklaters LLP) for ClaimantsRichard Coleman KC and Philip Ahlquist (instructed by Russell-Cooke LLP) for DefendantHearing Hearing dates: 3-5, 9-10 February 2026
Approved JudgmentThis judgment was handed down remotely at 10.30am on 17 th March 2026 by circulation to the parties or their representatives by e-mail and by release to the National Archives..............................

Paul Stanley KC:

[1]When a retailer accepts a card payment from a customer, its bank pays a fee—known as a multilateral interchange fee (“MIF”)—to the bank that issued the card. The fee is usually set by the operator of the card payment system, though not paid to it. Although the MIF is not directly paid by the retailer, it is likely to be reflected in the charges that the retailer’s bank makes for allowing the retailer to accept card payments.[2]For more than a decade, litigation about MIFs has been taking place in (at least) England, Europe, and the United States. The complaints have been based on alleged breaches of competition law, with retailers alleging that Visa (and MasterCard) set MIFs at artificially high prices. In England, those complaints have been based on domestic and EU law. In the US, they have been based on the Sherman Act. They have involved many retailers, and raised complex and controversial legal and factual issues.[3]The claimant, Luxottica Retail UK Ltd (“Luxottica”), is part of a large corporate group. It sells eyewear and accepted card payments directly and through concessions at department stores in the UK, Jersey, and Ireland. The claimants—whom I will call “Visa”—operated the Visa card system, and set MIFs. In 2017, Luxottica issued a claim form in the High Court, claiming damages for MIFs it had paid. That claim form was never served. But it led to settlement discussions. In October2 2020, Luxottica and Visa agreed on a settlement figure of £200,000. In the succeeding months they agreed a written settlement agreement, which was executed on 20 January 2021.[4]Meanwhile, a then-unrelated company, GrandVision NV (“GrandVision”) and various other entities, had also begun its own MIF claim against Visa in 2018. In mid-2019, Luxottica’s parent company had agreed in principle to acquire a controlling interest in GrandVision. That acquisition completed in July 2021. A few months later, Visa told Luxottica that it thought the settlement agreement had, thanks to GrandVision’s acquisition by the Luxottica group, also settled GrandVision’s claims. It wanted Luxottica to see to it that they should be brought to an end, and to compensate Visa if they were not. This, Visa says, results from the clear terms of the settlement agreement. Luxottica says this is quite wrong—indeed “absurd”—and that if it is what the settlement agreement means then it is the result of “sharp practice” on Visa’s part, and should not be permitted.[5]I therefore have to decide three questions. First, as a matter of construction, does the settlement agreement have the effect for which Visa contends, or does it—as Luxottica maintains—leave GrandVision’s claims untouched? Second, if it does encompass GrandVision’s claims, would enforcement in those terms reward “sharp practice” which the law will prevent? Third, if the agreement is enforceable, what relief should be given to Visa?[6]I have concluded that the settlement agreement does oblige Luxottica to ensure that the GrandVision claims are withdrawn, and to indemnify Visa for them. It is not unconscionable to enforce those obligations. Visa is entitled to declaratory relief, and to damages; but I decline to grant specific performance.

The evidence

[7]I heard more extensive evidence than would be appropriate in relation to a question of construction. The ostensible reason for this was the existence of the sharp practice allegation. That left the door ajar for witnesses to offer opinions on the meaning of the contract, what they had been seeking to achieve, the commercial fairness or unfairness of that, and many things which would be inadmissible in support of any argument about construction. That had spread like woodworm throughout the statements, and neither party asked me to rule any of it, as evidence, inadmissible. Inevitably, cross-examiners took up the cudgels on some of these points.[8]I have approached this evidence as follows: i) In case it matters for any appeal, I make findings about any potentially material aspect of a witness’s (or a party’s) subjective knowledge, belief, or opinion at the relevant time, reached on a balance of probability. ii) However, in so far as witnesses were formulating or challenging arguments about the settlement agreement’s true legal meaning, I go no further than to decide whether those were opinions held at the time. I keep my own conclusions about the agreement’s meaning separate. Whether witnesses agree or disagree with my views does not matter. iii) I have explained which aspects of the witness and documentary evidence I regard as relevant and admissible to construction, and which I do not. Where evidence is not relevant or admissible, I have not relied on it in reaching my conclusions about construction.[9]A second respect in which the evidence calls for careful handling arises from Visa’s assertion of privilege over most communications with its lawyers. Visa has steadfastly resisted waiving privilege, as is its right. But it has often sought to adduce evidence of what a lawyer knew, or thought, or considered, without waiving privilege, scattering the expression “without waiver of privilege, I can say …” or words to similar effect through its statements. The overall effect of this was, it appeared to me, often to encourage me to reach conclusions about the lawyer’s views, without quite revealing, though often hinting, at the process by which they had been formed.[10]For the most part, Luxottica did not attempt to go behind these assertions of privilege. On two occasions Mr Coleman KC, who represented Luxottica, persisted in a question to which objection was taken. In both cases, I ruled that the question should be answered. Despite the cooperation with which the pitfalls were negotiated, I was left with a sense that the content of privileged communications was being implied or hinted at, but without giving access to material from which the witness in question derived the alleged knowledge or belief, or by which it could be tested.[11]I should therefore be clear about how I approach this evidence, and inferences from it. I can—as a matter of law—draw no inference from the fact that privilege is asserted. I have not done so. The rule that adverse inferences cannot be drawn from the assertion of privilege, however, only goes so far. Privilege restricts the evidence available; it offers no positive foothold for adverse inference. But it does not prevent the court from drawing inferences from the evidence that is otherwise available. The assertion that a statement cannot be explained without describing privileged communications is not a trump card which constrains the court to accept the statement as accurate. Conclusory statements, especially vaguely conclusory statements, unsupported by documentary evidence or clear detail, are apt to carry little weight.[12]A third salient aspect is that Visa’s evidence comes from lawyers who remain engaged in a struggle over MIFs. They have an interest—not just in relation to this case but for thousands of others—in maintaining strategic positions. Luxottica, of course, also has an interest; so does Fieldfisher, who acted for it. All the witnesses I heard from were truthful. But many, especially on Visa’s side, had rehearsed talking points. They often appeared as advocates for their employer’s or client’s position. After years of close involvement in the mass tort claim that MIFs represents, they could hardly be expected to be objective. Some of the talking points (such as Visa’s insistence that it believes MIF claims to be fundamentally bad in law and settles only for commercial reasons) are irrelevant to the issues here. But in some respects, notably the insistence on the central importance Visa now attributes to group-wide settlements leading to perfect and eternal peace, I think that the pressures of this litigation, and of the MIF litigation in general, have given the witnesses a distorted view of reality. The point is not restricted to Visa’s witnesses, though my assessment was that it affected them more than Luxottica’s.[13]I thought all the witnesses were trying to tell the truth as they saw and recalled it. None gave evidence that I thought was deliberately misleading. But none of them was completely reliable when it came to points of detail. That is not surprising, given the time that has passed since the relevant events, and given that this agreement was not a momentous matter for any of them.[14]In brief, so far as Visa’s witnesses are concerned: i) Mr Cassels is an experienced partner at Linklaters. He gave his evidence fairly and directly. He was one of the senior lawyers supervising the MIF litigation. He was intermittently involved in the settlement with Luxottica, but it was clear from his evidence that Luxottica’s claim was not a high priority, and he largely left competent but more junior colleagues to deal with it. As one would expect, he had no real independent recollection—aside from what contemporaneous documents told him—of how it had progressed. ii) Ms Williams was, at the relevant time, a managing associate at Linklaters, on the cusp of partnership. The MIF litigation had occupied much of her active career, and she was steeped in it. She was, in practical terms, the most senior lawyer at Linklaters dealing with the Luxottica settlement terms. But she had many other—and more pressing—commitments on her time and attention. I do not mean that she gave Luxottica’s claim or its resolution less than the attention it deserved. But it did not need much. It was, in the context of the MIF litigation, an early and not especially threatening claim, advancing no novel legal argument, of relatively low value, with every indication that the claimant wished to settle. By this stage, Linklaters had a battle-tested way of handling that sort of claim, with standard-form letters and agreements which could be deployed with economy of effort. iii) Mr Winfield-Chislett was, at the time, a lawyer at Visa. He had trained and practised briefly at a well-regarded firm before joining Visa in 2017, and was obviously conscientious and intelligent, though not (at the time) deeply experienced in the MIF litigation, which he became involved with in mid-2020. For similar reasons to Ms Williams, he had no reason to pay special attention to Luxottica’s claim or its settlement, so long as that was within the range of values and on the general terms that Visa regarded as acceptable. iv) Ms Holtz was the most senior lawyer at Visa dealing with the MIF claims. Like Ms Williams, she was steeped in them. Of the witnesses called by Visa, her evidence seemed the most heavily varnished. Her experience of the MIF litigation has robbed her of objectivity. She will not contemplate (at least publicly) any suggestion that MIF claimants might have any legitimate complaint. I got the impression that she regarded them, and those who represent them, as opportunists grasping for whatever they can obtain. A witness who believes that retailers, such as Luxottica, who have obtained any compensation have got more than they deserved, and that those who obtain none get only what they deserve, will find it almost impossible to bring a balanced perspective. v) Mr Stokes retired as Executive Director, Competition Law, at Visa in 2017, but continued as a consultant until 2017. His evidence was not materially challenged.[15]I turn to Luxottica’s witnesses: i) Mr Pimlott is an experienced solicitor, who was at the relevant time a partner of Fieldfisher LLP, which represented Luxottica. His evidence struck me as completely straightforward. He accepted that Luxottica’s claim was part of a rather experimental venture (“dipping our toe in the water”) into MIF litigation for his firm and its clients. Although he and his colleagues had evidently considered the various decisions relating to MIFs which were being delivered over the relevant period, Fieldfisher did not aspire to be at the cutting-edge of the MIF claims, and had not explored all the by-ways. The firm’s practical focus was on reaching reasonable settlements which would yield compensation for clients who had paid MIFs on UK business. Like the other lawyers, his involvement had been intermittent; he had relied on other competent lawyers in the team; his recollection of detail was imperfect. ii) Mr Bacon became responsible for dealing with Luxottica’s claim from around May 2020. His expertise was in property management. He had only a rough idea what the MIF litigation was about, and relied entirely on Fieldfisher’s guidance for information about it, and on Mr Melani for key decisions. His evidence was completely truthful, but he understandably remembers little. iii) Mr Melani, who was a director of Luxottica, sometimes found it difficult to answer the questions he was being asked without going down side-roads. I did not think this was because he was being deliberately evasive. It was because he could see where the question was leading, and was keen to address its implications as well as its immediate substance. He obviously has no accurate detailed recollection of day-by-day events, but at best a general impression. So far as that went, I found it credible and fairly reported. As with Mr Bacon, I was not left with any sense that Mr Melani had considered Luxottica’s claim deeply. As far as he was concerned, it was effectively an asset—which Luxottica understood had some value, but which it was not willing to devote much time or money to maximising. That is not surprising, since Luxottica’s business was selling spectacles, not bringing litigation. He approached it in that pragmatic commercial spirit. The facts MIFs

The facts

[16]Visa operates payment card schemes, in the UK and elsewhere, for both debit and credit cards. The schemes are helpfully described in Sainsbury’s Supermarkets Ltd v Mastercard Inc [2020] UKSC 24, [2020] 4 All ER 588. They involve four main parties as well as the payment scheme operator. At one end are the retailer (“merchant”) and the purchaser (“cardholder”). The cardholder obtains a payment card from a bank (the “card issuer”). The issuer agrees to provide the cardholder with a payment card, and sets the terms and fees the cardholder will pay for using it. The merchant obtains payment services from another bank (the “merchant acquirer”). The merchant acquirer charges a fee (the “merchant service charge”) to the merchant. When a cardholder uses the card, its card issuer settles the payment with the merchant’s merchant acquirer. That happens on terms including a fee paid by the merchant acquirer to the card issuer. This is the MIF, which can be set at different rates for different categories of transaction. For instance, it may depend on whether the transaction relates to a debit card or a credit card, on whether the merchant acquirer and the card issuer are in the same country or region, on whether the payment is contactless or signed-for. Unless a particular merchant acquirer and card issuer have a separate bilateral arrangement, they deal on terms set by the scheme’s operator—and in practice this is what happens. The scheme operator does not receive the MIF (it earns its return through various other fees). The MIF is paid by the merchant acquirer. But it is likely to affect the fees it charges to the merchant, which may or may not in turn affect the price it charges its customers. The extent to which MIFs were “passed on” to retailers and then “passed on” by them was one area of dispute in the litigation over MIFs.

The MIF litigation

[17]In December 2007, the European Commission decided that Mastercard’s MIFs had—going back to 1992—been in breach of the provisions of European law (now Art 101 TFEU) which prohibit agreements that have the object or effect of preventing, restricting, or distorting competition within the internal market. Successive appeals by Mastercard to the General Court and the Court of Justice failed: Case T-111/08 MasterCard Inc v European Commission EU:T:2012:260, Case C-382/P MasterCard Inc v European Commission EU:C:2014:2201. (No EU-level decision finding infringement was ever taken against Visa.)[18]That led directly or indirectly to regulatory action in the EU, and to various competition claims in the English courts (where EU law at that time applied directly, and where there was and is an equivalent domestic provision to Article 101 TFEU in Section 2 of the Competition Act 1998).[19]The regulatory action was a regulation, adopted in 2015, fixing maximum MIFs on many intra-EU card transactions: Regulation (EU) 2015/751. One of its effects was that it was widely thought—during the period I am concerned with, at least—that challenge to intra-EU MIFs after 2015 was likely to be difficult, and the focus therefore lay on the MIFs that had been charged up to 2015.[20]The English litigation had mixed fortunes at different stages: i) Sainsbury’s began a claim against Mastercard in the Chancery Division. That claim related to UK MIFs. It was transferred to the Competition Appeal Tribunal in 2015. In July 2016, the CAT decided that Mastercard’s UK MIFs had been unlawful, and awarded damages: [2016] CAT 11. ii) Asda and Morrisons brought claims against Mastercard, in the Commercial Court. Those claims concerned UK MIFs, EEA MIFs, and Irish domestic MIFs. The trial was heard by Popplewell J (as he then was) in 2016. He delivered judgment in January 2017, dismissing the claims. iii) Sainsbury’s brought a separate claim, relating to UK MIFs fixed under Visa’s rules, in the Commercial Court. The trial was heard by Phillips J (as he then was) in late 2016, and early 2017. On 30 November 2017, he dismissed those claims [2018] 2 All ER 611. He also, and separately, found that if the MIFs had been restrictive, they would not have qualified for exemption. His reasoning was different from Popplewell J’s. iv) All three cases then proceeded to a combined hearing in the Court of Appeal. That court disagreed with aspects of all the judgments that were before it: [2018] EWCA Civ 1536, [2019] 1 All ER 903. It held that the MIFs set by Visa and Mastercard had restricted competition between merchant acquirers. But it also held that questions relating to the application of Article 101(3)—which may permit, in some circumstances, the justification of otherwise prohibited agreements—should be reconsidered. It remitted them in all the cases to the CAT. v) That decision was appealed to the Supreme Court. On 17 June 2020, the Supreme Court endorsed the Court of Appeal’s conclusion that MIFs had restricted competition. It reversed the decision to remit the issue of justification to the CAT in one of the three cases. In these respects, therefore, its judgment was beneficial to merchant claimants. However, another aspect of its decision concerned quantum, and how difficult it would be for scheme operators to show that retailers had passed on costs to their customers, through the prices they charged. In this respect, the Supreme Court accepted that a “broad axe” might be appropriately employed. That was beneficial to Visa. Furthermore, even in relation to exemption, the Supreme Court’s judgment did not close the door to arguments about exemption under Article 101(3). So, even after three first instance cases and two appeals, there remained a lot to play for in individual cases.[21]There was also litigation in the US, under the Sherman Act. I have less information about the history or nature of this litigation, which was not being handled by Linklaters. It is apparent from the Second Amended and Supplemental Complaint and Demand for Jury Trial that there were proceedings, entitled In Re Payment Card Interchange Fee and Merchant Discount Antitrust Litigation,which were being handled as Multi-District Litigation in the US District Court for the Eastern District of New York. They seem to have begun in 2013, although the amended complaint was filed in October 2017. The defendants included Visa Inc and Mastercard Inc. The plaintiffs were a variety of retailers, many household names. They included Luxottica US Holdings Corporation, with various subsidiaries including Sunglass Hut Trading LLC, and Oakley Inc, described as “wholesalers and retailers of iconic sun and prescription eyewear” who operate “more than 4,000 retail stores in the United States”. They alleged that Luxottica US Holdings Corporation has “been forced to pay Defendants’ supracompetitive interchange fees … [and] has been and will be injured in its business or property as a result of the unlawful conduct alleged herein”.[22]I know almost nothing about the history of this Sherman Act complaint. US Antitrust law is not identical to EU or UK competition law. But there undoubtedly was an overlap of concern. The core allegation was that MIFs had been set by Visa at higher levels than would have prevailed in a competitive market, and that retailers (including US companies that handled the Luxottica group’s business there) had suffered injury. They sought damages (including trebled damages) and injunctive relief for the future.[23]In different ways, the witnesses on all sides showed that they understood the MIF litigation to have some distinctive characteristics. Mr Melani described it as a “class action”. That is not technically correct; but it had the features of what US lawyers often call a “mass tort claim”. There were many claimants, and even more potential claimants. Their cases were in some respects similar or identical (particularly as to liability). But there were relevant individual differences when it came to quantum.[24]Visa faced many claims (in the end, I was told, more than a thousand—the number of possible claims must have been far, far higher). Collectively they presented a high worst-case exposure, and the principles to be applied had implications for Visa’s business model. It was therefore worth Visa’s while to devote large resources to the claims, and by 2017 the team at Linklaters handling them must have understood their ins-and-outs very thoroughly. Individuals’ claims, however, were of much lower value. A claim for £1.7m sounds like a lot, but the investment of money and management time to fight a full-scale competition law case over such a sum might have been commercially marginal.[25]What emerged were patterns familiar in mass tort cases. Claimants with the most to gain—such as the supermarkets, who initially started the ball rolling over MIFs—or collective actions make the running. Smaller claimants follow in their wake, and on their coattails. Firms representing them try to build portfolios of cases which can be efficiently run, and to acquire experience in the defendant’s approach which helps settle those cases favourably. Defendants try to thwart this, and manage the litigation as a whole. They aim to choose the cases that go to trial, to use developments in the litigation to settle other cases, disrupt attempts to get claimants to act collectively or share cost, and discourage the emergence of specialist claimant-side firms with large “books”.[26]Such litigation often mutates over time. A judicial decision may lead to new arguments being developed, new claimants being found, or large numbers dropping away. From a defendant’s point of view, settlements at opportune moments may reduce risk desirably; but known generosity may encourage copycats and, at worst, settlement sums may be used by claimants or firms to fund fresh waves of cases. Defendants settling such cases do so not only with an eye to the immediate benefit of a settlement, but to the medium- and long-term effects on the whole problem.[27]The MIF litigation was of this sort. That is how Visa understood it. It accords with Mr Melani’s description of how Luxottica understood it. I am sure it is how Fieldfisher understood it too. That broad general knowledge—that the Luxottica settlement agreement was not simply the settlement of an individual claim, but of a claim brought in the context of mass tort litigation over MIF fees in various forms—forms part of the factual matrix available to both parties.[28]There are two other aspects of the MIF-litigation that form part of the relevant context. MIFs were not merely historical. Visa continues to set them, albeit now at rates consistent with an EU Regulation. Its ongoing ability to do so was commercially significant. And neither MIFs, nor the arguments about them, were very retailer-specific: most large retailers would have been able to put together some sort of claim, although the precise arguments and quantum would have differed depending on the business, its location, and clientele.

Luxottica and the Luxottica Group

[29]Luxottica was a UK company, but it was part of a larger group. When the settlement agreement was concluded the group’s ultimate parent was a French company, EssilorLuxottica SA, which in turn had two intermediate holding companies, Essilor International (a French company) and Luxottica Group SpA (an Italian company). That structure derived from the combination of two previously separate groups (Essilor, and Luxottica) in around 2018. The group had numerous subsidiaries in many jurisdictions: Australia, Brazil, Canada, Chile, China, Colombia, France, Germany, Greece, Hong Kong, India, Ireland, Israel, Italy, Japan, Korea, Malaysia, Mexico, the Netherlands, Peru, Portugal, Russia, Singapore, South Africa, Sweden, Switzerland, Thailand, Turkey, the USA, Ukraine, the UAE, and the UK.[30]Luxottica’s main business—also carried on by some, but not all, the other companies in the group—was retailing sunglasses, spectacles, contact lenses and accessories. Its brands included Sunglass Hut, Oakley, and David Clulow opticians. It was a substantial enterprise, with reported revenue of £133m and an operating profit of £6.5m in 2019. That, however, was a small fraction of the total group revenue (EUR 17.4bn) and profit (EUR 2.8bn) in the same period.[31]I do not consider that the relevant context for interpretation includes assumed knowledge of EssilorLuxottica’s corporate structure in all its details. But it does, I think, include the readily available information that Luxottica was a relatively small part of a very large corporate group, with worldwide retail operations.

Luxottica’s High Court claim

[32]Luxottica’s claim was issued in the Chancery Division on 21 July 2017. This was after the CAT and Popplewell J’s decision, before Phillips J’s, and before any appeals. It combined (for convenience, and perhaps to save court fees) Luxottica’s claim with that of an entirely unrelated company which just happened to be another client of Fieldfisher. At this early stage, the litigation was being handled within Luxottica by Mr Colin Allen.[33]The Defendants were three Mastercard companies, and the four Visa companies who are claimants in this case, described as the “Visa defendants”. Luxottica was described as a “fashion retailer[] which accept[s] payment by Visa”. The claims were described in this way:
“The claims are for damages, arising from and in relation to infringements of Article 101 [TFEU] and/or Article 53 of the Agreement on the European Economic Area and/or Chapter 1 of the Competition Act 1998, as amended. The claims arise from restrictive arrangements, amounting to agreements, concerted practice and/or decision by associations of undertakings for the setting and implementation of [MIFs] on credit and debit card payments which were and remain part of the fees levied on the Claimants pursuant to … the Visa payment card system for their respective credit and debit cards and in which … the Visa Defendants … were and/or remain involved. … The Claimants claim as damages the amount of any overcharge found to exist by the Court.”
[34]The claim as so defined was not confined to MIFs paid on intra UK transactions but could have included other MIFs, if Luxottica had paid any. It was, however, confined to fees relating to Luxottica’s card transactions, and not those borne by any other company within Luxottica’s group.[35]Three days after the claim form was issued, Fieldfisher wrote to Linklaters. They explained that the claim form had been issued “for protective reasons” (to prevent claims being reduced by limitation periods expiring). Their letter identified Luxottica as a retailer selling its products through “wholly-owned retail premises and concessions throughout the UK” and through two online stores, and identified some of its brands. It set out the essential argument about the lawfulness of MIFs. That passage of the letter expressly referred (and only referred) to UK MIFs and the UK market (“UK MIF” was mentioned six times, and no other MIF was mentioned). When it came to quantum, Luxottica was said to have paid “UK interchange fees” on transactions totalling around £52m, resulting in an alleged overcharge of £321,548, and to have suffered estimated overcharge through concessions at four department stores totalling £428,553. (One of those stores was, in fact, in Jersey and therefore not technically in the UK, as Mr Piccinin pointed out to Mr Melani. But the sum involved was tiny.) The letter said that “our Clients’ claims are for UK credit card transactions only”. It asked Visa to extend the time for service of the claim form while the parties awaited decisions on permission to appeal for the CAT’s Sainsbury’s judgment and Popplewell J’s judgment in Morrisons, and for Phillips J’s judgment to be delivered.[36]It was, therefore, clear to Visa that the claim being made was by a UK company for UK MIFs. Fieldfisher’s retainer was not limited to making such a claim—but that was the claim advanced, and I accept the evidence of Mr Pimlott, Mr Bacon, and Mr Melani that that is the only claim that Luxottica ever considered making. Whether the wider group might have wider claims was never considered.[37]So Visa knew (and this never changed) that Luxottica was only advancing claims based on UK MIFs. On the other hand, Visa knew that the claim form—which they had seen—could in theory have accommodated other claims (for instance in relation to EEA MIFs). It would not, however, have been wide enough to cover claims by any other Luxottica group company.[38]The claim form was never served. Its validity was successively extended and, as I describe below, the claim was ultimately settled before service.[39]The facts I have set out above (except the scope of Fieldfisher’s retainer as internally understood, and the extent to which Luxottica had considered other possible claims) were available to both parties, and form part of the objective factual context for construing the settlement agreement. Although Luxottica’s employees seem not to have been aware of the possibility of claiming for MIFs paid anywhere other than the UK, the reasonable objective observer would have been aware that it was a possibility. On the other hand, it is a matter of objective background knowledge available to both parties that no such claim had ever been advanced.

Visa’s settlement strategy

[40]Visa’s strategic approach to the litigation included a practice of settling claims when it was opportune to do so. Linklaters developed a standard form of settlement agreement, and some standard form letters which—presumably suitably adapted—could be used to make or accept offers.[41]One aspect of that settlement strategy involved a concern that settlements should not be undermined or exploited to Visa’s disadvantage. The evidence expressed this concern in various ways.[42]Visa wanted (as it was put by Mr Hugh Stokes) to cover claims by associated companies “to prevent the claimant(s) from circumventing the terms of the settlement agreement by bringing a similar claim via a related corporate entity” (my emphasis). As Ms Holtz put it in her statement, this was to prevent claimants from seeking “to undo the settlement by bringing effectively the same claim through another means, e.g. through another entity within their corporate group” (my emphasis). Mr Pimlott and Mr Melani gave evidence in which they both essentially accepted that they would have understood this to be an objective.[43]Some of the witnesses—including Ms Holtz in her oral evidence, Ms Williams in her witness statement and oral evidence, and Mr Winfield-Chislett—put the intention much wider. Mr Winfield-Chislett said that Visa’s purpose was to obtain the “maximum degree of comprehensiveness”, to “maximally remove … risk” and to do so by settling “any possible Interchange or Interchange-like claims”. Ms Williams explained that there was “no point in Visa expending significant time, money and energy settling with one entity in a corporate group if further claims swiftly pop up from other entities in the corporate group who know the amount that Visa would be willing to settle for”. It became a theme of this evidence that Visa’s commercial requirement was to know not only that the same claims would not resurface from elsewhere in the corporate group, or even that similar claims would not resurface—but that no claims relating to interchange fees—past, present, or future—could ever again be made by any company within a corporate group one member of which had settled a MIF claim, or from any company that ever had been part of that group, or from any company that ever became part of that group.[44]This broader (or deeper) strategy was sometimes equated with the “no undermining” strategy, but I do not accept that they are the same, or that one necessarily implies the other. The first is reasonably common, though blurry around the edges. When a defendant settles a particular claim, there are a variety of ways in which that settlement may be undermined quite directly (for instance by presenting essentially the same complaint in a new legal form, or by discovering new losses not previously claimed). A defendant will usually—in any litigation, especially when it is settled at an early stage—want to go beyond settling just a particular quantified claim arising out of a particular specified cause of action, so that the settlement reaches any sort of claim based on the events that have given rise to the dispute. How much further it will be necessary or possible to go will depend on circumstances. But that does not usually require anything as broad in its sweep as the “maximal protection” strategy, at least in the form that it was described.[45]The contemporaneous evidence does not, I think, demonstrate that the “maximal protection” strategy, with a cornerstone being complete and perpetual inter-group settlement, was central to Visa’s thinking at the time.[46]First, Mr Coleman submitted, if a core purpose of a settlement was to obtain group-wide permanent peace, Linklaters and Visa would have wanted to ensure that all group companies were party to the settlement agreements, or at the very least that parent companies were. As a practical matter, I think that is correct. Ms Williams was right that this was not legally essential, because in theory Visa could rely on the ability to enforce the subsidiary’s promise to ensure that no other group company sued, and indemnify it if it failed; and in theory it could rely on a warranty of authority. But any lawyer would see risks in that course if group-wide settlement was key. Most obviously, it would leave Visa with recourse only to the assets of a subsidiary, which might become insolvent, or be sold outside the group. If an effective and lasting group-wide settlement was a high priority, I would have expected Visa to want to deal with more than merely one subsidiary in a large group.[47]Secondly, if this had been Visa’s central objective, it would have been a high priority for Visa to identify the present and past members of the group for any settling claimant. If nothing else, it would have wanted to make sure that it had not previously settled with another group company, and to assess the value of the settlement to Visa by understanding the claims the group actually or potentially had. But Visa’s evidence is that there was no system for doing that. It was not done for Luxottica, and Linklaters did not work out whether a Luxottica US company was making claims about MIFs in the US. Nor—although this is of less significance—did Ms Williams react when she read that GrandVision was likely to be acquired by Luxottica. I accept her evidence that she did not. But the fact that she did not is telling: for if “group-wide” settlement had been the watchword, this would have been important intelligence, not uninteresting background.[48]Thirdly, if that had been a high priority, one would have expected Visa to specifically identify extant claims that were being settled by any agreement. It would rationally have wanted to assess the value of any settlement economically. It would also have wanted to do so to ensure that procedural steps were taken to bring all settled suits to an end, and stop running up costs. And it would probably have wanted to do so to identify known settled claims specifically in any settlement agreement, or at least to consider doing that. Visa’s witnesses emphasised the importance of clarity in a settlement. Although, inevitably, a settlement which covers unknown claims cannot identify those except in abstract terms, known claims can be identified, and clarity suggests that they should be.[49]My impression is that the “maximal protection” strategy has been played up in the witnesses’ minds because of the issues in this case. More specifically, I think that it has been given a concrete content (as if one of Visa’s central goals was to deal with claims just like the GrandVision claim) which I doubt it had. I accept that Visa (in general terms) wanted to obtain broad protection, beyond the immediate presenting claim, and extending at least some way into corporate groups. So much is indeed clear from the terms of the settlement agreement. But that is not quite the same thing as saying that it was specifically seeking protection from claims such as GrandVision’s, or that that was a prime and essential element of any settlement.[50]In any event, in my view, objective reasonable observers would not have assumed such a strategy. They would, however, have expected Visa to want to go at least somewhat beyond merely settling the claim as presented. They would have appreciated that Visa was likely to seek at least protection to cover similar claims, to prevent the settlement being undermined, and to make it effective in the context of mass tort litigation. But they would have understood the question “How much further?” to be a matter for negotiation, and would not have approached the agreement with a fixed preconception about how far it would go.[51]Luxottica submitted that Visa hoped that its settlements would be read as widely as possible, but that it did not think they would be. The evidence does not enable me to say that. Visa’s witnesses denied it; but since Visa did not waive privilege, their recollection is untestable. I do not think it matters. The context here was negotiation between well-represented commercial parties, far removed from any field where consumer-protection policy applies. It is not underhand for a party or a lawyer, in that context, to put forward terms with the hope of achieving a result, even with the knowledge that it may not be fully effective. Luxottica denied that it was suggesting that it would be, but suggested that this might have contributed to a failure to address any mistake that Luxottica made, and therefore have been relevant to its case on sharp practice. I do not see that this follows: if this would have been acceptable practice (as it was) it does not strengthen Luxottica’s case that Visa behaved in an underhand way.

This settlement

[52]Visa made its first offer to settle Luxottica’s High Court claim on 1 August 2017, in a letter from Linklaters. The offer was for £100,000, and for “full and final settlement of the Claim”. There was no reference in that letter to any other claims. It was met and countered by an offer to accept £300,000 on 15 August 2017. That was rejected, three days later, with Visa increasing its offer to £125,000 (still, at this stage, just for “the Claim”). The offer was not accepted.[53]On 22 January 2018, Linklaters renewed Visa’s £125,000 offer. This time the letter (following what was now Visa’s standard form) contained the following statement (emphasis added):
“The Offer is made subject to contract and is premised on the full and final settlement of the Claim and all similar claims by the claimant and/or other entities within its corporate group. If your client indicates an intention to accept the Offer, we shall provide a draft settlement agreement.”
[54]A week or so later, Fieldfisher responded with an offer of £200,000, saying that it was “presently the lowest Luxottica is prepared to accept”. That letter was not subject to contract, and was expressed as 65 percent of Luxottica’s “current estimated losses”. It was not accepted. But it led Visa—through Linklaters—to make an increased offer of £150,000 on 31 January 2018. That letter contained the language set out above. The counteroffer in turn was rejected, with Fieldfisher repeating the £200,000 offer.[55]There was then a telephone conversation between Mr Casells and Mr Pimlott, in which Mr Cassells increased Visa’s offer to £175,000. That was rejected by email on 21 March 2018, with Luxottica holding out for £200,000, which was in turn rejected. So there matters rested, with the parties expressly £25,000 apart on a “number” (there is no evidence that anything else was talked about). That remained the position in May, at which point Visa left its £175,000 offer on the table.[56]In September 2018, following the Court of Appeal’s judgment, Fieldfisher sent a letter to Visa increasing its settlement demand. Its claim had increased (because it was now based on a competitive interchange fee of zero), so that its total claim was said to be £1.8 million, and it was prepared to accept approximately £1.29 million in settlement. In due course the message came from Linklaters that, although Visa was willing to discuss settlement it was not willing to engage at those levels. There matters rested during 2019, while the Supreme Court appeal ran its course.[57]Mr Allen, who had up to that point been dealing with the claims at Luxottica, left the company in around September 2019. When Fieldfisher discovered this, in early 2020, they contacted Mr Melani and arranged to speak to him. He told Mr Pimlott that he was aware of the claim, and had been briefed by Mr Allen. In May 2020, Mr Bacon identified that he was “picking up” that claim.[58]Meanwhile, on 9 March 2020, Visa had made a further offer. This was in the period after the Supreme Court hearing had taken place, but before its judgment was given. Visa increased its offer to £200,000. The letter contained the comment about being based on the settlement of “similar claims” that I have set out above. Fieldfisher forwarded the email to Mr Melani, but seem not to have received instructions from him.[59]The Supreme Court’s judgment was given on 17 June 2020. Mr Pimlott emailed Mr Bacon to tell him that this was “good news” for the prospects of the claim. They gave more detailed advice a few days later. At this juncture, Fieldfisher was trying to put together some sort of funding arrangement for their various clients who had MIF claims. This led to internal discussions about Luxottica’s plans and appetite for the risk, management time, and cost of the litigation. By late September, it had become clear that Fieldfisher would not be able to arrange costs funding. On 14 October 2020, Luxottica decided to make a revised settlement offer. But—as Mr Bacon put it in an email of that date, “If they don’t accept then we have concluded internally that its [sic] not worth pursuing any longer”.[60]The offer was made on 16 October 2020. It was for £200,000 (i.e., at the level that Visa had itself been offering immediately before the Supreme Court judgment). It offered no detailed justification, beyond a reference to the “heavy evidential burden” of claimants in relation to the evidence of pass-on to their customers, to be reflected—alongside the “benefits of early settlement”—in a substantially discounted settlement sum.[61]On 23 October 2020, in a telephone conversation with Ms Dodds of Fieldfisher, Mr Cassels informally indicated that Visa intended to accept that offer. That was formally confirmed in a letter of 26 October 2020, which stated:
“Following the WP Call, we write to formally confirm that Visa is prepared, on a purely commercial basis, to accept your client’s offer of £200,000 (inclusive of interest and costs) in full and final settlement of the Claim. This acceptance remains subject to contract and is premised on the full and final settlement of the Claim and all similar claims by your client and/or other entities within its corporate group. We enclose a draft settlement agreement. We look forward to hearing from you.”
We enclose a draft settlement agreement. We look forward to hearing from you.”[62]The enclosed draft settlement agreement was—subject to one point that I discuss below—in substantially the same form as the version that was in due course signed. Fieldfisher forwarded it to Mr Bacon. He checked the sum, but did not otherwise study it.[63]The point on which Fieldfisher pushed back concerned the breadth with which “MIF-Related Claim” was defined. Fieldfisher already had experience of Visa’s standard-form of settlement agreement, which had been used for another client’s case. This enabled them to identify a risk that it might catch Luxottica’s claims against Mastercard. They discussed internally how best to avoid this, and settled on the introduction of words (“solely in their capacities as members or customer financial institutions of Visa and solely with respect to their Visa volume”) which would avoid that overreach, together with one other minor amendment. They sent a draft to Linklaters on 2 December 2020. The amendments were discussed in a telephone call between solicitors two days later. After various chasers, Linklaters confirmed Visa’s agreement to the amendments on 12 January 2021, with what were described (accurately) as “minor drafting tweaks”.[64]Fieldfisher took those tweaks (which were all to the definition of MIF-Related Claim, and all related to the carve-out of non-Visa related MIF claims) back to Mr Bacon and Mr Melani on 12 January, advising that they were “acceptable” and asking for a signed version to be returned. The following day, Mr Bacon returned the executed document, signed by Mr Melani. There was some debate at trial about how far Mr Melani’s authority went. Luxottica pleaded that Luxottica did not have apparent or usual authority to agree to settle claims for third parties, and that Mr Melani did not have apparent or usual authority to agree to obligations for Luxottica such as ensuring that related companies would not commence or continue claims. Those points are relied on, however, not to say that the settlement agreement was invalid (the agreement is admitted), but as part of the factual matrix about how it is to be construed. I do not think, however, that they are useful for that purpose. Luxottica expressly warranted that it had authority to do what the agreement did; Mr Melani impliedly warranted that he had authority to sign for Luxottica; it was a matter for both to satisfy themselves that they had such authority. To start with the scope of Luxottica’s (and Mr Melani’s) usual authority and then construe an admittedly valid agreement so that it contains only terms within the “usual” authority would be to work back to front. In any case, although I see the force of the point about Luxottica’s authority, I do not agree with the point about Mr Melani’s. I see no reason why a director would not usually have authority to agree guarantee and indemnity provisions on behalf of his company, even if they relate to third parties. The more general point—that on Visa’s case the settlement agreement imposed remarkably broad obligations—remains a valid consideration; but I do not find the analysis is assisted by looking at it through the lens of usual or apparent authority.[65]On 20 January 2021, Linklaters told Fieldfisher that they had Visa’s signed counterpart; Fieldfisher agreed to release Luxottica’s signed version. The settlement was then complete.

The US claim

[66]I have already described the outline of Luxottica US’s claim at paragraph [21] above. I now turn to the evidence about what Luxottica (i.e. Luxottica UK) knew about it, and what Visa (and Linklaters) knew about it.[67]The claim remained pending throughout the period with which I was concerned. The plaintiff was at all material times a member of Luxottica’s corporate group, and an associated company of Luxottica for the purposes of section 256 of the Companies Act 2006. Although I have no evidence about the merits of the case, I would infer that it was a claim that had substantial potential value—an inference that I draw partly from the absence of any evidence to the contrary, partly from the obvious fact that the US market was a large market and Luxottica’s US entity was likely to have had substantial card-payment volume, and partly from the confidential information I have about the amount paid to settle it later. I am not in a position to give even a range of values: but by “substantial” I mean that I infer that it is a claim that Visa could rationally have expected to pay “real money” to settle, and the plaintiff would rationally have expected to receive “real money” to settle—not simply a minimal or nuisance-value payment.[68]It must be the case that, at all material times, Visa itself (through Visa Inc, which was a defendant in the US litigation) knew all those facts. As to Linklaters, Ms Williams accepts that she was made aware of the US claim in 2019, but says (in effect) that she had forgotten about it by late 2020. I accept that is likely.[69]However, she was again alerted to it in early 2021. Visa accepts that Ms Williams, Ms Olomolaiye, and Mr Smyth (two other associates who worked on the settlement agreement) were aware of the US claim “at or before the Parties’ entry into the settlement agreement”. Ms Williams explains that on 11 January 2021 she “became aware for the first time [of] … a potential argument that the US Claim might fall within the scope of the Settlement Agreement”. At, or shortly after that point, Mr Smyth obtained corporate documents which would have enabled them to check whether the US claim did fall within the scope of the agreement. Ms Williams’ evidence was that Mr Smyth had those documents, but that he did not check them until after the agreement had been executed, and that she did not ask him to do so. She says that she, and her team, did not ascertain whether the case was “live”.[70]It is clear, then, despite the professed non-waiver of privilege, that Linklaters had(a) identified the US claim made by a “Luxottica” named company and(b) done so in the context of recognition that the settlement agreement, if executed, might apply to that claim. But Linklaters—on this evidence—neither carried out the admittedly simple checks that would have confirmed that the US plaintiff was related to Luxottica, or the status of its claim, and did not say anything to Visa about it. Nor did they raise it with Luxottica.[71]Ms Williams originally maintained that she could know nothing more than that this “might” be a claim by an associated company. That struck me as unrealistic. Although it was theoretically possible that there was an entirely unrelated retailer with the same name as Luxottica in the United States, it must have seemed likely that the companies were related, as Ms Williams ultimately accepted. She did not know whether the claim had settled. But, again, in the absence of any information suggesting that it had, I do not think that anyone in Ms Williams’ position would have assumed it had settled.[72]Ms Williams accepted that she had received information about the “value of the claim, or the alleged value of the claim” but said that it was nowhere near the figure of USD 750 million that appears to have been asserted by the plaintiff as its maximum value. I cannot begin to assess any precise figure. But I do not think Ms Williams disputed that the claim was “substantial” in the sense of being potentially worth a more than de minimis amount, and an amount which would have been material to whether £200,000 was an appropriate sum for settlement of both Luxottica’s High Court claim and the US claim. In any case, I find that she did know that. It seems to me to be common sense given(a) the fact that the US litigation, which proceeded for many years (as Ms Williams knew), was evidently not amenable to being simply disposed of summarily,(b) the inherent likelihood that Luxottica’s US subsidiaries would have been substantial indirect payers of MIFs,(c) the absence of any evidence to the contrary, and(d) what I know, confidentially, about its subsequent settlement.[73]Why then did she do nothing about it? There seem to be two possibilities.[74]The first possibility is that Ms Williams, suspecting that the US claim was unknown to Luxottica, deliberately refrained from finding out more about it (or whether Luxottica knew about it), suspecting that the settlement agreement might be re-opened if the point was raised, and wanting to keep Luxottica in the dark. Ms Williams denied that was so. Although Luxottica sometimes came close to making the allegation, there was no pleaded case of “blind eye” knowledge, and I accept that this was not Ms Williams’ motivation.[75]The second possibility is that Ms Williams did not see the US claim as very important, especially for her, since she was not dealing with the US litigation. It might or might not be swept up in the Luxottica settlement, if that was agreed. If it was, that might be a bonus. If it was not, the Luxottica settlement was still acceptable. This possibility is consistent with Ms Williams’ evidence that she “did not conduct any contractual analysis at the time to determine whether, if the US Claim was still a live claim at the time, it would have been captured by the terms of the Settlement Agreement. In other words, I did not conduct a merits assessment of the potential argument concerning the US Claim during the relevant period.”[76]The difficulty is in squaring that evidence with(a) Ms Williams’ insistence that a key part of Visa’s settlement strategy was obtaining group-wide release and(b) her insistence that this is what she already understood the settlement agreement (which was in a standard form she was familiar with) achieved and(c) the fact that, around this time, Linklaters’ specific focus was on the words of the definition of MIF-Related Claim, which included a reference to claims “in any jurisdiction”. If that evidence is correct, then there really was no “merits assessment” of a “potential argument” required: it would have been obvious to Ms Williams that, if a live claim, the US claim would be covered. Moreover, if her evidence about the settlement strategy is correct, it would have been important for Visa to find out whether it was covered, and to make sure that it was covered.[77]This is also an area where I am especially hampered by having no access to any contemporaneous documents showing internal communications within Linklaters. Without, ostensibly, waiving privilege, Visa has made assertions about Ms Williams’ state of mind, her assessment of the US claim’s value, and indeed about communications within the Linklaters team and the reasons for them. But it has not disclosed any documents showing or evidencing those communications.[78]Ultimately, I accept Ms Williams’ evidence that she did not focus much on the US claim in part because I do not accept her current recollection about Visa’s settlement strategy. As I have already said (para [45]), I do not think that permanent group-wide peace was as central, or as clearly focused, as it now appears to Ms Williams to have been. The conclusion I draw from these events is that the central purpose of the settlement, from Visa’s point of view, was to settle the UK claim and similar UK- and EU-law based claims; further protection, while “nice to have”, was not critical. The approach was to draft broadly, address any pushback that provoked from settling defendants, and deal with the remoter consequences later.[79]So much for Visa and Linklaters. What of Luxottica? In point of fact, none of those acting for Luxottica knew about the US claim. It is true that both Mr Bacon and Mr Melani gave evidence in their witness statements that there were others (not them) “at the Defendant” (i.e. Luxottica UK) who knew about the US claim. But they did not, and nor did Fieldfisher.[80]Would the objective observer treat knowledge of the US claim as assumed common knowledge? On this occasion, the settlement agreement itself gives an answer, for it contains a statement that Luxottica is not aware of any Associated Company with a MIF-Related Claim (see paragraph [133] below). When it comes to construing the terms of the settlement agreement, the reader would assume that that statement is true, because the rest of the agreement must be drafted on the assumption that it is true. And in fact, I think, it was true. For the purposes of construction, I do not think it would be right, objectively, to assume common knowledge just because someone at Luxottica who was not handling the UK claim happened to know about the US claim. I doubt that such knowledge would, on principles of agency, be attributed to Luxottica, since the individuals who had it—not identified—seem not to have been engaged in that case or its settlement. But even if it were, Visa could not know that. The distinctive feature of “factual matrix” is that it consists not simply of things that the parties each happen to know, but things that each party knows, and each expects the other to know too. The US claim was not shared background knowledge in that sense.[81]I consider this conclusion valid despite the fact that aspects of the US claim are set out in paragraph 12 of the re-re-amended defence, under the heading (some paragraphs above) of “The factual matrix”, which has not been specifically denied in the reply. That paragraph pleads matters which plainly cannot have been shared knowledge (such as that the claim settled in June 2025), and combines a factual plea with a variety of submissions about the meaning of the settlement agreement. But in any event, it pleads in terms that the US Claim was “a MIF-Related Claim within the meaning of the Settlement Agreement” and that it was made by “Associated Companies” of Luxottica. If that is so, then given the terms of clause 6.3, there seems to me to be no escape from the conclusion that the assumption to be made, for the purposes of interpreting the settlement agreement, is that Luxottica was not aware of it.

GrandVision’s claim

[82]GrandVision NV and 46 related companies issued a claim form in the High Court on 4 June 2018. The claim alleged breach of EU, EEA, UK, and domestic competition law in relation to fees (not, in the claim form, expressly limited to MIFs). Particulars of claim, served in September 2020, showed that the claim was for damages alleged to result from the way Visa had set MIFs, and went beyond UK MIFs. The claim is said to be for more than £10 million.

The acquisition of GrandVision

[83]When GrandVision’s claim was first made, GrandVision—another specialist optical group—was not in any way related to Luxottica. On 31 July 2019, it was announced that Luxottica’s controlling shareholder, EssilorLuxottica SA, had reached an agreement to buy a controlling interest in GrandVision, and that it expected to acquire the remaining shares by mandatory public cash offer. The acquisition was, however, subject to merger clearance, and was not expected to close until 1–2 years later. In fact, it completed on 1 July 2021.[84]It is therefore common ground that GrandVision and its subsidiaries were not, at the time of the settlement agreement, in any way related to Luxottica, but would become so if and when the acquisition took place.[85]Mr Melani knew that GrandVision was, if things panned out, to be acquired by Luxottica’s ultimate shareholder. He did not, however, know about GrandVision’s competition law claim.[86]Visa did know about GrandVision’s competition law claim. It was suggested that Visa must have known about the proposed acquisition, including because it was reported in a newsletters which Ms Holtz and others received, commenting on various aspects of EU competition policy, and in an MLex news update circulated to Mr Cassels and Ms Williams. She read it, but did not remember making much of it. I find that evidence convincing. The witnesses seemed to think the newsletters provided useful gossip about comings and goings at the European Commission, and the MLex updates could be of background interest. But their news of transactions subject to merger control is not the sort of thing anyone would be likely to remember, and (as Ms Williams said) when she was sent the MLex news update in December 2019 there were no active settlement discussions with Luxottica. Nor does the fact that a different team from Linklaters was advising GrandVision mean that any relevant person had actual knowledge.[87]I do not, therefore, think there is anything in the GrandVision claim that could be regarded as mutually shared factual context. Visa held one piece of the jigsaw puzzle (that GrandVision was making a claim). Luxottica held another (that GrandVision might become a related company). Neither was in a position to join the two, or expect the other to join the two. And unless joined they had no relevance to the construction of the settlement agreement. The objective reader of the settlement would have known—as a general matter—that retailers often did have MIF claims, and that Luxottica was a large corporate group that sometimes acquired or was acquired by other retailers. Those facts together made it objectively plausible that at some future point a company with an existing MIF claim might come into the group; but that was, objectively, an abstract possibility rather than an imminent probability.[88]At any rate, it is common ground that it was never discussed between the parties during the negotiation of the settlement. There is no evidence that it was considered internally by either. It was only after the acquisition completed that Visa raised the contention that the settlement agreement affected GrandVision’s claim, which led ultimately to this case.

Subjective beliefs

[89]I turn, finally, to the parties’ subjective beliefs about the scope of the settlement agreement as far as claims other than Luxottica’s own High Court claim is concerned. This makes no difference to construction. It may be relevant to the sharp practice defence. It is an area where the particular issue now raised has, I think, significantly shaped (and distorted) the witnesses’ current views.[90]Mr Bacon, Mr Melani, and Mr Pimlott did not give any thought to whether the agreement would cover the US claim, or the GrandVision claim, or any similar sort of claim. That being so, their present view about what they would have thought if they had considered it is bound to be hard to disentangle from their present desire that it should not be. All of them apart from Mr Bacon realised (as is obvious from the settlement agreement, even on a cursory reading, and from the covering letter) that it went beyond merely settling the Luxottica UK claim. (Mr Bacon did not study the agreement, and can therefore have had no idea what it meant.) None of them gave any detailed consideration to how much further. That is, in Mr Pimlott’s case, hardly surprising since he had not been told that there were any other claims in Luxottica’s corporate group, or any in the offing. And, in Mr Melani’s case, it is not surprising because he understood MIF claims to be a UK-specific kind of claim, and had in any event already decided that Luxottica should take whatever it could get. I am far from satisfied that if he had thought about it at the time, he would have been able to conclude—based on a 20-minute read-through of the agreement—that it covered only the very transactions that had been the subject of the High Court action. But apart from the detailed thought that Fieldfisher gave to ensuring the Visa settlement did not inadvertently apply to the Mastercard claims, Fieldfisher does not seem to have given an overall assessment of the draft settlement agreement’s effects, and Luxottica did not ask for one.[91]I accept, therefore, that Luxottica’s representatives had no positive belief that the settlement would cover the US claim or a claim like the GrandVision claim. But that was not because, having thought about that, they had decided it did not. It was simply that they had not thought about it. It was not something looked at and misunderstood, but something overlooked, even as a possibility.[92]As to Visa, I have already given my reasons for believing that Visa’s representatives and lawyers were less fixated on the breadth of the settlement than they are now inclined to think they were. There is no evidence that any specific consideration was given to whether it would apply in GrandVision’s case, and their own evidence is that Linklaters had raised merely as a possibility that it might cover the US claim and had not considered that point factually or as a matter of contractual analysis. There is some evidence of a previous occasion when a case like GrandVision’s had arisen, and Visa had persuaded the claimant to withdraw the claim. But without a properly detailed account of that, and with no evidence that it was seen as salient, I can make nothing much of it.[93]So, overall, my conclusion is that both sides knew that the agreement went further than the existing claims, but neither side had attempted either to precisely map the outer fringes of its effect, or its impact on specific alternative claims. To ask whether one or the other thought it would apply to either the US claim or the GrandVision claim is to ask, as if it were a matter of historical fact, a hypothetical question. To the extent that the point had been considered at all, it was only by Linklaters, whose answer (for the US claim only) was “maybe”.

Interpretation: Principles

[94]The principles governing contractual interpretation are not in dispute. They were recently summarised by Lord Burrows JSC in Providence Building Services Ltd v Hexagon Housing Association Ltd [2026] UKSC 1, [2026] 1 WLR 538 at [21]–[23] as follows (some citations omitted):
“[21] The aim of contractual interpretation is to ascertain ‘the meaning which [the contract] would convey to a reasonable person having all the [relevant] background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract.’ … business (or commercial) common sense may be relevant. In contrast, declarations of the subjective intentions of the parties and, for reasons of practical policy, previous negotiations cannot be used in determining what the contractual language means. [22] In Arnold v Britton [2015] UKSC 36; [2015] AC 1619, the Supreme Court clarified that the words used by the parties are of primary importance so that one must be careful to avoid placing too much weight on business common sense (or purpose) at the expense of the words used; and one must be astute not to rewrite the contract so as to protect one of the parties from having entered into a bad bargain. [23] In Wood v Capita Insurance Services Ltd [2017] UKSC 24; [2017] AC 1173, Lord Hodge, with whom the other Supreme Court Justices agreed, pointed out … that contractual interpretation ‘involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its contractual consequences are investigated’.”

The importance of context

[95]Meaning is always contextual. One does not start by asking what words mean divorced from context and then ask whether the context changes them. This point has been repeatedly made for more than fifty years. In Prenn v Simmonds [1971] 1 WLR 1381 (HL) at 1383, Lord Wilberforce said that:
“In order for the agreement … to be understood, it must be placed in its context. The time has long passed when agreements … were isolated from the matrix of fact in which they were set and interpreted purely on internal linguistic considerations.”
In Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 (HL), Lord Hoffmann said, at 913:
“The meaning which a document (or any other utterance) would convey to a reasonable man is not the same thing as the meaning of its words. The meaning of words is a matter of dictionaries and grammars; the meaning of the document is what the parties using those words against the relevant background would reasonably have been understood to mean. The background may not merely enable the reasonable man to choose between the possible meanings of words which are ambiguous but even (as occasionally happens in ordinary life) to conclude that the parties must, for whatever reason, have used the wrong words or syntax.”
In Bank of Credit and Commerce International SA v Ali [2001] UKHL 8, [2002] 1 AC 251, Lord Nicholls said, at [26]:
“The meaning to be given to the words used in a contract is the meaning which ought reasonably to be ascribed to those words having due regard to the purpose of the contract and the circumstances in which the contract is made. This general principle is as much applicable to a general release as to any other contractual term. Why ever should it not be?”
In R (Westminster City Council) v National Asylum Support Service [2002] 1 WLR 2956, Lord Steyn said (at [5]): “The starting point is that language in all legal texts conveys meaning according to the circumstances in which it was used. It follows that the context must always be identified and considered before the process of construction or during it. It is therefore wrong to say that the court may only resort to evidence of the contextual scene when an ambiguity has arisen”.[96]Examples could be multiplied. In this, “legal” interpretation does not depart from the ordinary way we understand utterances, but follows it. Wittgenstein gave a famous example (Philosophical Investigations §70):
“Someone says to me: ‘Shew the children a game.’ I teach them gaming with dice, and the other says ‘I didn’t mean that sort of game.’ Must the exclusion of the game with dice have come before his mind when he gave me the order?”
[97]Context was central to the House of Lords’ decision in Bank of Credit and Commerce International SA v Ali [2001] UKHL 8, [2002] 1 AC 251. An employee, on accepting an enhanced redundancy settlement, had given a release. The words used were extremely wide:
“all or any claims whether under statute, common law or in equity of whatsoever nature that exist or may exist …”
. All members of the House of Lords agreed that “all or any … of whatsoever nature” could not be read without some implied qualification—for instance they could not have applied to a claim for money owed as part of a banking relationship. All agreed that the relevant qualification was to be found in the context (the settlement of potential claims by an employee who was being made redundant). They differed only about the ambit of the qualification that the context entailed. Some (see Lord Bingham at [10]–[17]) found the relevant context by asking the question “what claims would a person think he might well be settling in this context”, and considering (see [17]) what the settling party “knew or could know that he had” to give up. For Lord Hoffmann, the relevant context consisted of the relationship out of which the settlement arose (employment).[98]One way of describing this is to say that something is implied: the reference to “any claims” implied “any claims relating to my employment” or “any of the sort of claims a competent person might expect that relate to my employment”. Mr Piccinin said that this cannot be so: the implication of terms is not a matter of interpretation, but separate from it. I agree with Mr Coleman that this presents too categorical a distinction. The question whether a term is implied is different from the question what an express term means. But it is both harmless and natural to speak of some meaning being implied. That says only that the words mean something different, in the context that they are used and because of that context, than they might mean in another context. I see nothing heretical in saying that, in BCCI, the widely worded release was impliedly limited to certain categories of claim, without jumping through the hoops of implying a term. Indeed, such language was used by the members of the House of Lords in BCCI: see Lord Bingham at [18] (“subject to some implied limitations”), Lord Hoffmann at [38] (“limitations in scope to be inferred from the background”).[99]So context is always important. That context must include all that the reasonable observer would believe that both the parties would each know, but avoiding “idiosyncratic speculation” (Jak Property Jersey Ltd v Together Commercial Finance Ltd [2025] EWHC 2442 (Ch) at [34] (Hildyard J)).[100]On the other hand, as the Supreme Court stressed in Arnold v Britton, the court’s task is to determine the meaning of the words in context. When people frame rules (or define contractual terms) by reference to generalisations, they usually have an objective (to use an example from Frederick Schauer’s Playing by the Rules (OUP 1991), “no dogs in the restaurant” to prevent customers being disturbed). The objective may help decide whether a penumbral case falls inside or outside the rule, or whether there are implied exceptions to its linguistically permissible breadth. But we cannot simply treat the rule as if it was a statement of an objective (“no disturbances”): we must reckon with the fact that a screaming human baby—however disturbing—is not a “dog”; and that a police dog—however well-trained—is. Some degree of over- and under-inclusiveness is inherent in choosing to use abstract generalised categories. But fidelity to language serves a purpose in commercial contracts, promoting predictability and certainty. If we are allowed to make words mean anything—to look through the actual language the parties have chosen to employ to the known or presumed objectives—we undermine that important goal.[101]Take the objective of catching, in a settlement agreement, “settlement-undermining” claims. The parties could simply say “… and any claim that would undermine the settlement”. They would be foregoing any attempt to specify what claims count as doing that, leaving that to be done by an open-textured term. That is one choice. But they may prefer to define what claims are to be prohibited more categorically. If they do that, they may end up with a definition which includes some claims that are not very similar, or excludes some that are; but they will avoid some arguments. This choice (open-textured statement of objective vs. defined categorical generalisation) is often faced by those drafting contracts. Each technique has its advantages and disadvantages. Skilled drafters choose deliberately. So, if categorical definition has been chosen, the court should not simply treat it as a roundabout way of expressing an objective and expand or contract it accordingly. The decision to define a category, rather than simply express an objective, should be respected. To return to the example of dogs being prohibited from a restaurant, our assumptions about the purpose and social context might lead us to conclude that, by implication, guide dogs are not within its true meaning. But we should be reluctant to entertain arguments that it does not apply to spaniels, or that it does apply to colicky infants or pet hamsters. Those conclusions would look through the words, not at them.[102]There are other well-known problems with making presumed objectives do too much work, compared to chosen language. Commercial contracts have multiple “authors” (we are concerned with what the parties are to be taken to have intended, not what one party is taken to have intended). Their objectives are not always aligned, and sometimes opposed. So the conclusion that a particular interpretation works against one party’s best interest does not take us very far. A commercial contract will often be a compromise between opposing interests: see The Rio Assu (No 2) [1999] 1 Lloyd’s Rep 115, 124 (Waller LJ). We also have to reckon with the fact that people sometimes make bad deals, but the law enforces the bad deals they make. If I agree to buy something for £1,000 which is only worth £800, it is no good arguing that “£1,000 was just a fancy way of saying ‘whatever this thing is worth’ so the price is actually £800”.[103]There are further reasons to tread cautiously around perceptions of “commercial purpose” and “commercial effect”. A judge’s assumptions about commercial purpose—often shaped in hindsight, by the facts of the particular case—may not be reliable. Commercial parties may take risks that surprise lawyers. Moreover, “commercial consequence” and “commercial purpose” often cannot be reliably ascertained apart from the parties’ chosen words (which is why the process must be “iterative”). In addition, there are very few contracts which operate perfectly, and it can be hard to know whether a particular imperfection is part of a tolerable overall package, or so serious that it should cause us to revise our assessment of meaning.[104]The final feature of contractual interpretation is that we constrain, sometimes artificially, the range of material that we permit the reasonable observer to consider. Some of these constraints are inherent in the viewpoint that the law chooses to adopt. We do not consider what either party thought the contract meant, because that is simply not data that bears on the question to be decided. But a few of them are not. English law insists that post-contractual conduct cannot be used as a guide to interpretation: it imagines the reasonable observer basing their assessment on the context as it existed at the moment the contract was made. A second limitation is that where the result is a self-contained written agreement, what the parties said during negotiations leading to it cannot be used as part of the context.[105]I should say a little more about this exclusionary rule, and its reach, because it was the subject of debate. The essential rule is clear. It was stated by Lord Hoffmann in ICS, and revisited and confirmed in Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] AC 1101. But how far does it go? In Oceanbulk Shipping & Trading SA v TMT Asia Ltd [2010] UKSC 44, [2011] 1 AC 662, the Supreme Court held that without prejudice communications (by definition, statements made during negotiations) might be admitted in evidence to prove relevant context. I think that the ratio and reach of that decision is best expressed in Lord Phillips PSC’s brief concurring judgment (at [48]): “When construing a contract between two parties, evidence of facts within their common knowledge is admissible where those facts have a bearing on the meaning that should be given to the words of the contract. This is so even where the knowledge of those facts is conveyed by one party to the other in the course of negotiations that are conducted ‘without prejudice’”.[106]So, if it is in question whether the parties knew something, the fact that they discussed it during negotiations may be admissible to show that it was shared knowledge. This does not mean that statements by the parties about how they intend the contract to apply are admissible. They are not. Nor are facts about the negotiations from which the court is being asked to infer how one party, or the other, or even both, intended the contract to apply (so, for instance, proposed amendments, deployed for that purpose).[107]To give an example of something clearly the wrong side of this line, it would be impermissible to take the informal description given in Linklaters’ letters (“… premised on the full and final settlement of the claim and all similar claims by the claimant and/or other entities within its corporate group”) and use that to construe the settlement agreement to mean just that and no more than that.[108]The controversial question before me was how that applies to things like “the parties discussed only Luxottica’s High Court claim”, or “the parties negotiated the price by reference to the value of that claim and nothing more”. Are these statements of subjective intent (or tantamount to them)? Or “background facts”? If pressed to decide, I would have considered that they fall on the wrong side of the line, and should not be considered. As it happens, however, the points for which the evidence was relied on all emerge from the text of the settlement agreement itself, so that it makes no practical difference.

Construction of the settlement agreement

[109]The iterative process of interpretation must start somewhere. I start with a high-level tour of the settlement agreement’s terms, focusing on their language and internal structure.[110]The parties to the settlement agreement are, on one side Luxottica (i.e., Luxottica Retail UK Ltd), and on the other side the Visa companies that are claimants in this case, and were defendants in Luxottica’s claim. No other person is expressed to be a party to the agreement. The attestation to Mr Melani’s signature is that he signed “on behalf of Luxottica Retail UK Ltd”. The legal position of various non-parties is, however, expressed to be affected by the agreement. Luxottica agrees to release claims “on behalf of each Claimant Associated Company” (clause 5.1, to which I shall return). And Luxottica agrees to release claims not only against the Visa parties but against “any member of the Visa Organisation”, and “Bank” and any “other party” in certain capacities (the definition of MIF-Related Claim, to which again I shall return). Of those various non-parties, some have third party rights (clause 18.2, which extends such a right to members of the Visa Organisation, the VE Member Representative, and Banks). Luxottica’s Associated Companies do not have such a right.[111]Having identified the parties, the settlement agreement proceeds to four recitals. The first three identify “the Claim”:
“(A) The Claimant [Luxottica] is the claimant in proceedings issued against Visa before the English High Court under Claim No HC-2017-002077 (the ‘Claim’). (B) In the Claim, the Claimant alleges that Visa have engaged in anti-competitive conduct in relation to the arrangements for setting and implementing multilateral interchange fees for certain types of Visa-Branded Card transactions (as defined below) in the UK. The Claimant further alleges that Visa are jointly and severally liable for losses allegedly suffered by the Claimant as a result of such conduct. (C) Visa deny that they are liable to the Claimant in respect of any aspect of the Claim or at all.”
[112]These three recitals place “the Claim” (that is Luxottica’s English claim, relating to MIFs applied “in the UK”) front and centre. No reference is made to any other claim, save in the last three words of recital C (“or at all”). No reference is made to anyone else’s claim.[113]The fourth recital is in the following terms:
“(D) In the interests of reaching a commercial resolution of the Claim and of conclusively resolving any and all other MIF-Related Claims (as defined below, together with the Claim, the ‘Settled Claims’) that the Claimant and/or any Claimant Associated Company have or may have, Visa have agreed to pay, and the Claimant has agreed to accept, the Settlement Payment (as defined below) in full and final settlement of the Settled Claims on the terms set out below.”
[114]This recital does two things. First, it explains the purpose of the agreement, which is two-fold: to settle “the Claim” and to resolve “any and all other MIF-Related Claims … that the Claimant and/or any Claimant Associated Company have or may have”. Secondly, in stating that purpose, it defines the Settled Claims, though not in a self-contained way, because it relies on further definitions of “MIF-Related Claim” and “Associated Company”.[115]There was some debate about whether the placement of the definition—after the word “Claims” and before “that”—is important. Luxottica suggested that “Settled Claims” means all MIF-Related Claims (everyone’s, anywhere, ever), and that the restrictive clause that follows (“that the Claimant …”) is not really part of the definition of “Settled Claims”, but an expression of the intention to settle only some Settled Claims. That is a tortured approach, advocated by Luxottica only to encourage the introduction of an unstated qualification to MIF-Related Claims. I agree that there is a limitation on what “MIF-Related Claims” are Settled Claims; but there is no need to look for any implied limitation. It is express. “Settled Claims” consist of(a) the Settled Claim and(b) any and all other MIF-Related Claims that the Claimant and/or any Claimant Associated Company have or may have.[116]The primary shared objective was to settle a specific claim, i.e. the claim of Luxottica relating to UK MIFs. That this is the primary context does not require recourse to what the parties said during the settlement discussions. It is clear on the face of the settlement agreement. A secondary shared objective was to “conclusively resolve” other claims, including at least some claims of “Claimant Associated Companies”. Readers would understand the objectives to have that order of priority, not least because Recital D follows hard on the heels of three recitals focusing attention on the Claim, and because of the order in which the objectives are presented. But they would be in no doubt that the purpose extends beyond settlement of Luxottica’s existing claim. They would also be in no doubt that, to identify how far further it extended, it would be necessary to consult other parts of the agreement, especially the definition of MIF-Related Claim (which is flagged) and the terms, which would explain how those claims were “conclusively resolved”. Although the words “as defined below” were not added to “Claimant Associated Company”, the use of capitals would suggest that this would prove to be a defined term, and it would certainly be apparent to the reader that the scope of this otherwise ambiguous phrase would require further examination.[117]At this point in reading, there would be one point of ambiguity, involving “have or may have”. “Have” looks to the concrete present. “May have” is ambiguous: it could mean “might (now) have” or it could mean “might (in the future) have”, or it could mean both of those things. Simply reading Recital D, any of those meanings is possible. But it would be clear that the intention is definitive finality (“conclusively resolve”), and that—so long as they are within the definition—it appears to be comprehensive (“any and all”).[118]The next port of call would, I think, be the definitions section, where I would start with the definition of MIF-Related Claim. It is a long definition, which in turn refers to others:
“ ‘MIF-Related Claim’ means any actual or potential claim, counterclaim, demand, action, cause of action, set-off, right, suit, arbitration, inquiry or proceeding of any nature whatsoever (other than in relation to an Acquirer Issue) in any jurisdiction, whether past, present or future, whether known or unknown at the time of this Agreement, whether or not contemplated or foreseen, whether against any member of the Visa Organisation, any Bank or any other party solely in their capacities as member or customer financial institutions of the Visa Organisation and solely with respect to their Visa-Branded Card transaction volume, concerning, or in any way relating to, directly or indirectly: (a) any MIF or other Merchant Fee applicable to any Visa-Branded Card transaction of any nature whatsoever in any jurisdiction irrespective of (i) where the merchant outlet and card issuer are located and, (ii) whether the merchant outlet and the card issuer are based in the same or different jurisdictions or other regions (for example, without limitation, any inter-regional MIF); and/or (b) any Rule, provided that such Rule either now applies, or at any time before the date of this Agreement has applied, to Visa-Branded Card transactions, or may in the future apply to Visa-Branded Card transactions and is of a similar nature and/or effect to any Rule which now applies, or at any time before the date of this agreement has applied, to Visa-Branded Card transactions”
[119]The immediate impression given by this clause is of comprehensive, indeed pleonastic, width. The thesaurus has been emptied of synonyms for “claim”. The parties have made it clear that the claim can be “actual or potential”, that it can relate to proceedings “of any nature whatsoever”, that it may arise in “any jurisdiction”, that it may be “unknown”, “uncontemplated”, or “unforeseen”, and that it may be “past, present or future”. The ambiguity I identified in paragraph [117] is therefore resolved in favour of the broadest possible reading:
“may have” means both “presently possible” and “future”
. Any conceivable limitation on the type of claim has been anticipated and rejected. So, too, the link to MIFs is broadly stated (“concerning, or in any way related to, directly or indirectly … any MIF …any Rule …”). There are only three express limitations. First, an express carve-out for “Acquirer Issues” (defined, elsewhere, as “omission, error or negligent act” by an acquirer about the amount to be deducted, i.e. errors by the acquirers in operating the account). Second, a qualification on the applicability of the definition to Banks and third parties to make it clear that the relevant claims only concern Visa cards. Third, a qualification in relation to Visa’s rules, for the future, to prevent the definition from applying to new rules unless they are “of a similar nature and/or effect” to existing ones.[120]This shows that MIF-Related Claims extend beyond “the Claim” in at least four directions. First, they include claims in other jurisdictions, and arising from MIF payments made in other jurisdictions—so beyond UK MIFS, and beyond a claim in the English courts. Second, they extend to claims against other companies within the Visa organisation, and to banks and financial institutions handling Visa-branded card payments. Third, they include not only existing claims, but claims which could arise in the future, including because of action not yet taken or contemplated (such as new rules). Fourth, the intention is to include those claims whether or not they are known, contemplated, or foreseen, so that Luxottica is taking the risk that it may be prejudiced by the settlement in ways that it has not yet identified.[121]Some of those extensions are very common: settlements often cover unknown and unforeseen claims; they often cover claims in other jurisdictions than the one which is currently the focus of attention; and they quite often cover some claims against third parties (which can drag a settling defendant back into litigation through contribution proceedings). The breadth of the extension to third parties here, however, is notably wide. The most striking—and not usual—aspect is the inclusion not merely of unknown claims but of future claims arising out of the same general subject-matter, and it is not (linguistically) totally comfortable to describe an agreement addressing future conduct as “settling” or “resolving” a claim. But the intention is clear, and a reason is also apparent: Visa wished to continue to apply MIFs, and Luxottica was agreeing to let it do so.[122]What that definition does not address, however, is the meaning of “Claimant Associated Company”: it tells us what kinds of claim are covered, but not whose claims are covered. There is, in fact, no definition of “Claimant Associated Company” as such. But there is a definition of “Associated Company”:
“ ‘Associated Company’ shall have the same meaning as in Section 256 of the Companies Act 2006, and shall include past, present or future Associated Companies (together with successors) without any geographical limitation whatsoever”. “Claimant Associated Company” would therefore be understood to mean a company which is, under section 256 of the Companies Act 2006, associated with Luxottica. Section 256 provides as follows: “For the purposes of this part—(a) bodies corporate are associated if one is a subsidiary of the other or both are subsidiaries of the same body corporate; and (b) companies are associated if one is a subsidiary of the other or both are subsidiaries of the same body corporate.”
Subsidiary is defined in section 1159. A company is a subsidiary of another company (its holding company) if the holding company holds a majority of voting rights, or is a member with power to appoint or remove a majority of its board of directors, or is a member and party to an agreement which enable it to control the majority of the voting rights. Indirect subsidiaries are subsidiaries. The practical effect of this is that, at any moment, every company within Luxottica’s group from its ultimate holding company to Luxottica’s own indirect subsidiaries, including all sibling companies, is “Associated”. “For the purposes of this part—(a) bodies corporate are associated if one is a subsidiary of the other or both are subsidiaries of the same body corporate; and (b) companies are associated if one is a subsidiary of the other or both are subsidiaries of the same body corporate.”[123]Armed with that knowledge, one may return to Recital D. The first impression is of an intention to settle Luxottica’s asserted English claim and to “conclusively resolve … any and all” of a very wide class of claim—present or future—of any company that was, is, or becomes part of Luxottica’s corporate group. However, I agree with Luxottica that the reader would find such a result at first blush remarkably wide, and might put a question-mark in the margin at that point. I shall revisit it later.[124]The next task, having seen the apparent breadth with which “Settled Claim” is defined would be to see how the definition is used. If the intention is to “conclusively resolve” such claims, what is the mechanism by which that is done?[125]The first indication is in clause 4.2, which provides that the Settlement Payment shall be paid “in full and final settlement of the Settled Claims”. That does not take one much further. But it does show that the payment is in consideration of both the Claim and of the other Settled Claims.[126]The second indication is in clause 5.1 which provides:
“The Claimant on its own behalf, and for and on behalf of each Claimant Associated Company, agrees to fully and finally release, and irrevocably waive, the Settled Claims.”
[127]This provision would give the reader pause in two respects. First, the definition of Associated Company, if read with its maximum breadth, includes all past Associated Companies and all future Associated Companies. But it seems improbable that it should be read so widely here. Consider three examples.(a) Take a company, formerly part of Luxottica’s group, but now liquidated. Clearly Luxottica could not, even theoretically, do anything “on behalf of” that non-existent entity.(b) Take a company, formerly part of Luxottica’s group, but now owned outside it. Although theoretically Luxottica could (if properly authorised) act on behalf of such a company, it would be unlikely to do so. So the parties seem to be saying that Luxottica is doing something it is unlikely to do.(c) Take future companies, which might either exist—but be outside the group—at the date of the agreement, or not yet be in existence. It is hard to see how Luxottica could possibly act on behalf of them, because they would not even be identifiable. There are, then, two possibilities: either one understands that “Claimant Associated Company” here means “present Claimant Associated Company”, or one reads this as an agreement to obtain authority to settle claims in the future. To my mind, the former is the easier and more natural reading.[128]The second point is related. The terms used (“fully and finally release, and irrevocably waive”) typically refer to existing claims, not future ones. However, in that respect at least, I think the reader might surmise—given the definition of MIF-Related Claim which is unmistakeably future facing—that this clause is also future-facing: Luxottica will waive future claims.[129]In any event, those provisions are supplemented by those of clause 7, which I need to set out almost in full:
“7.1 The Claimant shall not, and shall ensure that its Associated Companies do not, initiate, bring, pursue, commence or continue any claim for the recovery of damages or any other remedy in respect of any Settled Claim. 7.2 The Claimant shall not, and shall ensure that its Associated Companies do not, purport to sell, assign or transfer any Settled Claim to any other person. 7.3 The Claimant shall not, and shall ensure that its Associated Companies will not, give any assistance whatsoever (save as required by law or regulation) to any party who is or might be contemplating pursuing (or is pursuing or might pursue) a MIF-Related Claim in the pursuit of such a claim. For the avoidance of doubt, the fact that the Claimant and/or its Associated Companies are or have been represented by the same law firm(s) as any such party shall not amount to a breach of this Clause 7.3. … 7.5 In the event that (a) a Claimant breaches any of Clauses 7.1 to 7.4 of this Agreement or (b) any Claimant Associated Company takes any action which the Claimant is required to ensure that such Associated Company shall not take pursuant to Clauses 7.1 to 7.4 of this Agreement, then the Claimant shall be liable to indemnify, and keep indemnified, on an after-tax basis, the Visa Organisation, the Banks and the VE Member Representative in relation to any liabilities or losses which any of them may incur or suffer in connection with any MIF-Related Claim that ensues or is assisted by the Claimant or any Claimant Associated Company.”
[130]Three points would strike the reader of these clauses. First, whereas clause 5.1 uses words focused on the past and the present, these provisions are forward-looking. There is no difficulty in accommodating “future” Associated Companies and “future claims” in them, and they positively invite that reading.[131]The second point is that—again in contrast to clause 5.1—present authority on Luxottica’s behalf to make these promises is not problematic. These are obligations of result: Luxottica will “ensure” that Associated Companies do not (in the future) bring Settled Claims, or assist in bringing MIF-Related Claims. How it will do so is left for it to work out. It will indemnify Visa and others if it cannot achieve that. Luxottica needs no present authority to do any of those things, any more than a liability insurer needs authority to indemnify from the victims of its assured’s torts. These are, no doubt, extensive and potentially onerous obligations, but their difficulties are practical not legal.[132]The third point concerns clause 7.3 (on which I invited written submissions after the hearing). Luxottica and its “Associated Companies” are not to give “any assistance whatsoever” to any “party” who is making a MIF-Related Claim. “Associated Companies” here can easily include future companies; the parties agree it does. The uncapitalised word “party” must be being used to mean “person”, including “third party”, not “party to the agreement”. MIF-Related Claims include, but are not limited to, Settled Claims. Luxottica is promising, therefore, that it will not help a third party to bring a MIF-Related Claim against Visa, and that it will make sure no other company in its group does either. If so, it would be quite odd if Associated Companies were free to make such claims themselves. Leave aside any question of self-assistance (whether GrandVision, in making its claim, is “assisting” itself – a question that need not be decided, since Visa does not rely on clause 7.3 directly). Leave aside the practicalities (how likely is any group company to bring any claim without some assistance from another). Even if one overlooks those issues, it would be incongruous that Luxottica should be obliged to “ensure” that GrandVision should not assist third parties with their MIF-Related Claims, but not under any obligation to ensure that they do not make them itself. So clause 7.3 supports the idea that definition of “Settled Claim” and the obligations under clauses 7.1 and 7.5 extend to any MIF-Related Claims of any Associated Company.[133]In my view, those are the clauses in the settlement agreement which bear most directly on the question before me. The remaining points are secondary: i) By clause 6.1, Luxottica warrants that it has not “sold, assigned, transferred or otherwise disposed of any interest in the Settled Claims”. That might suggest an assumption that they are essentially its claims. But the suggestion is equivocal, for “any interest” might mean “any interest it ever had”, so that there could be Settled Claims in which Luxottica never had any interest to sell, assign, transfer, or dispose of. ii) By clause 6.3, Luxottica warrants that it is “not aware of any Claimant Associated Company that has a MIF-Related Claim”. “Claimant Associated Company” here cannot naturally refer to future Associated Companies, because the warranty speaks at the date of the agreement to the position then. So this, like clause 5.1, is a case where “Associated Company” cannot be used in its widest possible meaning. Mr Coleman argued that this warranty supported his case because I should assume that the parties both knew about the US claim, and that they expected the warranty to be true, and therefore that they did not think the US claim was covered by the agreement. I do not accept this. The warranty related not to “Settled Claims” but to “MIF-Related Claims”. Whether or not the US claim was a Settled Claim, it must have been within this category (as, indeed, paragraph 12 of the re-re-amended defence asserts). One can only make sense of clause 5.1 on the assumption (true, in fact) that Luxottica was not aware of the US claim. So, as I have already said, this provision means that it would be wrong—whatever the true position—to approach interpretation of the agreement on the assumption that there were many actual or potential MIF-Related claim within the business: the agreement’s assumption is that Luxottica knows of none. iii) Clause 10 provides for the dismissal of Luxottica’s High Court claim. No provision is made for the disposal of any other claim. That is consistent with a primary focus being on Luxottica’s claim, and with the warranty that Luxottica is not aware of any other MIF-Related Claim within the group. iv) Clause 11 contains a requirement to keep the terms of the settlement confidential. There are exceptions to this which include communication “by the Claimant to directors, officers or advisers of an Associated Company strictly to the extent necessary to ensure compliance with the terms of this Agreement”. That is, it might be said, an odd restriction on the ability of a purported agent to communicate with its purported principals, and arguably gives rise to some other possible commercial difficulties, which I consider below. The exceptions also include a general ability to communicate “with the prior written consent of all parties”. v) Clause 11.4 permits “any Party” to confirm that the “Parties are no longer in dispute with each other and have settled the Claim to the satisfaction of all Parties on confidential terms”. On the face of it, however, that would not permit Visa to confirm that it had settled with any company other than Luxottica, and would not permit any of Luxottica’s Associated Companies to confirm that they had settled.[134]Although some of those detailed points suggest a primary focus on Luxottica’s UK claim, none does more than provide a hint beyond that. None is directly inconsistent with the view that clauses 7.1 and 7.5 bite on any MIF-Related Claim of any company within Luxottica’s corporate group.

A preliminary view

[135]In my view, as a “first iteration”, reading the settlement agreement in context, the objective reader with knowledge of the context would conclude that “Settled Claim” is defined, deliberately widely, to include MIF-Related Claims of companies that in the future become Associated Companies of Luxottica, and that the obligations under clauses 7.1 and 7.5 will apply to such claims. That would be for the following reasons: i) Recital D makes it clear—and clause 7.1 confirms—that the aim of the agreement went beyond settling Luxottica’s UK Claim, and that it extended to claims held by Associated Companies. It was also clear that it intended more than just a backward-looking settlement, but a “conclusive resolution”, in advance, of claims that might be made in the future. This is not part of typical settlement agreements, but unmistakeably a feature here. ii) The definition of Associated Company as “past, present or future” could not be mechanically applied wherever that term was used—because some of the contexts in which it was used were by their nature present or backward looking. But in the context of the definition of Settled Claim, a forward-looking interpretation is consistent with both the language of Recital D (“have or may have”) and the plainly expressed intention of the definition of MIF-Related Claim, which is both backward and forward-looking. iii) Of the other provisions of the agreement, some exhibit some tension or discomfort with this reading—but really only to the extent of showing that it was not the only or main purpose of the agreement to deal with those claims. None is flatly inconsistent with it. And one (clause 7.3) is hard to square, purposively, with any other reading.[136]Anyone could see that Visa had good reasons to extend the settlement beyond Luxottica’s existing claims, including to ensure that its MIFs—applied as a continuing practice—would never again become a bone of contention. There were also rational reasons for Visa to want to extend it to other companies in Luxottica’s group, who might otherwise exploit the information from the settlement. In saying this, I am not saying that the objective reader of the agreement would have thought it essential for the settlement to reach as far as it seemed to. But that is not the question. The objective reader would have understood that this would be a matter for negotiation, would understand why Visa might draft protectively in a settlement which it intended to achieve long-term strategic benefit. The objective reader would see from the terms of the agreement that that is how it had been drafted. On those terms, it was on a first reading the most straightforward and compelling interpretation.

Testing that view

[137]That view, however, must be tested, because there is no doubt the settlement agreement, so construed, has a wide reach. I turn, then, to the arguments which Luxottica advanced to say that the reader’s first impression should be revised by a deeper consideration of context and consequence.[138]Luxottica suggests that the apparent breadth of “Settled Claims”, and therefore the reach of clause 7.1, is cut down in one or more of three ways. The first proposed limitation is that they should exclude “claims of Associated Companies that are unrelated to Luxottica UK’s business”. The second proposed limitation is that they should exclude “MIF-Related claims of companies that are not, and have not previously been, Associated Companies of Luxottica as at the date of the Settlement Agreement”. The third proposed limitation is that they should exclude “the MIF-Related Claims that the parties are to be taken to know at the time of the Settlement Agreement were pending in the courts”.[139]Each proposed restriction depends on slightly different arguments, but they have two things in common. First, what Luxottica submits is extravagance, amounting to commercial absurdity, unless some appropriate limitation is impliedly present. Second, the need for the settlement agreement’s terms, however wide their language, to be anchored to the primary objective, of settling Luxottica’s High Court claim.[140]Luxottica is right to say that the agreement, without qualification, may operate very beneficially to Visa. One would expect anyone to think hard before agreeing to indemnify Visa against liability (however and whenever arising) to any company that ever was or ever becomes part of a large corporate family. How—even theoretically—could Luxottica know what those claims might be? To take an example canvassed in argument, if the Luxottica group happened to be acquired by another major retailer (say, a large online retailer) which had made extensive historical MIF claims on the day before judgment on those claims was delivered, Visa’s construction would require Luxottica to indemnify it. All this for a mere £200,000! In other words, the future value to Visa (and cost to Luxottica) of this agreement would be hard to quantify, and therefore hard to “price”, since it might well depend on circumstances entirely outside Luxottica’s control or contemplation. Is this not so highly improbable—so palpably uncommercial—a result that one should struggle hard to avoid it?[141]Mr Piccinin suggested Luxottica could sort out such problems in due diligence during acquisitions, or perhaps by structuring acquisitions or sales so that relevant claims never entered the group. There are, however, difficulties with that submission. The confidentiality terms of the agreement would have prevented Luxottica from even warning its parent company or a purchaser that it was about to enter into a transaction which would, surprisingly, have this effect, unless Visa agreed to that being done. But why should Visa agree? And how—unless it happened to be involved in the acquisition process—would Luxottica even know to ask Visa to agree?[142]Luxottica’s second overarching point is that the primary focus of the settlement agreement is Luxottica’s claim. That is the subject of three of the recitals. It is the only claim specifically mentioned. It is the only claim for whose procedural resolution the agreement provides. Luxottica is the only party to the agreement. The only dispute that the parties are at liberty to confirm has been settled is the dispute between them. True, as Luxottica accept, the agreement goes beyond just that. But to treat the settlement as, first and foremost, a treaty of perpetual peace between two corporate groups would be to get things out of proportion. It should be treated as first and foremost a settlement of Luxottica’s claims, and only peripherally a settlement that goes further. Mr Coleman developed that submission by reference to Lord Denning’s famous dictum about the need for some contractual provisions to be signposted in red. Luxottica might say that this is almost more troublesome: the red hands are there, but they point to one thing (the settlement of Luxottica’s claim) when, buried in the definitions and the detail, is something quite different. Mr Coleman emphasises the many cases, referred to in BCCI, which have encouraged the construction of wide releases by reference to the claims that were most salient when the releases were agreed.[143]These points have sufficient force to warrant a hard second look at the agreement. But I do not think that they ultimately justify the limitations for which Luxottica advocates.

Is the settlement limited to Luxottica UK’s business?

[144]This submission, in my view, founders against several linguistic and contextual rocks. The most striking of these is the conspicuous lack of any such limitation, simple as it would have been to state. It is not as if the contract does not make exceptions. It does so by carving “Acquirer Issues” from the definition of MIF-Related Claim. It does so again by carving out claims against Banks or third parties that are not “solely in their capacities” as members of the Visa Organisation and with respect to Visa-Branded Card Transactions. And it does so again when defining which Rules are relevant in identifying MIF-Related claims. It does not begin to do so when it comes to defining “Settled Claims”, simple as that would have been to achieve. The words of Recital D resist limitation with respect to the type of claim: it is “any and all” (not some) MIF Related Claims that the Claimant “and/or any” Associated Company “have or may have”.[145]I do not think that the objective reader would regard this as an accident or an obvious error of the Chartbrook sort. A limitation which excluded “MIF-Related Claims unless they are related to Luxottica UK’s business” would have been an invitation to future argument. What does it mean? Would a claim by (say) a European subsidiary which paid Luxottica a commission on UK sales be so related? If a company within the group that specialised in online sales for both Luxottica and other group companies has a claim, is that “unrelated” to Luxottica’s business? Or partly so? Does Luxottica’s “business” include those of subsidiaries? Is it to be defined by reference to subject-matter (“selling sunglasses and spectacles”), or geography (“in the UK, Jersey and Denmark”), or brand?[146]It is precisely to avoid questions like this that parties sometimes prefer a categorical definition, rather than an open-textured one. Luxottica’s proposed limitation would cut across the words that have been used, whose overwhelming impression is of width, with circumscribed and express exceptions. It would run against the grain of the drafting technique chosen. It would introduce what is—especially from Visa’s perspective—unacceptable uncertainty and unpredictability into the agreement.[147]Nor am I satisfied that commercial good sense demands it. Luxottica has—understandably—emphasised how a broad interpretation may operate unfairly to it. It has a point. But there are countervailing factors too, and if one is asking whether a proposed construction is mandated by business common sense, one must take a balanced view of how matters might appear in advance to both parties, not one dominated by hindsight and the interests of one. One must be satisfied not only that the agreement, if broadly construed, is disadvantageous to Luxottica, but that the proposed limitation is one that both parties presumably wanted, and objectively intended.[148]One aspect of the “absurdity” that Luxottica alleges is that it entails settlement of existing claims for present and past group companies, including (by a side-wind) the US claim. However, so far as the past is concerned, Luxottica could have tried to find out what MIF-related change had slipped under the cushions of its corporate sofa, or what risks the indemnity presented. It warranted that it was not aware of any other group company with a MIF-Related Claim. The objective reader would not assume that it had given that warranty without checking the position (though that, in fact, is what it had done). If one assumes that the settling party has done prudent due diligence, its decision (at least as far as the past is concerned) to commit to a fresh start may be commercially justifiable. And its settlement “on behalf of” other group companies of their claims occasions little surprise if it does not think they have any. Moreover, Visa had good reasons not to want to find (as Ms Williams put it) that as soon as it settled with one company other claims popped up from other group companies, buoyed by their sibling’s success and armed with the knowledge that Visa was willing to write a cheque. To leave it to Luxottica to announce, at some future time, what further claims had been unearthed within the group would have had obvious disadvantages for Visa.[149]In that context, the specific facts about the US claim are not informative. For reasons I have already given, one cannot, in construing the agreement, assume that both parties knew about it. As to conduct after the agreement was entered into, the fact that Visa did not, in the end, maintain that the US claim had been settled as a side-effect of the settlement agreement seems, in common sense terms, impressive. But Luxottica did not suggest I could place any relevant weight on it. Rightly, because it is inadmissible: it is post-contractual conduct, not alleged to give rise to any sort of estoppel. That Visa did not rely on the settlement agreement to dispose of the US claim does not show that it could not have done so.[150]As to the future, the commercial considerations are again less clearcut than Luxottica suggests. Any attempt to draw a line under MIFs so that, going forward, Visa could know where it stood would have been of doubtful value if it applied only to Luxottica’s business; for the group could by a little corporate imagination easily direct the flow of commerce to companies whose business would be their own. Indeed, on Luxottica’s construction it might have been enough for Luxottica to create a subsidiary and divert all its business there to enable it, once again, to challenge MIFs going forward. Its parent assuredly could. One can easily see why that might be insufficient for Visa. That the parties might have found suitable limiting words which would have captured some group-company claims but excluded others does not mean that they needed to, or that Luxottica’s proposed limitation is obvious or implicit. One can see why Visa might well prefer a comprehensive definition, leaving it for Luxottica to propose any limitations, which could have been crisply defined.[151]Luxottica relied, by analogy, on the approach taken in BCCI. I do not think one can simply read over from that case to this. Every member of the House of Lords there agreed that there is no special rule of construction of settlement agreements, and that some of the language from older cases, albeit instructive, went further than it should. In general terms, I accept, the agreement gives plenty of reasons to think that its central purpose and immediate occasion was the settlement of Luxottica’s UK claims (the choice of parties, the confidentiality provisions, the specific focus on its dismissal, the absence of provision for enforcement by any other Luxottica group company). But it also gives plenty of reasons to think that it goes wider.[152]For, if there are indications of focus on one specific claim, there is equally an incontrovertible intention to sweep more widely than just that claim or claimant. This case is different from BCCI, in four ways. The first difference is that although one may fairly say that the primary purpose is the resolution of Luxottica’s UK claim, it is beyond question that there were subsidiary purposes of conclusively resolving a category of other claims. The second difference is that whereas in BCCI the category of claim which the majority held had not been settled was unprecedented and “uncontemplated”, there was nothing remotely unprecedented or uncontemplated about either(a) other companies in Luxottica’s group having MIF claims or(b) companies with MIF claims joining Luxottica’s group. Both those possibilities were eminently foreseeable. The third difference is that the language chosen here makes it clear that it is intended, quite specifically, to cover unforeseen and uncontemplated claims (“whether or not contemplated or foreseen”). The fourth difference is the detailed definition of MIF-Related Claim. That is (unlike the language in BCCI which was so broad that some implicit limitation was required to make sense of it) a thorough attempt to define a particular category of claim precisely, but widely.[153]In short, although I agree that the primary focus is on Luxottica’s UK claim, the settlement agreement in plain terms goes wider, and it is not realistically limited to other claims “relating to Luxottica UK’s business”. It may well be that those were the claims that were most salient or, in the parties’ view, the most likely to be caught. But the terms resist a reading that is limited to such claims only, and commercial considerations do not clearly demand it.[154]Mr Coleman highlighted some subsidiary textual points where, he submitted, the agreement’s terms ran against the grain of any effect as broad as Visa say the agreement had. Would so broad an intention not entail making other companies party? Could Luxottica really promise to control other group companies? Could it really have authority on their behalf? Would broader exceptions to confidentiality not be required?[155]I accept the force of those points to the extent that I agree that they undermine the contention that group-wide perpetual settlement was a primary purpose. But that does not mean that some version of it was not a subsidiary purpose. The protection that Visa obtained was (here I agree with Mr Coleman) not bullet-proof; it depended on Luxottica’s solvency, and on its practical ability to influence other group companies. But it does not follow, legally or practically, that the agreement was implicitly limited to “Luxottica UK’s business”, when that was never said. If one assumes (reasonably, as it seems to me) that Luxottica will have considered what claims the group had and might acquire, it could decide to give the commitments it did. Is the settlement limited to the claims of companies that are Associated Companies at the agreement’s date?[156]Visa stressed that the definition of “Associated Company” looks forward and backward (“past, present or future”). But I think deeper analysis is required. For in some cases the immediate context shows that a term must be used in a more restricted sense. For instance, it is not credible to imagine that when Luxottica said that it had the power to enter into the settlement agreement on behalf of “Associated Companies” it meant that it had present power to conclude it on behalf of future Associated Companies, including those that had no existence at the time of the agreement. Even when a definition is not expressly stated to apply “unless the context otherwise requires”, if the context does otherwise require it will be necessary to take it into account.[157]In my view, it is useful to distinguish between four different categories of claim: i) Claims by companies that were part of the group at the date of the agreement, and which already existed (existing claims of existing group companies). ii) Claims by companies that were part of the group at the date of the agreement, but only arose after the agreement was concluded (future claims of existing group companies). iii) Claims by companies that were not part of the group at the date of the agreement, but which already existed when the agreement was signed (existing claims of future group companies). iv) Claims by companies that were not part of the group at the date of the agreement, and which arose later (future claims of future group companies). Such claims might be further subdivided into claims that arose after the agreement but before the company became Associated, and those that arose after the company became Associated.[158]If one looks at Recital D, it places the claims into two categories: those that relevant companies have and those that they may have. The latter category may include future claims; but the former cannot.[159]When it refers to claims that Associated Companies “have” therefore, the agreement refers to the position at the time of the agreement. That expression cannot refer to future claims.[160]Could it refer to existing claims of future Associated Companies? Linguistically, just about: one could say of (say) GrandVision “this was an Associated Company at the time the agreement was concluded because it was going to become one: it was of the genus ‘Associated Company’ and the species ‘future Associated Company’, and it had a claim”. But it would require a rigid sort of legal Calvinism to adopt that approach, treating GrandVision as an “Associated Company” that had a claim in January because it was predestined to become one in July. To my mind the natural reading of “MIF-Related Claim … that the Claimant or any Associated Company … have” is implicitly limited to existing claims of existing Associated Companies. The context of the words “Claims … that Associated Company … have” requires one to look at present (and perhaps past) Associated Companies.[161]What about “may have”? Once one appreciates—as is quite clear—that MIF-Related Claims include claims that have not yet arisen, “may have” must include not only the future claims of companies that were Associated Companies at the time of the agreement, but those of companies that may become Associated Companies later. That indeed, is the obvious purpose of including “future” Associated Companies in the definition. And it is a readily comprehensible purpose even within a narrow “anti-avoidance” conception of the parties’ purpose. Otherwise, Luxottica could re-declare war on Visa by the simple expedient of creating a new subsidiary. It follows that the future claims both of existing and future Associated Companies were caught.[162]I see no uncertainty about that, nor any difficulty with it. Mr Coleman said that it is odd and uncommercial to describe as a “Settled Claim”, or to make subject to “release” or “waiver” a claim that has not yet arisen by a person perhaps not yet in being. But the intention is unmistakeable in the definition of MIF-Related Claim, and it was dealt with not just by the “release” and “waiver” provisions, but by the terms of clause 7, which apply quite easily to future claims. So the infelicity is the choice of the word “Settled” to refer to them. But that takes nobody very far, for it is simply a convenient shorthand for what would otherwise have needed to be some cumbersome term such as “settled and conclusively resolved claims”.[163]Nor do I think that, in this respect, one can sensibly distinguish between the claims of existing companies and those of companies yet to be created, or that it can matter whether a company becomes an Associated Company by being created within the Luxottica group, or being absorbed into it later.[164]To that extent, the submission that the whole of Recital D applies only to “present Associated Companies” is not one that I can accept. Far from that being a reading that the context requires, it is a reading the context repels.[165]That, however, does not resolve the question completely. For the reasons I have given, so far as existing claims are concerned (“have”) the definition does naturally look to the claims of companies that are(or, perhaps, have been) Associated Companieswhen the agreement is concluded. So it might be said that although there is a clear answer for categories (i), (ii), and (iv), there is an unresolved ambiguity for category (iii). When a company becomes an Associated Company, its prospective claims will thereafter fall within the definition; but should its past claims? One could read the definition as a compressed way of saying that a Settled Claim is just(a) the Claim,(b) any MIF-Related Claims that any company that is an Associated Company at the time of the agreement then (“have”), and(c) any MIF-Related Claims that any company acquires in the future at a time when it is an Associated Company (“might have”).[166]That construction would mitigate the commercial problems that Luxottica relies on, and decide this case in Luxottica’s favour. It would strike a commercial balance. It has some intuitive attraction. It is the sort of distinction parties might have made. But that does not mean it is the correct construction—for it is not my job to make the agreement one that might appeal to me, but to decide what it means.[167]I have ultimately concluded that the interpretation is, as a matter of language, too rarefied, and as a matter of the interpretation of the agreement in its totality, cuts too far across its other provisions and general tenor. Granted, the claims of a future Associated Company were not, when the agreement was executed, claims that an Associated Company “had”. But there is nothing in “may have” that requires or encourages one to distinguish between claims that fall within the definition because they are acquired later, and claims that fall within the definition because a company becomes associated later: “may have” can easily cover both categories. Words are added; but they tend to comprehensiveness (“any and all”), not restriction. If one asks whether GrandVision’s claim is within the class of any and all claims that a future Associated Company “may have”, the answer must be that it is. The proof of that particular pudding being that GrandVision has become such an Associated Company and does have such a claim.[168]It would be going too far to allow commercial considerations to drive one to a hair-splitting approach to the words of Recital D. In the context of the GrandVision claim—which was being actively pursued as a third-party claim when the agreement was concluded—the consequences of a broad interpretation seem surprising. But one may twist the kaleidoscope only a little to make very different pictures, which would make an interpretation which excluded such claims unattractive. Suppose, for example, that Luxottica acquired an insolvent retailer whose only asset was a large MIF-Related Claim which had not been actively pressed, wishing to use its knowledge of Visa’s willingness to settle to obtain some ready cash. On a narrow interpretation of Recital D which excluded the existing claims of future Associated Companies, Luxottica would be free to do just that, for the newly acquired company’s claim would not be one that any Associated Company “had” when the settlement agreement was signed. Wouldn’t Visa want to prevent that? And what warrant is there for supposing that it did not? Not much, if one looks at the language in context.[169]In short, although I do understand why it might be surprising that the settlement agreement applies to GrandVision’s claim, I do not think that enables one—legitimately—to construe Recital D so that “Settled Claims” excludes claims by companies that become “Associated” only after the agreement is concluded. That would require a strained construction. I cannot be confident that it is needed to make the agreement work for both parties so as to reflect what they must both have intended. It would cross the line between construing the agreement and (from Luxottica’s perspective) “improving” it, and (from Visa’s) making it worse.

Exclusion of claims that were known at the date of the agreement but not mentioned

[170]Luxottica’s submission in relation to this possible restriction is that one should assume that general words of release cover only unknown claims, and that known claims are included only if they were specifically identified. I cannot accept this. No doubt parties will often specifically identify claims that they know about and intend to settle. But general words do more than simply cater for the unknown. For instance, they also deal with the risk that something once known may have been forgotten or overlooked. I see no warrant for treating general words as implicitly excluding particular, known claims which fall within them, especially when it then requires an examination of whether a particular claim was known at the date of the agreement. Nor is that how the agreement is drafted: Settled Claims are not defined simply to cover claims that are unknown, unforeseen, or uncontemplated, but to cover “any and all” MIF-Related Claims including (as that definition makes clear) the known, foreseen, and contemplated.[171]Furthermore, if one did ask that question, I do not think the GrandVision claim would be such a claim. The parties before me tended to the view that it was. But that involved putting together different aspects of partial knowledge. Visa knew about the claim; but it did not know, and cannot be taken to have known, that it was a claim by an Associated Company of Luxottica. (When the agreement was concluded, it was not. It might never have been.) Luxottica, through Mr Melani, knew of the possibility—not then a certainty—that GrandVision might become an Associated Company, but not of the claim. So that claim was—so far as the overlapping assumed knowledge of both parties is concerned—one that a company which might become an Associated Company in the future might have. Even if the general defining words had excluded the mutually known existing claims of Associated Companies, the GrandVision claim would not have been excluded. Summary[172]I do not accept any of Luxottica’s implied limitations on the apparent breadth of Recital D or clauses 7.1 and 7.5. Read as a whole and in context, they cover claims other than those that relate to Luxottica’s own business. They cover claims (whenever arising) by companies which only become Associated Companies after the agreement was executed. They cover such claims even if they were known to both parties at the date of the agreement—though in fact the GrandVision claim was not.

Sharp practice

[173]I turn to the law on “sharp practice”, which was forcefully developed for Luxottica by Mr Ahlquist.[174]In BCCI v Ali [2000] 3 All ER 51 (CA), a majority of the Court of Appeal (Scott V-C and Chadwick LJ), although finding that the release had extended to the claim at issue, held that there is an equitable principle by which equity would relieve against the “unconscionable” enforcement of releases. A paradigm example of such a case, as they saw it, was one in which A had persuaded B to accept general words which would release a claim which A knew B had, but B did not know they had, where A knew that B was ignorant of the claim.[175]Scott V-C discerned the origins of such an equitable principle in Taylor Fashions Ltd v Liverpool Victoria Trustees Co Ltd [1982] QB 133 (conventionally regarded as an early case of proprietary estoppel), in the principles governing rectification for unilateral mistake (Thomas Bates & Son Ltd v Wyndham’s (Lingerie) Ltd [1981] 1 WLR 505 (CA)), and in Lyall v Edwards (1861) H&N 337, 158 ER 139. He expressed the principle thus (at [25]):
“The court will not allow a general release to be enforced so as to bar a claim of which the releasor had been unaware if so to enforce it would in all the circumstances be unconscionable.”
Chadwick LJ, too, relied on Taylor Fashions. He gave a more precise statement of the terms in which he considered the doctrine would apply (at [81]):
“I would hold that, where (i) the releasee, say A, knows of facts which give rise to a claim (whether or not he believes that claim to be well-founded as a matter of law), (ii) A deliberately conceals those facts from the releasor, say B, in circumstances where A knows or believes that B cannot discover them for himself, and (iii) B does not know those facts, then A cannot rely on a general release from B as a defence to a claim based on those facts, notwithstanding that, as a matter of construction the words of the release would include all unidentified claims. A cannot rely on the release because, in the circumstances described, it would be unconscionable for him to do so.”
Buxton LJ expressed doubts on the existence of such a general equity, and said he would wish to reserve his opinion on it ([97]).[176]In the House of Lords, the majority’s conclusion on construction meant that this did not matter. However, both Lord Nicholls and Lord Hoffmann gave some endorsement or encouragement to this approach (Lord Bingham expressly gave no opinion).[177]Lord Nicholls said rather little ([32]):
“Materially different is the case where the party to whom the release was given knew that the other party had or might have a claim and knew also that the other party was ignorant of this. In some circumstances seeking and taking a general release in such a case, without disclosing the existence of the claim or possible claim, could be unacceptable sharp practice. When this is so, the law would be defective if it did not provide a remedy.”
Unlike Scott V-C and Chadwick LJ, he did not think that BCCI was such a case. In that respect, at least, he must have been disagreeing with the breadth of the principle that they had stated. In particular, he seems to have thought that it was not enough that BCCI had knowledge of the facts giving rise to a claim, because it did not have any idea that they gave rise to a claim (see [33]).[178]Lord Hoffmann started with a general discussion of the topic (at [69]):
“It is not difficult to imply an obligation upon the beneficiary of [a general release] to disclose the existence of claims of which he actually knows and which he also realises may not be known to the other party. There are different ways in which it can be put. One may say, for example, that inviting a person to enter into a release in general terms implies a representation that one is not aware of any specific claims which the other party may not know about. That would preserve the purity of the principle that there is no positive duty of disclosure. Or one could say, as the old Chancery judges did, that reliance upon such a release is against conscience when the beneficiary has been guilty of a suppressio veri or suggestio falsi. On a principle of law like this, I think it is legitimate to go back to authority, to Lord Keeper Henley in Salkeld v Vernon 1 Eden 64, 69 [28 ER 608].”
[179]Thus far, at least, Lord Hoffmann’s dicta need not to be read as stating any special doctrine for releases. A context-specific duty to speak would be consistent with ordinary principles of misrepresentation—as might references to “suppressio veri” and “suggestio falsi”. Lord Hoffmann might be saying no more than was said by Rix LJ in ING Bank NV v Ros Roca SA [2011] EWCA Civ 353, [2012] 1 WLR 472 at [92]–[93]:
“92. Outside the insurance context, there is no obligation in general to bring difficulties and defects to the attention of a contract partner or prospective contract partner. Caveat emptor reflects a basic facet of English commercial law (the growth of consumer law has been moving in a different direction). Nor is there any general notion, as there is in the civil law, of a duty of good faith in commercial affairs, however much individual concepts of English common law, such as that of the reasonable man, and of waiver and estoppel itself, may be said to reflect such a notion. In such circumstances, silence is golden, for where there is no obligation to speak, silence gives no hostages to fortune. If, however, the contractor speaks, then he may have to live up to what he says; so also where what is unsaid is sufficiently closely connected with what he has said to render what has been left unsaid misleading. 93. Nevertheless, particular circumstances can make a difference, and it is possible to formulate a general principle as to why that should be so. Thus in Moorgate Mercantile Co Ltd v Twitchings [1977] AC 890, 903 Lord Wilberforce, in a dissenting speech but which in this respect has borne fruit, spoke of the possibility that, in a particular situation which affected two parties, a reasonable man would expect the other party, ‘acting honestly and responsibly’ either to make something known or face the consequences of not doing so.”
[180]It is tempting to read Lord Hoffmann’s comments along these lines. True, English law generally allows silence to be golden, but not always. There are circumstances when either an implied duty to speak, or understanding silence as a species of fraud (‘suppressio veri’ or ‘suggestio falsi’), gives purchase to doctrines such as estoppel, or misrepresentation. General releases may sometimes exemplify such circumstances. That would not require a special doctrine for releases, but the operation of ordinary contractual and equitable doctrines in the circumstances of a particular general release.[181]Lord Hoffmann, however, went on (at [70]–[71]) to say this:
“70. In principle, therefore, I agree with what I consider Sir Richard Scott V-C … to have meant … and with Chadwick LJ, that a person cannot be allowed to rely upon a release in general terms if he knew that the other party had a claim and knew that the other party was not aware that he had a claim. I do not propose any wider principle: there is obviously room in the dealing of the market for legitimately taking advantage of the known ignorance of the other party. But, both on principle and authority, I think that a release of rights is a situation in which the court should not allow a party to do so. On the other hand, if the context shows that the parties intended a general release for good consideration of rights unknown to both of them, I can see nothing unfair in such a transaction. 71. It follows that in my opinion the principle that a party to a general release cannot take advantage of a suggestio falsi or suppressio veri, of what would ordinarily be regarded as sharp practice, is sufficient to deal with any unfairness which may be caused by such releases.”
[182]This does make it look as if there is something special, in Lord Hoffmann’s mind, about general releases as such. All counsel before me agreed that was so. The limits of such a principle would need to be drawn carefully, for it is in tension with how English contract law operates (not, usually, in contract-specific silos, nor by reference to general principles of good faith directly applied). Nor does it seem highly productive of certainty, as BCCI shows. In that case, of the four very experienced judges who sat in the Court of Appeal and the House of Lords who expressed a view on unconscionability and thought there was such a principle, two (Scott V-C and Chadwick LJ) thought that enforcement would be unconscionable; but two (Lord Nicholls and Lord Hoffmann) thought it would not be.[183]Apart from its application (incorrectly, according to Lord Nicholls and Lord Hoffmann) by the Court of Appeal in BCCI, the putative doctrine does not seem to have borne modern fruit. In Maranello Rosso Ltd v Lohomij BV [2022] EWCA Civ 1667, at [66] Phillips LJ agreed with a judge’s conclusion that, on the facts of the case, the releasing party had been aware at least of the possibility that it had claims for deliberate breach of duty, had chosen not to investigate them, and had decided to release them. He continued at [66]:
“I would add that, where a release is construed as covering unknown claims in fraud, dishonesty and conspiracy relating to the defined subject matter … such a construction entails a finding that the parties mutually intended to settle such claims. That would seem to leave little scope for a finding that one of the parties was guilty of sharp practice in relation to the existence of such a claim.”
[184]A similar conclusion was reached in Riley v National Westminster Bank plc [2024] EWCA Civ 833, at [79]. Similar points, that where a releasor plainly accepts the risk that it will release unknown claims (which the other party may know about) the doctrine cannot apply, were made by Knowles J in Tchenguiz v Grant Thornton UK LLP [2016] EWHC 3727 (Comm) at [58], and by Moulder J in Yukos Hydrocarbons Investments Ltd v Georgiades [2020] EWHC 173, at [225].[185]At first sight, this emphasis on the agreement’s terms seems strange. After all, is not the whole point of the doctrine that it will apply to claims that do fall within the scope of the release as properly construed? That is just when it is needed. The point being made, however, is more subtle and fundamental.[186]In many cases parties settle cases with a known inequality of information. Fraud cases are a prime example, but not the only one. When a claimant who has alleged deliberate misconduct settles, there are obvious reasons to suspect deliberate misconduct, and to suppose that the defendant knows more about it than the claimant does. Even in less colourful cases, a settling claimant will often know that it has not yet fully bottomed out all its possible claims. It may prefer not to do so, taking the bird in the hand for the two that may be in the bush. Defendants often know more; that may be part of the impetus to settle. If, in such circumstances, a claimant is willing to settle not only known but also unknown claims, it is not reasonable to suppose that it does so on the assumption that it is on a level playing field with the defendant, so far as knowledge of possible claims is concerned. They are “known unknowns”. To say that it is “suppressio veri” or “suggestio falsi” for a defendant not to make a clean breast to the claimant, and that a settlement on any other basis will be vulnerable to non-enforcement, would risk making such cases incapable of settlement. It would impose unrealistic standards of what the parties would honestly and responsibly expect, given the risks that each agreed to take. If the parties have expressly settled on the basis that there may be other claims of which one may be aware and the other may not be, silence is not sharp practice.[187]There was one further issue of debate. How does the “sharp practice” doctrine relate to rectification? I agree with Mr Ahlquist that one should not draw a bright line between sharp practice about facts and sharp practice about law. Both may be relevant, as BCCI itself and Salkeld v Vernon (an old case to which Lord Hoffmann referred) suggest. But it would be surprising if someone who could not establish the requirements for rectification for unilateral mistake (notably, knowledge of the mistake) could repackage essentially the same argument as a “sharp practice” point. One perhaps telling point is that, in the context of unilateral mistake—a doctrine which itself rests on unconscionability—we distinguish between attempts to exploit known mistakes about what a contract means, and cases where there is no more than a known risk of error, which are not sufficient. Mr Ahlquist may be right that certainty is not required. But once one is in the area merely of “known risk”, especially where the risk is one that the releasing party has expressly agreed to take, we are some way short of anything that could be regarded as “unconscionable”.[188]If the sharp practice argument applies, I would expect it usually to apply to specific claims. If the suppressio veri is the concealment of a claim, it is the enforcement of the release in relation to that claim which would be unconscionable. If a party wishes to render the whole agreement ineffective, it needs a suitable broader basis for that (such as rescission for misrepresentation). If a party wishes to enforce the agreement but on different terms, it must establish a valid case for rectification.

Relevant facts

[189]Luxottica’s closing written argument set out some propositions on which it founded its case on sharp practice. I shall consider them in turn. Luxottica started with Visa’s settlement practice and strategy. It asked me to find(1) that the strategy had been developed by Visa and Linklaters together,(2) that the settlement agreement was understood by Ms Williams and(3) that it involved paying “only a fair settlement sum for a particular claim … but including some very wide wording in the standard form of the Settlement Agreement in the speculative hope that it might extend to all the MIF-Related Claims of the claimant’s corporate group”.[190]I accept (1) and (2). Indeed, they are common ground. One might go further on the evidence and say that Ms Williams believed that the Settlement Agreement would cover other claims within the corporate group. It is her evidence, which I accept, that she did think that.[191]I do not accept proposition (3), either in its terms or its implications. Its implication is that Visa put forward its draft as the Greeks dragged their wooden horse to the gates of Troy, knowing that it concealed a trap. There is no basis for such a conclusion. Visa put forward the terms it was proposed to accept, both financially and otherwise. It was under no obligation to make them “fair”, and it did not—expressly or by implication—say that they were “fair”. It left it to Luxottica, which was professionally represented, to decide whether they were good enough for it to accept, and to decide what they would entail. It made it clear that it sought to resolve similar claims by group companies, and it would have been obvious to anyone that the agreement defined what that meant. That was not sharp practice. It is how negotiated settlements are made.[192]The fact that, had Luxottica sought to negotiate, Visa might have been prepared either to pay more (for a wider release and indemnity) or to carve out existing claims, makes no difference. It is not sharp practice to settle for less than the most you would be prepared to offer.[193]Propositions (4) and (5) are that Ms Williams was aware of the US Claim in December 2019, and that in the period from 11 January to 20 January 2021 she and Linklaters were aware of a potential argument that it would be released and settled by the settlement agreement.[194]Proposition (4) is admittedly correct, though of marginal relevance since it is the later period that matters. Proposition (5) is also correct; it is accepted that this very possibility had been raised internally.[195]Proposition (6) is that Ms Williams and Linklaters “were aware that the US Claim was likely to be, or could well be, a substantial pending claim brought by Associated Companies of Luxottica UK but did not seek to determine these points”. Proposition (7) repeats the point that the claim was understood to be “substantial”.[196]For reasons that I have already given, I also accept these propositions.[197]Proposition (8) is that the possibility of the US claim being settled under the Settlement Agreement was material information that Linklaters should have communicated to Visa before the settlement agreement was executed.[198]I do not understand why the materiality to Visa matters. If the purpose is to show that Linklaters’ knowledge counts as Visa’s for the “sharp practice” argument, I do not understand that to be disputed. It cannot be in doubt, not just because (as I agree) the information was material to the merits of the Luxottica settlement, but because if—as I think Ms Holtz at least seemed to suggest—the whole thing was simply being left in Linklaters’ hands, then that would also suffice to make their knowledge attributable to their client. And also because, if one is asking what Visa itself knew, Visa Inc at least must have known as much, and presumably much more, about the US claim than Linklaters, and Ms Holtz and Mr Winfield-Chislett both insist that they expected any settlement with Luxottica to settle all related claims.[199]Proposition (9) is that “Luxottica UK and its representatives made a mistake in failing to appreciate that the Settlement Agreement was not restricted to the Luxottica UK Claim”.[200]I do not accept that proposition in the terms that it is advanced. Luxottica’s witnesses (apart from Mr Bacon, but he had not read the agreement) accepted that they knew that the Settlement Agreement was not, as such, restricted to the “Luxottica UK Claim”, and that it at least went somewhat further to cover “similar” claims. Even Mr Melani thought that it covered “any other money from Visa because that is done for Luxottica UK”. None thought that it was specifically limited to just the claims that had been made.[201]If what is being alleged is a narrower mistake, namely a mistake in thinking that the settlement would not cover the US claim, Luxottica simply had not considered that possibility. The reality is not that Luxottica’s representatives mistakenly thought that the agreement did not settle anything other than the UK Claim—a mistake which would have been remarkable to anyone who read either Recital D or the accompanying letter with its reference to “similar claims”. The reality is that they gave no sustained thought to how much further it went, or the implications of that. They did not consider the US claim because they were not aware of it, even as a possibility (since both Mr Melani and Mr Bacon thought that the MIF litigation was UK- and Luxottica-specific).[202]Proposition (10) that Luxottica invites me to accept is a very long one, with several parts. I shall deal with it piece by piece.[203]The first sub-proposition advanced is that Ms Williams “knew [between 11 and 20 January 2021] that Luxottica UK probably did not intend the settlement agreement to cover MIF Related Claims of Associated Companies that had no connection with Luxottica UK’s business”.[204]I do not accept this. Ms Williams, subjectively, thought that the agreement did cover such claims. Her reasons for thinking that were consistent with the terms she had drafted. At no point in the negotiation did Luxottica say anything to her which suggested that they understood the agreement to relate only to Luxottica’s UK business, and she had no reason to think they did have that intention.[205]The remaining sub-propositions under this heading are various ways of putting a narrower proposition, which is that Ms Williams must have realised that Luxottica might not have appreciated that the settlement agreement would cover the US claim, and would not have intended or expected that it would do so, and cannot really have intended that it would. These are not disposed of by my conclusion in the previous paragraph (which deals with only one reason why the US claim might have been ignored). Their essential foundation is that Ms Williams should have found it surprising for a subsidiary to settle a substantial claim in a different jurisdiction without at least mentioning that claim or addressing it in terms, or asking for at least some money for it.[206]In my view, looking at the matter objectively, Ms Williams should have appreciated, as a risk, that Luxottica might be blundering in some way. Her evidence was that the terms of the settlement were clear, and she assumed that Luxottica knew what they were doing, including settling the US claim if it was a live claim of an Associated Company. But that evidence sits uncomfortably with her insistence that she had not carried out a “contractual analysis” to work out whether such a claim would be covered. And it sits uncomfortably with the evidence that Linklaters had identified only a potential argument that the US claim was covered. If Visa did not know whether the agreement settled the US claim, why should Luxottica?[207]Moreover, although Ms Williams said that “if anyone was going to be aware of its own claim in the US, it would be Luxottica”, the US claim was not Luxottica’s “own claim” but a claim made somewhere else by another company. If Ms Williams, acting for Visa, knew little about it, why assume that Fieldfisher, acting for Luxottica UK, would do so? Furthermore, Ms Williams was aware that the agreement contained warranties including a warranty that Luxottica was not aware of any MIF-Related Claim brought by an Associated Company. She said, in answer to a question I asked, that she had no reason to think the warranty was not accurate. But, if so, that could only be because either the US claim was not such a claim (possible, but unlikely) or because Luxottica was not, unlike Ms Williams, aware of it.[208]In my view, therefore, objectively speaking, a person in Ms Williams’ position would have been aware that Luxottica might not appreciate the effect that the settlement would have on the US claim. The most likely explanations for that—given the terms of the agreement—would be that Luxottica UK was not aware of that claim or had not thought about it. In other words, Ms Williams was objectively aware of a number of possibilities: maybe Luxottica knew about the US claim and expected it to be settled; maybe Luxottica did not know about the US claim; maybe Luxottica knew about the US claim but did not realise it would be settled; maybe there was no live US claim to know about; maybe the US claim was irrelevant.[209]So much for the objective position. What of the subjective one? Ms Williams’ evidence was that she gave the matter little thought. That was mainly because—to put it roughly—she thought it was Luxottica’s issue, not hers; and she had other things on her mind. I accept that it was not a matter of profound concern for her. But I would infer from the very limited material available that she consciously appreciated at least as one possibility that Luxottica had overlooked the issue. Since Linklaters did not know whether the US claim was going to be settled, but thought it might, it was a tiny step to suppose that Luxottica did not know either. Ms Williams may well have thought that was not her problem, but that does not mean that it would not have crossed her mind that it could be someone else’s problem. In the absence of disclosure of documents to show the internal thinking or to explain how Ms Williams could have been confident that Luxottica was aware of the issue her team had spotted but not resolved, I think it more likely than not that she realised that it was possible that Luxottica had not noticed what she had, because that is what I would expect an experienced litigator in her position to have done. Litigators spend a lot of time guessing what their counterparts are thinking. The same, I think, goes for other members of Ms Williams’ team, notably Mr Smyth, who did not give evidence.[210]Proposition (11) for which Luxottica contends is that there was no urgency which prevented the investigation of the US claim. I agree, particularly given how simple and quick it would have been for Mr Smyth to check whether the plaintiff in the US claim was an Associated Company, for Linklaters to ascertain whether that case had settled, and (if need be) for Linklaters to raise the point with Luxottica. However, I cannot see it matters, since Visa does not rely on any pressure of time as an explanation for what happened.[211]Proposition (12) is that Ms Williams took no steps to resolve the uncertainty, either internally or externally. That is common ground.[212]Proposition (13) is that Ms Williams did not take these steps because “the discovery that a claim in the US that had not been discussed in the negotiations or mentioned in the Settlement Agreement might be covered by the wording of the Settlement Agreement was consistent with Visa’s settlement strategy”. That is, delicately dressed up, an allegation of blind-eye knowledge, although no such claim is pleaded—of the “Trojan horse” approach to settlement. For reasons I have already given, even if it is open to Luxottica to allege this, I reject it. Ms Williams was aware, I agree, that the outer reach of the settlement agreement’s effect might be unclear, either factually or legally. She did not know whether it included the US claim, though she suspected it might. But she did not raise that not because she feared the answer, or feared the question, but because she regarded it as an issue for Luxottica to resolve for itself.[213]Proposition (14), though presented as a factual issue, asks me to decide that Ms Williams’ conduct was unconscionable. I address that point in the next section.[214]In summary, Ms Williams was aware of a risk that Luxottica would be settling (or indemnifying Visa) against the US claim in ignorance of the fact that that is what it was going to do, or without having considered its implications. She did not, however, know that it was so: it was possible that the US claim might turn out already to have settled or to have nothing to do with Luxottica. It was also possible that Luxottica might be doing so with its eyes wide open. She did not do anything to find out which of these things was correct, but that was not because she preferred not to know, but because she did not see it as a critical issue for Visa, and felt justified in leaving it to Luxottica and its advisers.

Was this sharp practice justifying relief?

[215]Although, logically, the first question is whether there was sharp practice at all, I prefer to begin by asking whether, even if I found that there had been sharp practice which might have led the court to refuse to permit Visa to enforce the agreement so far as the US claim was concerned, it would make a difference to the GrandVision claim. I conclude that it would not.[216]On the view most favourable to Luxottica, Ms Williams knew that Luxottica mightbe making a mistake about the meaning of the agreement as it applied to the US claim, or that it might be in ignorance of the existence or characteristics of the US claim. The first mistake could have its origins in any number of misinterpretations including (on Luxottica’s own case) at least three, for there are three “implied limitations” that Luxottica suggests might be used to cut down the breadth of Recital D. They would not be sufficient to justify rectification for unilateral mistake, as Luxottica accepts. The latter mistake, which would be factual, might at most be relevant to the enforcement of the settlement agreement over the US claim itself, but not any further.[217]Given those facts, “sharp practice” might have been relevant if Visa had been seeking to enforce the agreement in respect of the US claim. But it is not: it has settled that claim, and does not seek to recover what it paid. There is no reason why knowledge of a possible misconstruction of the agreement—only one of a number of possible mistakes or blunders that Luxottica might have been making—should prevent enforcement in relation to a claim which Visa was not aware of, and a risk that it had no reason to think Luxottica had not noticed. If Luxottica wants to say that, as a general matter, the agreement does not apply to a whole category of claim, that is an argument that either sounds in rectification, or fails.[218]In any event, even viewing the facts in a way that is favourable to Luxottica, I do not think there was sharp practice here. The settlement agreement was framed, expressly, to catch both known and unknown claims. It was unmistakeably not limited to settling just the claims of Luxottica in the High Court action. Visa had made no bones about that. Luxottica was professionally advised, and its advisers had proposed drafting amendments to a key definition which showed that they had considered—as one would expect—the draft terms. Luxottica was not in any sense vulnerable. It was not looking, and had no business to look, to Linklaters or Visa to advise it. Negotiations were at arm’s length. The risk that Luxottica might be wrong about the consequences for its business was one that Visa was, in objective good conscience, allowed to leave it for Luxottica to assess. Ms Williams was not baiting a trap, or exploiting a known error. She was simply opening a door into which Luxottica chose to step.

Relief

[219]It follows that Visa is entitled to relief. Visa seeks three categories of relief: a declaration, damages, and specific performance.[220]There is little difficulty in principle with a declaration or damages. Visa is entitled to damages for breach of the indemnity, and it is plainly desirable that the court should declare the position. There is no need to regard the damages as being “in lieu of specific performance”: they are simply damages for breach of the obligation in clause 7.1, and the indemnity in clause 7.5.[221]The only issue of principle on this point is whether Visa is entitled to recover the costs of these proceedings as part of the indemnity, so that they should be assessed on an indemnity basis. The argument that they should be rests on the contention that they are “losses which [Visa] may incur or suffer in connection with any MIF-Related Claim that ensues or is assisted by the Claimant or any Claimant Associated Company”.[222]I do not accept this argument. I agree that “in connection with” are broad words, but they must be read in context. To treat a claim to enforce the indemnity as itself a claim that is “in connection with” one to which the indemnity applies stretches them too far. If the parties had wished to provide for the costs of enforcing the settlement agreement to be subject to an indemnity, they would have done so specifically. They expressly referred to claims to enforce the agreement itself in clause 5.2, treating them as separate from (and outside) the settlement. I shall therefore assess the costs of this claim, to the extent that they are recoverable by Visa, on whatever basis would otherwise have been appropriate.[223]Specific performance is controversial. On behalf of Visa, Ms Mayer (who argued this aspect of the case for Visa) submits that breach of a settlement agreement is an established case in which specific performance may be granted. She accepts that it might be refused if Luxottica could show that it was impossible for it to “ensure” that the GrandVision claim is withdrawn. But, she says, it has not been shown to be impossible, and it is likely that if Luxottica asks GrandVision to withdraw its claim, it will. Specific performance will show the court means business, and if there is any real problem it can be dealt with later, or during contempt proceedings.[224]Neither party was able to cite any case that seems to me to cast very much light on the issue, and certainly none that remotely resembles this. There is authority that an obligation to do a particular thing may be specifically enforced even if that requires the consent of a third party, unless it is very unlikely that consent will be given. But no authority was cited to me dealing with the specific enforcement of broad obligations to ensure that third parties act in particular ways.[225]If the claim being brought contrary to the settlement agreement were Luxottica’s own claim, I would grant specific performance because damages are not a fully adequate remedy where proceedings are improperly brought. But I do not think it appropriate to order specific performance of an obligation to “ensure” that someone else does something, when that thing is not within the legal power of the party against whom enforcement is sought. This is not because I regard that as “impossible”, or as “acting in vain”. It is rather that an obligation in general terms to “ensure” that someone else does something, with no concrete idea for how that would be “ensured”, is not appropriately ordered. For specific performance to be appropriate, in my view, it needs to be clear what the defendant must do, so that the court can judge—without undue difficulty—whether it has done those things. Anything less is unfair to the defendant and an imposition on the court. If the GrandVision claim were Luxottica’s, that would be clear (or could be made clear). But an obligation to “ensure” that third parties, over whom the defendant has admittedly at most some indirect persuasive influence, will do something is not one for which specific performance is well suited. Luxottica has a strong financial incentive to do what it can to get the GrandVision proceedings withdrawn. To add a nonspecific obligation to “ensure” their withdrawal, backed by the contempt jurisdiction, without clearly specified steps to be taken, risks injustice and interminable further argument.[226]As a matter of the principled exercise of discretion, therefore, I decline to order specific performance.

Conclusion

[227]For the reasons I have given, the GrandVision claim has, since GrandVision’s acquisition, been a “Settled Claim” within the meaning of the settlement agreement, with all the consequences that entails under clauses 7.1 and 7.5. I shall grant appropriate declaratory relief, and assess damages (if they are not agreed) or give directions for their assessment. I shall not grant specific performance. I shall hear counsel on consequential matters, and costs.