England and Wales High Court (Patents Court) Decisions

IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
INTELLECTUAL PROPERTY LIST
PATENTS COURT

		Rolls Building Fetter Lane, London EC4
		24 November 2017

MR DANIEL ALEXANDER QC

INTRODUCTION AND SUMMARY

1. There are two applications before the court. First, an application relating to so- called Mayne Pharma disclosure (after Mayne Pharma Pty Ltd v Debiopharm SA [2006] EWHC 164 (Pat), [2007] FSR 37). Second, an application concerning whether permission should be given to conduct testing on certain confidential samples obtained from a process inspection and to admit further experts into an inner confidentiality club and if so on what terms.
2. The applications have a common leitmotif: how protective should the court be of a party's confidential information when it is said by an opposing party that it may be relevant to legal proceedings? For one of the applications, it is privileged information. For the other, trade secrets. In each application, the court is invited to require the opposing party to disclose more of its confidential information than it has done already or to more people.
3. I gave judgment with reasons to follow on the second application on 9 November 2017. Judgment/reasons have been reserved on both applications partly because of the timing of the hearings and partly because an indication has been given that the matter may go further. The applications raise issues which are not entirely straightforward in the context of ensuring that important competing interests in modern English litigation, especially relating to patents, are appropriately addressed.

The claim

4. The applications arise in a claim for infringement of EP (UK) 1 444 036 ("the Patent"), owned by the claimant ("MEL").
5. MEL is a UK-based producer of chemicals including rare-earth mixed oxides ("REMO") which are the subject matter of the Patent. REMO products are used in a process called washcoating in which active materials, including REMO, are coated onto a ceramic substrate for use in automotive emissions catalysts. The REMO delivers oxygen when needed, taking oxygen from the atmosphere when the air/fuel mixture is lean (low in fuel, high in air) and conversely releasing oxygen when the air/fuel mixture is rich (high in fuel, low in air). The characteristics of REMO products determine how well that is done.
6. The main defendants (collectively "Neo") are part of a corporate group and, for most of the issues on these applications, there is no need to distinguish between them. When these proceedings started, the group was known as the "Molycorp" group and was headquartered in the US. Since then, the group has undergone restructuring and is now known as Neo. The second defendant makes the product that is the subject of the claim in China. MEL contends that the second defendant has used the patented process, and that all defendants have infringed the Patent by importing the product of the process into the UK and disposing of it here.
7. The action is set down for trial floating from 5 March 2018. The claim is said to be of considerable commercial significance for both sides (>£10,000,000 value) which is reflected in the fact that the combined costs of the proceedings on both sides already stand at about £2 million.

The Patent

8. The Patent concerns a method of producing zirconium-cerium-based mixed oxides. Claim 1, on which all other claims depend, is as follows:

"A process for preparing zirconium-cerium-based mixed oxides which comprises reacting an alkali with an aqueous solution of a soluble zirconium salt containing 0.42-0.7 mole of sulphate anion (SO4^2–) per mole of zirconium cation at a temperature of not greater than 50°C, in the presence of a soluble cerium salt to form a cerium-zirconium mixed hydroxide, and then calcining the cerium-zirconium mixed hydroxide to form a mixed oxide."

The description discloses a number of examples of processes which are said to lead to particularly successful production of such mixed oxides as well as a number of comparative examples, where conditions, starting materials and proportions were different and which are said not to achieve results as beneficial in various respects.

PROCEDURAL BACKGROUND

9. The background to these applications is important since they are both consequential upon previous orders made by Birss J and Arnold J in 2017. One aspect of the first application raises a direct question of interpretation of one of those orders.

Order and judgment of Birss J giving permission to serve out of the jurisdiction on 3 December 2015

10. It is necessary to start even further back, in 2015.
11. MEL required the permission of the court to serve the claim on the second defendant out of the jurisdiction. Accordingly, an application was made to the Patents Court supported by evidence designed to show that there was a serious issue to be tried, inter alia, of the technical issues relating to infringement. That application came before Birss J on 3 December 2015 and was founded on the fact that, by way of a trap purchase, MEL had obtained a sample of product manufactured by the second defendant in China and supplied by the first defendant to a customer in the UK.
12. MEL analysed this product, referred to as "TP1", to compare it with its own product produced by the patented process, alongside a control sample produced by a non-patented earlier method (so-called "G1"). These were tested in a set of so-called "fingerprint" experiments which had been developed over a considerable period and were referred to in general terms at para. 3.13 of the witness statement of Dr Moles of MEL. Dr Moles was a senior scientist working for MEL who is now retired and he gave his evidence on the basis of his knowledge of products of this kind. As explained in that statement, these experiments ("the Experiments") were not complete at that stage but it was said that sufficient results had been produced to raise a prima facie case that the second defendant's process was indeed the patented process.
13. Birss J accepted that the evidence showed a prima facie case of infringement and made the order for service out. In his reasons for making the order, Birss J summarised the technical position as follows (Magnesium Elektron Ltd v Molycorp Chemicals & Oxides (Europe) Ltd & Anor [2015] EWHC 3596 (Pat) (15 December 2015)):

"21. Mr Moles explains that the claimant has been working with the professors to develop experiments which identify what he refers to as a chemical fingerprint of REMO obtained directly by means of the patented process. The claimant says it has found that the patented process give the REMO produced by that method a unique nanostructure which in turn provides enhanced oxygen release kinetics. The nanostructure is identified by Raman spectroscopy and the oxygen release kinetics are identified by two techniques: Hydrogen Transient Isothermal Reduction (H₂TIR) and Transient Isothermal Isotopic Exchange (¹⁶O/¹⁸O). Draft protocols for these three experiments are exhibited by Mr Moles. The Raman spectroscopy reveals information about the physical bulk structure of the metal oxide solids focussing on the oxygen sub-lattice, H₂TIR evaluates the concentration of labile lattice oxygen removed from the metal oxide and the oxygen isotope technique probes the mobility of surface and bulk oxygen of the oxygen sub-lattice of the metal oxide.

22. Carrying out the full experiments is a major project and is taking time to complete. The samples tested so far are (i) a sample from the trap purchase ("TP1"), (ii) two samples both made by the patented process (S1 and S2), and (iii) a control sample (S8). Samples S1 and S2 differ in that while both are made by the patented method, S1 was made especially for the experiments while S2 was taken from a batch of commercial product. The control sample S8 was made by the claimant's previous, first generation process. There are other products mentioned in the exhibit (S3 to S7) but no results have been produced in evidence for them.

23. Mr Moles has discussed the initial results with Professor Efstathiou and Professor Boghosian. They have told him that TP1 behaves almost identically to S1 and S2 and that all these three samples behave very differently from S8. The full written report is not yet available and will be produced in mid-January 2016 once all the data has been captured. Mr Moles states that:

"The experiments on the trap purchase show that the second defendant's REMO has identical characteristics in terms of nanostructure and kinetics and therefore I believe that they must have been produced by the Patented process for I know of no other process which could produce such results."

14. The Experiments comprise three sets of tests: Hydrogen Transient Isothermal Reduction ("H₂-TIR"), Transient Isothermal Isotopic Exchange ("TIIE") analysis, and Raman spectroscopy. In each case, the experimental protocols were designed by, and the experiments carried out by, one expert. The results were then analysed and conclusions drawn by a separate expert (referred to as the "Peer Review" experts), as follows:

(1) Prof Efstathiou designed and carried out the H₂-TIR and TIIE experiments. These were reviewed by Prof Goodwin.

(2) Prof Boghosian designed and carried out the Raman spectroscopy. This was reviewed by Prof Martínez Arias.

15. There was, at that stage, limited material available to Birss J concerning the manner in which the Experiments had themselves been tested or validated to ensure that they were both sufficiently sensitive and sufficiently specific to be reliable "fingerprint" tests for product made and only made according to the process of the Patent.

APPLICATION 1 – MAYNE PHARMA DISCLOSURE

Order and judgment of Birss J at the case management conference on 27 April 2017

16. In substance, the first application invites the court to rule as to whether further disclosure should be given about the Experiments and whether what has been provided so far is enough. For this application, the procedural history must be taken up again nearly 18 months later with the case management conference ("CMC") held before Birss J on 27 April 2017. That was a considerable time after the initial exchange of pleadings had taken place in January and February 2016 partly as a result of considerable delay in serving the second defendant in China.
17. By the time of the CMC, the Experiments referred to by Dr Moles had been completed. Among the applications made at the CMC, Neo made an application for disclosure of the material connected to the Experiments, including privileged material on the Mayne Pharma basis. The material sought by Neo was summarised in the second statement of Mr Jenkins at paras. 19-21 and, as recorded in Birss J's judgment at para. [2] a list of materials, some specific and some general was sought. It is important to note that this application was made with a view to avoiding Neo even providing a PPD on the basis that, if the materials showed that there was no arguable case because the experiments were so flawed as to form no basis for the claim, the action should be struck out. Birss J did not accept this overall approach was correct and thought the defendants should provide a PPD in any event, which he ordered.
18. MEL's position at the CMC was that it proposed to disclose finalised protocols and results for the Experiments, details of the instructions to the experts and the expert reports on the Experiments. These proposals were embodied in paragraph 6 of the Birss J's CMC order which was drafted by reference to MEL's evidence for the CMC.

The judgment relating to Mayne Pharma disclosure

19. Birss J went on to consider whether MEL ought to go further and disclose additional material (sometimes referred to as "workup" material) relating to the experiments notwithstanding the fact that it may be privileged.
20. MEL resisted giving any such disclosure. Birss J therefore heard extended argument concerning the scope and effect of Pumfrey J's decision in Mayne Pharma. In particular, MEL submitted that Mayne Pharma disclosure was limited to cases where experiments were deployed in support of a case that a patent was invalid on the grounds that the prior art contained instructions which, if followed, would inevitably produce something within the claims (so-called "inevitable result" cases). The argument was advanced by MEL on the footing that, in such cases, a person deploying experimental results is also in effect deploying them to show that undertaking experiments in the way in which they were undertaken was inevitable and straightforward. That, it was said, gives rise in such cases to a need to see the relevant instructions to the person carrying them out and information as to how the experiments were conducted. MEL argued that this did not apply in other cases.
21. Birss J held that, even though Mayne Pharma was an "inevitable result" case, the principle that disclosure of material relating to experiments deployed in patent litigation was not limited to such cases (see [2017] EWHC 1024). He held that Mayne Pharma was "not as narrow as [Counsel for MEL] submits" and that Pumfrey J had been careful to make the basis upon which he was operating clear. Birss J referred to the fact that Mayne Pharma had addressed both the conflict of authorities relating to disclosure of material relating to experiments (Honeywell v. Appliance cf. Electrolux v. Black & Decker) and the relevant case law in Nea Karteria [1981] Comm LR 138 relating to waiver of privilege. Referring to Pumfrey J's exposition of the law, Birss J said at [19]:

"That paragraph [of Mayne Pharma] makes it clear that the context that Pumfrey J was concerned with was, as he said, the work up of an experiment. It does not say that the context in which Pumfrey J was concerned was the work up of an experiment used only to prove an inevitable result."

and at [22]

"In my judgment…the principle is not confined only to cases where the experiment is to show anticipation by inevitable result, but is concerned with the question of work-up of experiments in general. It is a narrow question because it is confined to work-up of experiments, but it is about experiments in patent cases irrespective of the legal conclusion on which they are being deployed to prove. That conclusion makes sense in any event…This principle does apply on the facts of this case. Accordingly there will come a time when work-up disclosure may well need to be provided by the claimant. However, the question is when."

22. Having considered the timing, Birss J declined to order work up disclosure immediately but said at [26]:

"The defendants, assuming the proceedings continue, will have to give a product and process description (PPD) and the claimant, assuming the proceedings continue, will have to give Mayne v. Debiopharm disclosure relating to the experiments which it has provided at this early stage."

Paragraph 10 of Birss J's CMC order

23. There was some discussion during argument on that occasion as to precisely what disclosure should be given and how but Birss J declined to provide detailed guidance on this. Instead, Birss J ordered at para. 10 of the CMC order:

"The Claimant shall, on or before 17 July 2017, give standard disclosure by list in form N265 (modified as appropriate) (including, for the avoidance of doubt, privileged material pursuant to Mayne Pharma v Debiopharm [2006] FSR 37) in relation to the documents directly associated with or underlying the evidence adduced under paragraph 6 of this Order".

24. The language was taken from the Mayne Pharma case. MEL did not appeal against that order. Nonetheless, MEL has purported in its skeleton argument to "reserve its right to contend that Mayne Pharma is limited to cases of inevitable result if this matter goes further". MEL has not sought to take that matter further and whether or not it is open to it to do so is a matter for question. It is, however, open to MEL to contend that, notwithstanding the order or given its terms, in the light of the material which has been provided no more should be ordered and, if necessary, the order varied to that end.

The materials provided pursuant to Birss J's order at the CMC

25. Following the CMC, MEL provided the materials which are summarised in MEL's skeleton argument as follows:

(1) Martínez Arias 1: This is a 15 page witness statement/expert report plus a lever arch file of exhibits. This was provided pursuant to para. 6 of Birss J's order.

This statement relates to the Raman spectroscopy experiments. The statement, together with exhibits, include full details of Prof Martínez Arias' instructions, the finalised protocols for the experiments, the finalised data from the experiments, and the Professor's conclusions at para. 3.64. The exhibits include his full technical report.

(2) Goodwin 1: This is a 19 page witness statement/expert report plus two lever arch files of exhibits. This was also provided pursuant to para. 6.

This statement relates to the H₂-TIR and TIIE experiments. The statement together with exhibits include full details of Prof Goodwin's instructions, the finalised protocols for the experiments, the finalised data from the experiments, and Prof Goodwin's conclusions at para. 3.40. The exhibits include his full technical report.

(3) Bould 5: This is a 55 page witness statement with seven lever arch files of exhibits.

Ms Bould is the partner at Pinsent Masons, MEL's solicitors, with conduct of the case. Her statement describes the full detail of the processes of carrying out the Experiments, including details of the instructions given to Professors Boghosian and Efstathiou and the iterations of the protocols.

Bould 5 also confirms that Ms Bould has considered the entirety of the documents and that she considers that the three witness statements and exhibits present the complete picture in relation to the matters covered by them.

(4) Raw data results of the Raman spectroscopy (32 lever arch files).

(5) Raw data results of the H₂-TIR experiments (12 lever arch files).

(6) Raw data results of the TIIE experiments (8 lever arch files).

In total this comprised 89 pages of narrative witness statements plus 62 lever arch files of documents. On any view, that is a formidable body of material relating to the Experiments. Part of the exhibits include the instructions notifying the experts that they may be asked further questions about the Experiments and there is no suggestion that they would not answer them. I have reviewed the reports and related material, albeit not in great detail, and it appears that they are the fruit of careful analysis with the experts being responsive to comments and suggestions as to how the Experiments could be improved. The material exhibited and the narrative in Bould 5 reveals an interactive process in developing the protocols in which Pinsent Masons was involved in various respects making suggestions of various kinds.

Costs of the exercise so far and projected costs

26. The evidence is compelling that the costs of provision of the material above have been significant. This has involved in excess of £120,000 in legal costs plus a substantial amount of time costs of MEL employees. It is also noteworthy that Neo has not identified any critical benefit to have emerged from that exercise at this stage. It is not in dispute that whatever the exact figures (estimated by MEL to exceed a further £100-£125,000 plus further party time but which is said by Neo to be less) the cost of the further disclosure in issue on this application would be substantial with the bulk devoted to the first category (identified below) involving examination of many thousand documents. MEL submits, in effect, that if still more is needed, the requirement to provide Mayne Pharma disclosure would have got out of hand and cannot be what that case or Birss J was contemplating when he made the order especially since Birss J has been one of the judges most concerned to ensure that disclosure is kept under control.

Neo's criticisms of this material and request for further disclosure

27. Nonetheless, Neo criticises what has been produced by MEL in a number of ways.
28. First, it says that there has been no list of documents on Form N265 and that the narrative provided in Bould 5 is an inadequate substitute. As to this, MEL accepts that it has not complied with the letter of Birss J's order in this respect. However it submits that rather than produce a list on form N265, it has produced the material in a form (narrative witness statements plus exhibits which includes a list of them) which is more digestible and helpful than a pure list of documents. It submits that its approach is preferable to producing "a vast collection of documents such as all the emails passing between Pinsent Masons, MEL and the experts" which it says would have had to be heavily redacted, which would have been excessively expensive and time-consuming.
29. Second, Neo contends that documents in specific categories have not been provided. Although one category of materials (raw data from the Experiments) has now been provided shortly before the hearing, Neo contends that MEL has omitted entirely to search for and produce documents in three categories and that it should have done pursuant to Birss J's order:

(1) Documents relating to the experiments before 2014;

(2) Documents prepared by lawyers with comments on the experiments;

(3) Drafts of the expert reports.

MEL contends that this material should not be disclosed for a range of reasons.

30. The upshot is that it is not in dispute that:

(1) Disclosure in precisely the terms of Birss J's order has not been provided;

(2) Documents have been provided which are said by MEL to be "better than" disclosure, including a large quantity of detailed material relating to the Experiments but which Neo disputes is "better" in any relevant respect;

(3) A great deal more material is sought on the footing that Birss J's order should be interpreted as having already ordered that it should be provided.

31. I now turn to the detail of the application and its genesis.
32. By letter dated 24 August 2017 from its solicitors, Neo requested that MEL disclose the four further categories of material. There is, however, no application on the part of Neo for specific disclosure of that material. That is because Neo treats the order of Birss J as already requiring disclosure of these documents.

Compliance with Birss J's order

33. The central issue on this application is whether the materials provided complied with the order of Birss J at the CMC and, if they did not, what further disclosure should be ordered. That is related to the question of whether Birss J's order should be retrospectively varied so as to "excuse" provision of non-conforming material (as the application seeks). MEL has made a pre-emptive application to the court for (as it puts it) "approval of what it has done, or else guidance as to precisely what more it is required to do".

The core issues including interpretation of Birss J's order and the scope of Mayne Pharma disclosure

34. Birss J's order provided that there should be standard disclosure of those materials of which the law required disclosure following the language of Mayne Pharma, without going into detail as to what that was. It was, if I may respectfully say so, rightly non-specific. That is confirmed by the exchanges between counsel at the hearing before Birss J, which I was taken through. Birss J decided the "macro" question of whether Mayne Pharma disclosure should be given in cases other than "inevitable result". I now have to decide the "micro" question of precisely what that requires in the present case, albeit having regard to the materials which have been provided. It follows that although the issue on this application can be framed as one of pure interpretation of Birss J's order, it is more realistically seen as dealing with the issue upon which Birss J declined to give substantive guidance namely the scope of Mayne Pharma disclosure, which is a matter of general procedural law rather than pure interpretation of the order.
35. The underlying questions are these. First, in what materials should privilege be taken to have been waived, potentially opening them up to disclosure and, second, should specifically further standard disclosure be required of them as a result? It is important to keep these two questions distinct because, even if privilege does not attach to certain materials because of waiver, it does not follow that standard disclosure of them will automatically be ordered. Questions of necessity, proportionality and so forth also come into play.
36. Underlying these points still further are two issues which arise in other contexts: how to determine the scope of waiver of privilege and how the court should approach disclosure in the light of the CPR. The latter issue is particularly important in the light of the efforts which have been taken in recent years to limit disclosure, including the current pilot schemes which aim to reduce it still further. The parties are far apart on both with MEL contending in substance that a narrow and focussed approach should be taken to both issues and Neo contending that a wider, more comprehensive, approach is appropriate not least to ensure against the risks of a partial presentation of information (in both senses of the word "partial"). Both sides appeal to procedural fairness in support of their positions. It is therefore necessary to examine the law relating to these issues and their particular application in the context of experiments in greater detail than would normally be desirable.

LAW

Principles

37. The principles applicable to this case lie at the intersection of several areas of procedural law: (i) privilege (ii) waiver of privilege (iii) waiver of privilege in material relating to experiments pursuant to Mayne Pharma (iv) disclosure and alternatives. It is therefore necessary to say something about each of these areas in turn since they each have an impact on the approach that should be taken to Birss J's order. In doing so, I have drawn for illustrative examples on some of the additional authorities not specifically mentioned by the parties but cited in the text of which extracts were provided or in other cited cases.

(i) Privilege

38. As counsel for MEL submits, legal professional privilege is a fundamental right, which has long been regarded as essential for the purposes of the rule of law and access to justice (see Three Rivers District Council v Bank of England (No. 6) [2004] UKHL 48, [2005] 1 AC 610, [23]-[34] (Lord Scott of Foscote) and General Mediterranean Holdings SA v Patel [1999] EWHC 832 (Comm), [2000] WLR 272, 280F-291E (Toulson J) where the principles were summarised.
39. Lord Taylor of Gosforth CJ said in R v Derby Magistrates' Court ex parte B [1996] AC 487, 507:

"The principle which runs through all these cases, and the many other cases which were cited, is that a man must be able to consult his lawyer in confidence, since otherwise he might hold back half the truth. The client must be sure that what he tells his lawyer in confidence will never be revealed without his consent. Legal professional privilege is thus much more than an ordinary rule or evidence, limited in its application to the facts of a particular case. It is a fundamental condition on which the administration of justice as a whole rests".

40. Legal professional privilege also extends to communications with a third party for the purposes of litigation because it is a single privilege encompassing both legal advice privilege and litigation privilege (see Three Rivers (No. 6) at [105] (Lord Carswell)). The strength of the principle is reinforced by the fact that, although legal professional privilege can be abrogated by statute, that can only be done by clear words (see General Mediterranean). It is not possible for the court to create exceptions to it.
41. In addition to its substantive importance, these principles also inform the interpretation of provisions said to affect privilege. The House of Lords has held that construction of a statute needs to be done against the background of these principles where a proposed interpretation may place them in jeopardy. Thus, in Special Commissioner and Another, Ex P Morgan Grenfell & Co Ltd, R v. [2002] UKHL 21 (16th May, 2002), Lord Hoffmann said:

"7. Two of the principles relevant to construction are not in dispute. First, LPP is a fundamental human right long established in the common law. It is a necessary corollary of the right of any person to obtain skilled advice about the law. Such advice cannot be effectively obtained unless the client is able to put all the facts before the adviser without fear that they may afterwards be disclosed and used to his prejudice. The cases establishing this principle are collected in the speech of Lord Taylor of Gosforth CJ in R v Derby Magistrates Court, Ex p B [1996] AC 487. It has been held by the European Court of Human Rights to be part of the right of privacy guaranteed by article 8 of the Convention (Campbell v United Kingdom (1992) 15 EHRR 137; Foxley v United Kingdom (2000) 31 EHRR 637) and held by the European Court of Justice to be a part of Community law: A M & S Europe Ltd v Commission of the European Communities (Case 155/79) [1983] QB 878.

8. Secondly, the courts will ordinarily construe general words in a statute, although literally capable of having some startling or unreasonable consequence, such as overriding fundamental human rights, as not having been intended to do so. An intention to override such rights must be expressly stated or appear by necessary implication. The speeches of Lord Steyn and myself in R v Secretary of State for the Home Department, Ex p Simms [2000] 2 AC 115 contain some discussion of this principle and its constitutional justification in the context of human rights. But the wider principle itself is hardly new. It can be traced back at least to Stradling v Morgan (1560) 1 Pl 199."

42. The same must apply to construction of an order of the court and should inform what order a court should make in future. As to the approach to construction generally, that set out by the Privy Council in Sans Souci Ltd v VRL Services Ltd (Jamaica) [2012] UKPC 6 (7 March 2012) namely that a judgment may be used to construe an order is applicable in principle but is not of great assistance because Birss J did not address the specific point in issue here in his judgment.

(ii) Waiver of privilege

43. While privilege is to the extent described above an absolute right, it may be voluntarily eroded or destroyed by waiver. The circumstances in which privilege is to be treated as waived is one of the age-old questions in civil litigation and often gives rise to difficulties. No less troublesome is the issue of scope of waiver of privilege: the effect of waiver of privilege in one document or part of a document on the privilege attaching to other documents or parts of them. An illustration of this comes from the treatment of the subject in Matthews & Malek on Disclosure, 5^th Edition (2017) ("Matthews & Malek") which says at para. 16.40 that where waiver of privilege before trial is concerned there is a conflict of authority "as to whether waiver is confined to the document or documents concerned…or goes wider [extending to "all material dealing with the transaction the subject of the documents concerned"]. It is well established that deployment of a document at trial may result in a waiver of privilege in other related material (see Nea Karteria Maritime Co v Atlantic & Great Lakes Steamship Corporation [1991] ComLR 138, 139 (Mustill J). The correct approach, derived from Nea Karteria, in such cases was summarised in Matthews & Malek in an oft cited passage:

"The general rule is that:

Where a person is deploying in court material which would otherwise be privileged, the opposite party and the court must have the opportunity of satisfying themselves that what the party has chosen to release from privilege represents the whole of the material relevant to the issue in question. To allow an individual item to be plucked out of context would be to risk injustice through its real weight or meaning being misunderstood.

The key word here is "deploying". A mere reference to a privileged document in an affidavit does not of itself amount to a waiver of privilege, and this is so even if the document referred to is being relied on for some purpose, for reliance in itself is said not to be the test. Instead, the test is whether the contents of the document are being relied on, rather than its effect. The problem is acute in cases where the maker of an affidavit or witness statement has to give details of the source of his information and belief, in order to comply with the rules of admissibility of such affidavit or witness statement. Provided that the maker does not quote the contents, or summarise them, but simply refers to the document's effect, there is apparently no waiver of privilege. This benevolent view has not been extended to the case where the maker refers to the document in order to comply with the party's need to give full and frank disclosure, e.g. on a without notice (ex parte) application".

44. This passage has been regularly approved including by the Court of Appeal. For example, in Dunlop Slazenger International Ltd v Joe Bloggs Sports Ltd [2003] EWCA Civ 901 at [11] (Waller LJ) and it was said to be "a convenient summary of the principle and the effect of former cases" in Lucas v Barking, Havering & Redbridge Hospitals NHS Trust [2003] EWCA Civ 1102, [2004] WLR 220, [19] (Waller LJ).
45. The passage nonetheless merits additional comment.
46. First, notwithstanding the strength of the rule protecting legal professional privilege, once there has been deployment of some otherwise privileged material, the courts are reluctant to permit selective deployment. But, the passage gives no specific guidance as to what further material is required or a formula to determine precisely how the scope of the waiver is to be determined.
47. Second, it has been repeatedly held (or it is implicit) that deployment of a document does not automatically open a floodgate to a wide destruction of privilege more broadly. The approach of Nea Karteria prevents individual documents or parts of documents from being put before the court so as to avoid the document deployed not being seen in its proper context. It is not, however, a rule which provides that, once a privileged document is deployed, all privileged documents related in some general way to that document or which can, in a general sense, be described as "relevant" to the issue must also be disclosed even if they deal directly with the subject matter in hand. That appears from Mustill J's consideration of two categories of documents in the Nea Karteria case itself. On one side were the questions put by a lawyer of which the answers were deployed so as to waive privilege in the answers: held - privilege waived in the questions as well. On the other side were instructions to the lawyer: held - privilege not waived in these documents forming part of the "prior history". Moreover, Mustill J referred to the proposition in the commonly cited passage as the principle "underlying the rule of practice exemplified by Bucknell v. British Transport Commission" [1956] 1 QB 187. In that much earlier case, a witness for the plaintiff was under cross-examination when counsel for the defendant asked him whether he had said X or Y in a statement. The witness agreed that he had. Counsel for the plaintiff then called for the statement but the defendant's counsel declined to provide it saying it retained its privilege. Production was ordered by Sellars J and the whole statement was put in evidence. The Court of Appeal unanimously dismissed an appeal. Denning LJ said at 190:

"…although this statement may well have been privileged from production and discovery in the hands of the [defendant] at one stage, nevertheless when it was used by cross-examining counsel in this way he waived privilege, certainly for that part which was used; and in a case of this kind, if the privilege is waived as to the part, it must, I think be waived also as to the whole. It would be most unfair that cross-examining counsel should use part of the document which was to his advantage and not allow anyone not even the judge or opposing counsel, a sight of the rest of the document, much of which might have been against him."

48. The "whole" was the whole witness statement in that case. There was no suggestion that deployment of that statement would open up a range of other documents on the basis that privilege had been waived in them as well because they were directly related to the issue.
49. However, in some cases citing the Nea Karteria passage, waiver of privilege in a document has been held to extend not just to the document as a whole but to other documents forming part of the transaction or communication in question. In Dunlop, where the rule was described as one designed to avoid "cherry picking", privilege was held to be waived in the whole of the communications, not just the whole of a single document. Thorpe LJ referred to "the whole of the privileged document or conversation which the contents have revealed" namely "documents relating to the communications surrounding the key date in February 2003 and redacting any advice to the defendants" at [19], although the court specifically excluded "advice from the solicitor i.e. client solicitor advice" (see e.g. at [17]). In one sense, of course, Nea Karteria was itself such a case of waiver of privilege in additional material but it was perhaps better seen as a situation in which a full set of both questions and answers was ordered to be disclosed in circumstances where a party was seeking to deploy the answers.
50. It is at this point that difficulties set in with applying the proposition in Nea Karteria. As soon as it is accepted that privilege may have been waived in other documents forming part of a transaction or communication where one of the documents of that transaction or overall communication is deployed, the principle of implied or consequential waiver has broken loose from its narrow origins in Bucknell or Nea Karteria. Once it is free from the bounds of a document (or a question/answer session), working out precisely how it should be limited is not easy. That is made more challenging by reflecting on the fact that what is contemplated is an implied or consequential waiver of an absolute right which is otherwise regarded as particularly strong.
51. The scope of this implied or consequential waiver is described in the case law in terms of potentially flexible meaning related to avoidance of unfairness. These terms have included: "unfair" to avoid parts of a document which might have been "against" the deployer being seen (Bucknell); "relevant to the issue… risk injustice…through real weight or meaning being misunderstood" (Nea Karteria); preventing "cherry picking" (Dunlop); "avoid unfairness or misunderstanding" (Fulham - cited below, relying on Paragon also cited below) and various other ways in which the scope of waiver has been defined in the cases. Of course, what may be required to avoid unfairness in any given case is itself open to debate: one person's cherry picking may be another person's reasonable and proportionate putting forward of only that which is relevant.
52. It is not possible to avoid these difficulties altogether by reference to waiver of privilege taking place in the "transaction" as a whole, since that begs a related question of how the boundaries of the "transaction" are to be determined. There is, as Matthews & Malek acknowledges at para. 16.40, "room for argument in any given case as to what constitutes the "transaction" in question" and whether, even if that can be determined, the boundaries of implied or consequential waiver of privilege are limited to that.
53. A helpful illustration of these issues comes from the judgment of Mann J in Fulham Leisure Holdings Ltd v. Nicholson Graham & Jones [2006] EWHC 158. Although it was not discussed by the parties before me, the passage in Matthews & Malek which refers to it was not suggested to be wrong in summarising the issues by reference to this authority. It is more useful to reproduce an extract from the case itself, since it contains a review of other cases in which the principle is discussed and applied. Mann J said:

"11. Based on the authorities which I am about to refer to, it seems to me that the relevant process should be as follows:

i) One should first identify the "transaction" in respect of which the disclosure has been made.

ii) That transaction may be identifiable simply from the nature of the disclosure made – for example, advice given by counsel on a single occasion.

iii) However, it may be apparent from that material, or from other available material, that the transaction is wider than that which is immediately apparent. If it does, then the whole of the wider transaction must be disclosed.

iv) When that has been done, further disclosure will be necessary if that is necessary in order to avoid unfairness or misunderstanding of what has been disclosed.

That chain is not articulated in terms in the authorities to which I am about to refer, but it seems to me that it is apparent from it.

12. The starting point to me seems to me to identify what Hobhouse J called the "transaction" in General Accident Fire and Life Assurance Corp Ltd v Tanter [1984] 1 WLR 100. In that case Hobhouse J was dealing with a wide ranging request for disclosure and inspection of otherwise privileged documents (covered by legal professional privilege), on the basis of use of one note at a trial. Hobhouse J refused to order that disclosure, and in the course of his judgment he considered the then existing authorities on the point. At page 111, he cited part of the judgment of Cotton LJ in Lyell v Kennedy 27 Ch D 1, which he described as "the cardinal quotation":

"There was this contention raised, which I have not forgotten, that the defendant had waived his privilege, and therefore could not claim it at all. That, in my opinion, was entirely fallacious. He had done this, he had said 'Whether I am entitled to protect them or not I will produce certain of the documents for which I had previously claimed privilege – I will waive that and I will produce them' but that did not prevent him relying on such protection with regard to others which he did not like to produce. It is not like the case of a man who discloses part of a conversation and then claims protection for the remainder, and we think there is no ground for the contention that there has been here a waiver of privilege."

13. It is apparent from that that it is open to a party to disclose some but not all of the documents falling within a privileged category. If further authority for that proposition is required, it is to be found in Paragon Finance v Freshfields [1990] 1WLR 1183 at p.1188D where Lord Bingham of Cornhill CJ said:

"While there is no rule that a party who waives privilege in relation to one communication is taken to waive privilege in relation to all, a party may not waive privilege in such a partial and selective manner that unfairness or misunderstanding may result."

14. The question therefore arises as to where the boundaries of the waiver are, or are taken to be. The first boundary is defined by reference to the "transaction" in relation to which disclosure is made. In the course of his judgment in Tanter Hobhouse J summarised various principles which applied and they included the following (at pp.114 – 5):

"Sixth, by adducing evidence at a trial one does get involved in potential further waiver. The underlying principle is one of fairness in the conduct of the trial and does not go further than that. The fact that this principle does not arise unless you adduce the evidence at the trial is clearly stated in the judgment of Mustill J [in the Nea Karteria case] and it was clearly raised by the facts in the Doland case and it was likewise raised by the facts in the Great Atlantic and Burnell cases. Further, if the evidence is adduced, then the extent of the waiver relates to the transaction to which that evidence goes. The extent of the transaction has to be examined and where it is what somebody said on a particular occasion, then that is the transaction. It is not the subject matter of those conversations. It does not extend to all matters relating to the subject matter of those conversations." Hobhouse J's principles, including that one, were cited with approval by the Court of Appeal in Tanap Investments (UK) Ltd v Tozer (unreported, 11^th October 1991, Balcombe and Taylor LJJ). It is from his sixth principle that the need to identify the transaction is apparent. It is the first means of ascertaining what limits there are to the waiver that has been made.

15. Earlier passages in Hobhouse J's judgment shed light on what he means by the "transaction". At page 113 he considers the judgment of Mustill J in the Nea Karteria case and says:

"Furthermore, he applies what for convenience can be called a single transaction test. He looks to see what is the issue in relation to which the material has been deployed. He held that the issue was what was said at the meeting between the lawyer and the Greek seaman, and the correct evaluation of that meeting, but he did not accept that it extended to other matters which did not form part of the transaction…. Furthermore, it is central to Mustill J's judgment that you must define what is the subject matter with which you are concerned. He did not treat the subject matter as all the things about which the seaman spoke, but he merely treated as the subject matter what had been said on that occasion."At page 114C, applying his principles to the facts of his case (which involved the disclosure of information about one conversation) he said:"What will have happened is that the putting into evidence of that document will have opened up the confidentiality of that transaction. It has not opened up the confidentiality of later privileged communications….."

That process enables the parties and the court to identify how far the disclosing party has gone (it being that party's right to limit waiver if he wishes to do so).

16. A similar approach can be detected in the judgment of Auld LJ in the Factortame case, though he uses different language. In that case (of which only an unpaginated transcript was made available to me) he started his analysis by referring to the:

"…classic judicial statement of principle…of Mustill J in Nea Karteria at 139: 'Where a party is deploying in court material which would otherwise be privileged, the opposite party and the court must have an opportunity of satisfying themselves that what the party has chosen to release from privilege represents the whole of the material relevant to the issue in question. To allow an individual item to be plucked out of context would be to risk injustice through its real weight or meaning being misunderstood.'"The most obvious application of that principle is in relation to a single document, where a party waives privilege as to part of it but seeks to withhold the rest of it…... It also extends to attempted partial waiver of privilege in respect of certain of a number of documents relevant to the same issue or transaction. Of course, the scope for unfairness depends on the breadth of the matter in issue or their severability if more than one, and on the exact relationship and/or relevance to such issue(s) of the documents respectively disclosed and sought to be withheld. It may or may not be that partial disclosure of documents going to a matter or matters in issue, say in an exchange of correspondence with legal advisers, would be unfair."Much depends on whether the party making partial disclosure seeks to represent by so doing that the disclosed documents go to part or the whole of an 'issue in question', the expression used by Mustill J in the passage from his judgment in Nea Karteria that I have cited. The issue may be confined to what was said or done in a single transaction or it may be more complex than that and extend over a series of connected events or transactions. In each case the question for the court is whether the matters in issue and the document or documents in respect of which partial disclosure has been made are respectively severable so that the partially disclosed material clearly does not bear on matters in issue in respect of which material is withheld. The more confined the issue, for example as to the content of a single document or conversation, the more difficult it is likely to be to withhold, by severance, part of the document or other documents relevant to the document or conversation."

17. In that passage, Auld LJ refers to the "transaction" but he also uses the expression "issue", which was the word used by Mustill J in Nea Karteria. He is apparently using the same concept as that deployed by Hobouse J.

18. What those citations show is that it is necessary to bear in mind two concepts. First of all, there is the actual transaction or act in respect of which disclosure is made. In order to identify the transaction, one has to look first at what it is in essence that the waiving party is seeking to disclose. It may be apparent from that alone that what is to be disclosed is obviously a single and complete "transaction" – for example, the advice given by a lawyer on a given occasion. In respect of disclosure before a trial that may be all that the non-disclosing party has to go on, because a wider context may not yet be apparent (or at least not until the exchange of witness statements). This may explain the contrast that Hobhouse J drew between disclosure before a trial and deployment at trial. However, in order to ascertain whether that is in fact correct one is in my view entitled to look to see the purpose for which the material is disclosed, or the point in the action to which it is said to go. That explains at least some of the references to the "issue" or "issues" in the judgments of Auld LJ and Mustill J. Mr Croxford submitted that the purpose of the disclosure played no part in a determination of how far the waiver went. I do not agree with that; in some cases it may provide a realistic, objectively determinable definition of the "transaction" in question. Once the transaction has been identified, then those cases show that the whole of the material relevant to that transaction must be disclosed. In my view it is not open to a waiving party to say that the transaction is simply what that party has chosen to disclose (again contrary to the substance of a submission made by Mr Croxford). The court will determine objectively what the real transaction is so that the scope of the waiver can be determined. If only part of the material involved in that transaction has been disclosed then further disclosure will be ordered and it can no longer be resisted on the basis of privilege.

19. Once the transaction has been identified and proper disclosure made of that, then the additional principles of fairness may come into play if it is apparent from the disclosure that has been made that it is in fact part of some bigger picture (not necessarily part of some bigger "transaction") and fairness, and the need not to mislead, requires further disclosure. The application of this principle will be very fact sensitive, and will therefore vary very much from case to case, as Auld LJ observed in the first paragraph of his judgment cited above. It is in this sense too that the disclosure may be partial. It is part of some greater whole, not necessarily part of some larger individual transaction. I confess, with all due respect, to having had some difficulty in understanding precisely what he meant in the last sentence of the citation, but I think that my analysis is consistent with his judgment.

20. I think that the point can be illustrated by how the principles worked in the Factortame case itself. One of the issues in that case was whether or not the United Kingdom's infringement of community law was intentional or reckless. The government had received advice from time to time on the legality of the legislation, and that went to the issue I have just referred to. The Secretary of State waived legal professional privilege in respect of legal advice up to a date in October 1987, but did not waive it for advice given after that date. That limitation was challenged, and it was said that in the light of the way that it took place, there should "in fairness" also be disclosure for a later period. The Court of Appeal held that the limitation of disclosure was not inconsistent with principle. The Secretary of State expressly stated that he would not suggest at the trial that his conduct after the October date was governed by the disclosed legal advice received before that date. Auld LJ held that in the light of that:

"It is not a case of partial disclosure in relation to his conduct throughout the period in issue, but one of clear severability over two periods within it and of the disclosed and undisclosed documents relating respectively to each period. If the Secretary of State keeps to Mr Richards' word, I can see no unfairness to the applicants…. If the Secretary of State does not seek to take an unfair advantage of his partial discovery at the trial, whether as a matter of evidence or argument, the applicants would be entitled to invite the trial judge to reopen the matter and determine whether there should be further disclosure."Auld LJ did not at that point identify what the particular transaction was for these purposes, but it is to be inferred that it was one of two things. It was either the giving of advice on each relevant occasion, or, more likely, the giving of advice on the legal issue in question across the period in question, which presumably comprised several events. I say that the latter is more likely, because Auld LJ took the advice across the period as in effect one entity. One can only do that if one looks to some extent at the purpose of the disclosure, that is to say the reliance that would be placed on it in the proceedings. Since advice was only being relied on in order to support the government's case on its mental state for a certain period, that both defined and limited the transaction in question. However, if it were to transpire that use was to be made of the discovery outside that period and purpose, then it might transpire that that use would be unfair and that further matters would be opened up. Whether that is because the waiver would be taken to have been extended by virtue of the unfairness of limiting it, or whether it is because there is a separate principle of unfairness operating alongside or concurrently with the extent of the actual waiver, does not matter for these purposes. That is how the scheme seems to me to operate."

54. As can be seen from this extract, Mann J did not find it wholly straightforward to reconcile the various authorities relating to this issue but drew from them an approach that involved identifying, first, the transaction and considering, second, whether there was any separate principle of unfairness operating alongside or concurrently.
55. It is also useful to refer to one more case discussed by Mann J and referred to in Matthews & Malek to illustrate a further aspect of the application of Nea Karteria. In Fulham at [7], Mann J cited R v Secretary of State for Transport Ex parte Factortame [1997] EWHC Admin 445 (7 May 1997). In that case Auld LJ referred to Nea Karteria and said (in the passage following that cited in the judgment of Mann J above):

"28. As Mr Richards observed, all or most of the reported cases deal with narrow issues of that sort. See e.g. the Great Atlantic, per Lord Templeman at 536D-F; Konigsberg, per Peter Gibson J at 1265C-G; and Derby v. Weldon (No 10), per Vinelott J at 918a-b. In Konigsberg the matter in issue was a transfer of land and the question was whether it was a gift or a sale. Peter Gibson J held that the party asserting that it was a sale and who had waived privilege in respect of a letter from her solicitor apparently supporting her case, could not claim privilege in respect of an affidavit sworn by him doubting it. Where the issue is broad, or there are several of them, or where the history giving rise to the litigation is long and/or complicated, partial disclosure which is clearly confined in its impact to one aspect of the case may well not require the all or nothing approach. In such a case it is not, in my view, apt terminology to ask whether a series of connected events or matters is a single "transaction" or series of separate "transactions" for this purpose. Where a party's conduct over a period of time is in issue the effect of partial disclosure of documents must depend on the particular facts of the case, usually as seen before trial in the pleadings and in other interlocutory battle-lines drawn by the parties.

29. As I have said, on this application Mr Richards has stated that the Secretary of State will not suggest at the trial that his conduct after 29th October 1987 in relation to the enactment and implementation of the 1988 Act was governed by the disclosed legal advice that he received before that date. It is not a case of partial disclosure in relation to his conduct throughout the period in issue, but one of clear severability of two periods within it and of the disclosed and undisclosed documents relating respectively to each period. If the Secretary of State keeps to Mr Richards' word I can see no unfairness to the applicants. The applicants and the Court know his stance, that of a party prepared to reveal the legal advice that he received as to his conduct over one period but not over another, with all the suspicion and adverse inference that that may engender. If the Secretary of State does seek to take an unfair advantage of his partial discovery at the trial, whether as a matter of evidence or argument, the applicants would be entitled to invite the trial judge to re-open the matter and determine whether there should be further disclosure."

56. Mann J noted that Auld LJ said that this authority justified resolving the situation by reference to the principle of "fairness". Factortame is a further illustration of the issue, discussed in Nea Karteria and canvassed in argument on the present application, that one of the most relevant factors in determining whether privilege should be treated as impliedly or consequentially waived is the purpose for which the document in question was deployed and the consequences which may flow if privilege is asserted. Factortame is also a good example. The Court of Appeal extracted a heavy price from the Crown in its attempt to preserve privilege in later legal advice. The Secretary of State was, in effect, forced to disavow any intention of arguing that the more favourable advice given earlier represented the position at a later, and arguably more relevant, time. Moreover, the Court of Appeal recognised that asserting privilege may engender "suspicion and adverse inference".

Implied or consequential waiver

57. One other point to emerge from Nea Karteria and illustrated elsewhere is that the court should treat the scope of waiver as determined in part by the scope of deployment. Again, Factortame provides a convenient illustration. Despite having disclosed the earlier privileged opinion, the Secretary of State was able to avoid disclosure of later privileged opinions by disavowing reliance on the earlier opinion as governing the later position. This shows that the approach to implied or consequential waiver is not necessarily a "once and for all" position where the court is required to determine, as a matter of quasi-historical fact, what the scope of waiver was. To the contrary, the law operates an element of consequential procedural control following deployment of a document where the scope of waiver is a function of the contents of the document and the nature of its deployment: "if you are going to rely on privileged document A to contend that it shows X, the scope of waiver is S(A,X); if you are going to rely on A to contend that it shows Y, the scope of waiver is S(A,Y) – which may not be the same". That is one reason why I have preferred to use the term "implied or consequential" waiver in this judgment.
58. The authorities subsequent to Nea Kerteria are therefore consistent with this somewhat more sophisticated approach to implied waiver which does not depend on mere deployment alone but takes account of (a) the material of which the deployed document forms part and (b) the representations express and implied made by the act of deploying the document. Moreover, they recognize that the potential unfairness of selective disclosure can be mitigated in the possibility that adverse inferences in certain cases may (and I emphasise, may) be drawn if privileged material is not disclosed. Moreover, some of the authorities are consistent with an approach that says that the scope of waiver will depend on what use is ultimately sought to be made of the material deployed which may not be self-evident before trial (see Factortame and to some extent in a different context Lucas v. Barking – see at [38]). Blanket waiver in classes of documents on the footing that part of the class has been deployed is not appropriate save in cases where, for example, a client is suing a solicitor and thereby putting the whole of the relationship in issue (see for example per Lord Bingham CJ for the Court of Appeal in Paragon Finance plc v Freshfields [1999] 1 WLR 1183 at 1188 and per Auld LJ in Facortame cited in Fulham above).

Relevance to specific contexts of waiver

59. In my view, the general approach of the law to the issue of waiver of privilege discussed above should inform the approach taken to the key issues on this application: (a) interpretation of Birss J's order and (b) the extent to which further disclosure should be ordered. Just as the approach to protection of privilege informs the interpretation of provisions said to abrogate it, the approach to treating privilege as waived restrictively and, in particular, only in so far as necessary, should inform interpretation of provisions relating to that subject (see Morgan Grenfell above). That is, in my view, all the more so since, in cases of this kind, the court is concerned not with express but with implied or consequential waiver of privilege which takes place by a combination of a party's voluntary act of deployment and the operation of law in treating other matter as also subject to the waiver. This is a situation in which the limitation of a fundamental right of privilege by the voluntary act of a party should in my view only be implied to the extent necessarily consequential upon that act to achieve the Nea Karteria purposes.
60. It is against that background of approaching the question of waiver of privilege narrowly and specifically in the general law that I turn to Mayne Pharma and the application of the principles to cases concerning experiments conducted for litigation.

(iii) Mayne Pharma disclosure relating to experiments in patent cases

61. Before dealing with Mayne Pharma disclosure, it is necessary to make some preliminary observations about experiments and how they differ from other kinds of documents which may be deployed in litigation. In the case of experiments, identifying something analogous to a transaction and identifying unfair "cherry picking" can be a quite different exercise from that involved in determining whether additional documents in a commercial transaction should be disclosed. That said, experiments are not so different in kind to any other evidence adduced in litigation that an analogous approach to that taken with other documents should not apply.

The varieties of experiment

62. Experiments are as diverse as the science that gives rise to them. The stereotype of a litmus test is far removed from the kinds of experiments normally deployed in patent proceedings. Such experiments can involve multiple stages and can be undertaken to test multiple propositions, including, quite frequently, the validity of the experimental protocol itself. Moreover, experiments in patent cases are often bespoke devised specially for the purpose of the case. Of course, in some situations, experiments can involve a standard say HPLC, NMR or mass spectrometer run using conventional equipment, but in many cases that is not so. Even where standard equipment is used, there may be many options available for testing a particular hypothesis including different ways of setting up and calibrating the equipment, different ways of preparing a sample under test, different test conditions, different approaches to interpretation of the results. Often where a suite of experiments is comparative as in this case, it may not be straightforward to devise a suitable test which may additionally involve identifying and sometimes producing for the first time reference standards, involving positive and negative controls, to ensure that a sample under test will be sensitively and specifically be identified as positive if and only if it has the relevant characteristics; in this case, having been produced by a specific process.
63. All experimental protocols require some degree of validation. Sometimes this comes from material "external" to the experiments or is not in question: proof that HPLC equipment is suitable for separating substances of a given kind may not be provided by the experiments themselves but by the fact that this is a technique known for years. However, sometimes this comes from material "internal" to the experiments where an experimental protocol is validated or calibrated by running a series of samples with a known characteristic through the test to establish a single or multiple controls. Sometimes both. It is frequently the case that test protocols are modified or adjusted to improve their ability to deliver reliable results.
64. Moreover, tests are usually run multiple times. In some cases, these are pure "dry runs" to check that the equipment is working and that the samples can in principle be analysed in that way. In other cases, they are multiple runs of the same experiment with the same sample, different samples, the same control different controls and so on. Some might be "trial runs". In some cases, it can be quite difficult to decide what is a "trial" run and what is a "real" run and, in cases where statistical validation may be required, what "n" is taken to be relevant may be of real significance. Test results are often analysed for statistical significance or even if that is not done formally, error bars are commonly produced from a set of variable results. It is not uncommon for a suite of test data from different individual sub-experiments to be aggregated in some way for the purpose of demonstrating a proposition. The question of what data should be aggregated can be a matter of legitimate scientific debate with arguments over whether doing so compounds or reduces errors, and so on.

Illustrations from the present case

65. The present case presents a good example of the diversity of issues experiments can raise and the number of different stages involved in producing results. I describe this in general terms because of the confidentiality issues involved, without identifying any specific characteristics of the Experiments, samples or results but I have taken into account the greater detail from which this higher level summary is derived in the later application of the law to the facts.

Sample preparation

66. Before getting to the nub of the Experiments themselves, there is a sampling protocol which has to be done and in this case was apparently designed to ensure among other things that the sample is not affected by powder loss, moisture ingress or contamination. There was a compositional analysis by X-ray fluorescence providing composition to a decimal place percentage. There was an analysis of specific surface area indicating some change after further calcination at a specific elevated temperature for a specific period. There was crystallographic analysis, which may or may not be open to different interpretations, as to whether or the extent to which calcination may or may not make a difference that is in any way significant to the characteristics and whether other calcination conditions may result in different characteristics.
67. For the actual Experiments, there were decisions to be made as to whether Raman spectroscopy, TIIE and H₂-TIR were appropriate tests at all for distinguishing routes of manufacture by reference (for example) to the different physical and chemical characteristics of the molecular structure including, specifically, that of the oxygen sub-lattice, the kinetics of oxygen exchange, and the redox characteristics, including the kinetics, of the final material. These are more sophisticated tests than those described in the Patent, which refers to X-ray crystallography for determining whether or not there is a single or twin crystalline phase in the material produced. Assuming that the basic premise of the experimental programme is correct namely that these analyses of the micro-structure and physical chemistry of the product can show up real differences in methods of manufacture in sufficient detail and with sufficient reliability, there are then further questions as to whether they are suitable for revealing specifically whether a method claimed in the Patent has been used or not.
68. Once one gets into the experimental protocols themselves, other issues arise including, at the most basic level, the choice and method of creation of comparative reference samples as well as test sample preparation (such as calcination conditions). For example, in the Experiments, some samples used as comparators for the purpose of evaluating TP1 were made in accordance with the patented process but were additionally made in accordance with a set of other more specific conditions which are not described in the Patent and which are said to be highly confidential (samples S1, S2, S4 and S5 – details of which are themselves confidential). One of these comparative samples is described as "optimized product". Other samples were made using the patented process or processes that differed from the patented process to a greater or lesser degree. Again the extent to which there are or might be differences between these samples other than those specifically claimed in the Patent may be uncontroversial but may be a matter for considerable debate.
69. That may be particularly the case with respect to one of the main comparative samples relied on, S8, which was made in accordance with an earlier process employed by MEL, used before the patented process was developed. In the Experiments, samples S3, S6, S7 and S8 were samples said not to be within Claim 1 of the patent (see e.g. Confidential Witness Statement of Martinez Arias 1, para. 3.69). So questions arising out of the Experiments may include whether TP1 was more or less similar in which test and by reference to what criteria than products said to be within or without the claims of the Patent respectively and the extent to which differences in methods of preparation other than those claimed could account for the differences/similarities.
70. If the action proceeds, these may be matters for trial. I would emphasise that I am not saying that these are points which would inevitably arise although counsel for Neo drew attention to these issues in general terms particularly by reference to the criteria of similarity and dissimilarity employed in the expert reports relating to the Raman spectroscropy. I use them merely as illustrative of the kinds of points which could arise on the substantial range of Experiments of this kind, which produced data filling multiple lever arch files.
71. Examination of the protocols, the expert evidence relating to them and the comments from the peer review experts recorded in the mass of material describing how the protocols were developed indicates that there were real choices to be made and that, for some of the tests, differences could be expected as a result of the type of preparation method employed and from changing parameters within the same method. It is also evident from the statements of the experts who conducted the tests that there was a dynamic exchange between a number of individuals providing technical input involving peer review experts and lawyers in developing and validating the protocols.
72. One characteristic of the evidence before the court at this stage including the witness statements of the experts and Bould 5 is that there is, if anything, an unusually high degree of transparency as to how the experimental regime was developed, what was done, how it was refined and in some cases, candid discussion about the issues that may arise with respect to it. It therefore provides a useful insight into the issues that often arise in these cases.
73. The upshot of this brief review is that trying to evaluate, in the terms of the existing case law, what exactly is "workup" or "preliminary" for what and how the boundaries between the various parts of the programme as a whole should be drawn is not straightforward.

Procedural aspects of experiments and the avenues for challenging them

74. Next, the procedural rules governing experiments deserve mention. Experiments conducted for the purpose of litigation fall into two categories: those done and deployed pursuant to an order for service of a notice of experiments and those which are deployed but not subject to that procedure. Experiments falling into the latter category tend to be those conducted for preliminary purposes (such as preliminary injunctions or applications, as in this case, for permission to serve a defendant out of the jurisdiction). In the case of experiments subject to a notice of experiments, there is a procedure established under the CPR Practice Direction under Rule 63 of which the primary provision is as follows:

7.1  A party seeking to establish any fact by experimental proof conducted for the purpose of litigation must, at least 21 days before service of the application notice for directions under paragraph 7.3, or within such other time as the court may direct, serve on all parties a notice–

(1) stating the facts which the party seeks to establish; and

(2) giving full particulars of the experiments proposed to establish them.

75. That does not apply to experiments not done under the notice procedure (as is the case here for the Experiments) but commonly parties will state what facts the experiments are sought to establish and there is no reason why the principles relevant to Mayne Pharma disclosure should not be the same.

Requirement to state facts to be established

76. As to this requirement, the person seeking to adduce experimental evidence has to state in its notice what facts it seeks to establish by the experiments. In some cases, even with experiments subject to the notice procedure, in addition to explicitly stated facts, the party relying on them may seek to use the experiments to prove facts which may not be explicit but which may be important or essential to the evidential value of the experiments. Experiments sought to establish that practising the teaching of the prior art inevitably leads to something falling within the claims and therefore invalidates the claims of an asserted patent by anticipation are paradigm cases. Deployment of such experiments can involve two propositions: (i) that undertaking the prior art inevitably and straightforwardly produces a product falling within the claims and (ii) that the experiment or the precise conditions under which the experiment is run would inevitably be chosen by a person skilled in the art from the prior art. There may be further implicit facts involved. For example, in the case of an experiment which is run multiple times but which gives variable results, it may be implicit in deploying a given set of runs that the variation is not significant (albeit not always in a formal statistical sense) and that the set of data which has been provided for the purpose of the notice of experiments is fairly representative of all results obtained from experimental runs of that kind. Another example may be that the criteria employed to determine whether the experiment has "worked" are appropriate criteria. There are many other possibilities.

Right to repetition and to serve responsive notice

77. For experiments subject to the notice procedure, an opposing party has a right to have the experiment repeated and to witness repetition (see CPD Practice Direction 63, para. 7.2). An opposing party also has a right to undertake and seek to adduce evidence of its own experiments, including experiments in reply to those done by the other side. That also applies to experiments not subject to a notice done for preliminary purposes.
78. Undertaking and witnessing the repetition of experiments is often a costly exercise. If it can be shown at the outset that the experiment relied on has not (for example) been sufficiently validated, an opposing party may reasonably decide not to undertake that exercise but may, instead, criticise the experiments as not being probative for that reason. There are therefore good reasons for having complete information about this at as early a stage as possible.
79. However, conversely, precisely because experiments are subject to challenge and often formal repetition, they are very different from (for example) a single document put forward as evidencing the nature of a transaction as a whole. Even if full data relating to an experiment is not put forward, there is often nothing to stop an opposing party running the same experiment and showing that it produces different results. Experiments are, to this extent, inherently more transparent than a good deal of other evidence.

Court control of experiments

80. The courts have in the past given experiments in patent cases mixed reviews. On the one hand, in Electrolux, there was extended discussion of the often limited value of experiments ("…experiments frequently prove to be a waste of time and effort" per Laddie J at [1996] FSR 595 at 613). On the other in Mayne Pharma, Pumfrey J referred at [15] to experiments as potentially providing "a fixed point against which the experts may themselves be assessed". Whatever view one takes of experiments in general, in order to have value, experimental evidence should be as solid as possible and criticisms of protocols and their significance should ideally be revealed and ironed out well before trial.
81. The court will exercise control over the admission of experiments (see CPD Practice Direction 63, para. 7.3): the freedom of a party to rely on any experiments it may have chosen to do is not unlimited. The court can refuse permission to rely on experiments if they are not likely to be of sufficient probative value. Again, if it can be shown that the experiments are unlikely to have significant value because the experimental protocol employed has not been adequately validated in important respects that may affect the court's decision as to whether evidence of them should be admitted at all given their value.
82. The consequence of all these points about experiments, substantive and procedural, is that what it is fair and unfair to reveal or disclose – and what constitutes "cherry picking" has to be seen against a different background from that which arises in many other cases concerned with waiver of privilege.

Mayne Pharma disclosure

83. Patent cases are no different to any other cases in that documents recording activity undertaken for the purpose of litigation attract privilege. Until they are deployed, they remain privileged. Once deployed, the question arises as to the extent to which, if at all, the effect of doing so is also to waive privilege in any other documents or material. The answer given in patent cases is in line with that in other cases although the patent case law has not always referred to all of the general authorities. In patent cases, as in any other, the opposite party and the court must have the opportunity of satisfying themselves that "what the party has chosen to release from privilege represents the whole of the material relevant to the issue in question". The problem arises in patent cases because, as in other cases, that proposition is itself somewhat imprecise: how are the boundaries of that which is relevant to the issue in question to be set?
84. In Mayne Pharma the claimant's case was that the patent was anticipated because the "inevitable result" of putting the prior art into practice was that one obtained the patented invention. A notice of experiments in support of its "inevitable result" anticipation case was served. There was then a dispute as to whether the choices made by the claimant in preparing the experimental protocol were inevitable. The protocol had been given to the expert and it had been presented to him as a fait accompli by the claimant's lawyers. He expressed the opinion that it was a fair and reasonable way of carrying out the process of the prior art. The defendant patentee sought disclosure of any (so-called) "work-up" experiments conducted in relation to the experiment adduced by the claimant. The claimant resisted on the grounds of privilege.
85. Pumfrey J considered the two conflicting decisions Honeywell Ltd v Appliance Components Ltd (unreported, 22 February 1996, Jacob J) and Electrolux Northern Ltd v Black & Decker [1996] FSR 595 (Laddie J). These judgments were not of complete assistance for a number of reasons. First, they concerned different experiments which had not been deployed by the party concerned and whether they should have been disclosed. In the case of Electrolux, these were experiments which had been abandoned. Second, in neither case was the general issue of privilege and its waiver argued and the case law relating to this was not addressed at all. Third, the factual context of the cases was very different to that of Mayne Pharma. In Mayne Pharma Pumfrey J was addressed on the Nea Kateria principles (albeit not apparently on all of the case law cited on the present application) and discussed Honeywell and Electrolux. He observed that the former case was concerned with a wider question than mere workup experiments and the latter case was concerned abandoned experiments. It is noteworthy that Pumfrey J rejected the suggestion that there was invariably no obligation to disclose additional material at all where an experiment had been deployed.
86. Pumfrey J applied the Nea Karteria principles to the case before him (see [18]). However, he emphasised that he was concerned only with the narrow question of workup saying that service of the notice of experiments:

"waives the privilege in work up experiments for that experiment"

on the Nea Karteria basis (emphasis added). He did not purport to extend his reasoning or conclusion to other experiments.

87. He noted at [20] that Nea Karteria: "strongly supports the argument of waiver, at least in respect of material directly associated with or underlying the evidence sought to be adduced." He said, at [21], that it is observed in Hollander on Documentary Evidence at 17-45 and following paragraphs that Mustill J's phrase "relevant to the issue in question" controls the extent of the waiver. Pumfrey J said:

"The scope of the waiver in any particular case is a question of fact. I have no difficulty in saying that the waiver in the present case which is effected by the giving of the notice of experiments in relation to an experiment conducted on the instruction of the lawyers for the purposes of litigation, extends to the workup of the particular experiment forming the subject matter of the notice. Whether it extends further I am more doubtful and I will not express a view."

88. Pumfrey J concluded at [23], that given that the purpose of the experiments was to show anticipation by inevitable result that the workup experiments had to be disclosed because privilege in them had been waived. Pumfrey J defined "workup" as follows at [1]:

"…preliminary investigation leading to the experiment forming the subject matter of a notice of experiments put in by Mayne."

89. While that definition was sufficiently clear for that case, its practical application in other cases can give rise to difficulty, particularly in the case of experiments which have multiple components or stages and may have been run multiple times in different ways.
90. In my view, in the light of the restrictive approach from the general authorities and in the light of Mayne Pharma itself, the disclosure required to be given does not extend beyond materials recording preliminary investigation leading to the particular experiment which is deployed in evidence and does not extend to other parts of an overall experimental programme even if the design of the experiment in question may have drawn on earlier experiments. That narrow approach is consistent with the authorities above which treat implied waiver of privilege restrictively in the general law. It also addresses the concerns expressed by Laddie J in Electrolux. I should say that I am not satisfied that even Jacob J in Honeywell was contemplating that there should always be disclosure of experiments as broadly as he may have been taken by Laddie J to have suggested. In Honeywell, Jacob J was primarily concerned with the question of whether there was an analogy with cases involving survey evidence in trade marks or passing off and whether an opposing party should be told whether other experiments had been done.

General points

91. The lesson from the above discussion is clear. There is often no wholly incontestable way of treating one aspect of an experimental programme as completely separable from another. The present case is a good example of such a situation. Some aspects of an overall programme in a complex case may involve deciding what experiments to undertake and what criteria to employ. Some aspects involve undertaking experiments. Some involve interpreting the results. In such a case it can be much harder to identify even what would constitute unfair "cherry picking". Cases before this court may be unusual in which the experimental programme is completely straightforward. In many cases of which this case is, again, an example, it is rarely as easy to say whether there would be real unfairness in preserving privilege in part of a complex multifactorial programme of development of litigation experiments as it is in the simple case where privilege is asserted in test runs 1-9 of a standard test when only test run 10 is deployed. In a complex case, what is "workup" and what is not and what is "directly" related to experiments ultimately deployed may be terminology over which there can be such legitimate disagreement that an order to provide that gives rise to over- or under- disclosure.

Straightforward Mayne Pharma cases

92. In my view, there are two kinds of case in which the Mayne Pharma approach can be clearly and easily applied.

"Inevitable result" cases

93. The first are "inevitable result" cases. These are a species of a genus of cases in which experiments are deployed as one part of what is really a two-part evidential submission: (a) that experiments of that kind would inevitably be selected and undertaken in the manner alleged (b) that the experiments inevitably have the results alleged. The experiments actually deployed go largely to the latter. In those cases, fairness requires information to be provided relating to the selection and conduct of the experiments and the manner in which they were carried out. That was the rationale for the actual decision in Mayne Pharma itself. It was also the rationale for a provision in an order relating to experiments made by agreement between the parties by Arnold J in a recent case discussed at the hearing involving a question of repetition of the teaching of the prior art (Anan Kasei Co Ltd v. Molycorp Chemicals and Oxides (Europe) Ltd, Order of 8 February 2017). The court ordered, following the parties' agreement, in effect, that all technical communications relating to design and performance of the experiments and work up experiments should be provided.
94. In those situations, one of the reasons for such a provision is that the selection and conduct of the experiments (and not just the results) is one of the key issues in the case. Indeed, a court might, in such a case, be justified in refusing permission to rely on the experiments altogether if suitable additional information about those matters was not provided, on the footing that, absent such information, the experiments taken alone are of insufficient probative value to the issue of whether undertaking the prior art would inevitably produce the patented invention.

"Completeness of data" cases

95. Second, in cases where an issue is whether an experiment is repeatable or exhibits significant variation in result, it may be critical to have information about that issue. Putting forward only part of a data set may lead to unfairness of the kind that Nea Karteria was addressing. In those circumstances, again, a court may be amply justified in not permitting a party relying on a set of test runs to deploy just the runs that suit its case.
96. In both the previous examples, the unfairness in selective deployment is clear. There is also in each a close analogy with the "transaction/whole document" cases in that the whole experimental programme can be regarded as analogous to the transaction in an "inevitable result" cases and the whole of the data set regarded as analogous to a whole document in "completeness of data" cases.

Other cases

97. However, other cases of which the present one is an example, are less straightforward. In cases where there may be multiple points potentially in issue albeit not clearly so at an interim stage, some of which relate to the data generated, some of which relate to the choice of conditions or protocols (especially where protocols have been refined over a period), it may be less clear that earlier or related material can be properly described as "workup" or "directly" related to the particular experiment deployed or that there would be any cherry picking in not disclosing it.
98. In such cases, a more cautious and focussed approach is required both to the question of scope of waiver of privilege and to the question of whether disclosure (standard or otherwise) should be ordered or whether other instruments are better suited. In the light of the discussion above, I do not believe that the court should treat such a situation as one of implied or consequential waiver of privilege in a quantity of precursor or surrounding material, on the basis that later material relating to experiments has been disclosed and can be described as "relevant", "directly" relating to or "underlying" in some sense. That is so even where it may be possible to find the answers to some potential questions about the validity of the material actually deployed by rummaging around in earlier material. That thinking is reinforced by the modern approach to disclosure and the move towards more bespoke solutions to the provision of information of which disclosure is only one way.

(iv) Disclosure and other ways of obtaining information

99. English law, like other common law systems, has long regarded the rights of litigants to subject their opponent's case to destructive testing both in trial and pre-trial as a vital part of the legal process. One of the key weapons for doing this has been the requirement to produce adverse documents in general and, recently, more limited, standard disclosure. However, it has also been recognised that disclosure can be an expensive and disproportionate sledgehammer even if subject to the limitations of standard disclosure pursuant to CPR 31. More sophisticated and surgical instruments have been introduced or refined to enable parties to find out critical information about an opponent's case well before trial. The Patents Court and its judges have often led the way in trying to limit disclosure (see the dissenting judgment of Jacob LJ in Nichia v. Argos referred to below).
100. That approach has found greater favour more generally. In his Foreword to Matthews & Malek, Lord Briggs summarises the position as follows:

"…there is a growing move towards earlier and more flexible case management…with an altogether more bespoke approach to the ambit of disclosure, rather than the 'one cap fits all' standard disclosure too easily adopted in the past as an unthinking default approach. This should assist in finding that happy but elusive compromise between attaining justice and proportionate expense on a case by case basis".

The options for disclosure

101. As regards disclosure, the need for more focussed and less costly options to standard disclosure has led to CPR Part 31 rule 5. This rule was discussed in a patent context by Birss J in Positec Power Tools (Europe) Ltd & Ors v Husqvarna AB [2016] EWHC 1061 (Pat) (10 May 2016) in a section which merits citation in full. He said:

"20. I turn to CPR Part 31 and rule 31.5 in particular. By r31.5(2), unless the court otherwise orders, paragraphs (3) to (8) of r31.5 apply to all multi-track claims other than claims including a claim for personal injuries. There is no reason to "otherwise order" in this case. Rules (3) to (6) provide for disclosure reports, EDQs and discussions before the CMC. Save for the point about the effect of Husqvarna's Disclosure Report, these paragraphs are not germane to the issue before me. The key provision is r31.5 (7) which provides as follows:

(7) At the first or any subsequent case management conference, the court will decide, having regard to the overriding objective and the need to limit disclosure to that which is necessary to deal with the case justly, which of the following orders to make in relation to disclosure –(a) an order dispensing with disclosure;(b) an order that a party disclose the documents on which it relies, and at the same time request any specific disclosure it requires from any other party;(c) an order that directs, where practicable, the disclosure to be given by each party on an issue by issue basis;(d) an order that each party disclose any documents which it is reasonable to suppose may contain information which enables that party to advance its own case or to damage that of any other party, or which leads to an enquiry which has either of those consequences;(e) an order that a party give standard disclosure;(f) any other order in relation to disclosure that the court considers appropriate.

21. Two things emerge from this. First is the reference to the overriding objective and the need to limit disclosure to that which is necessary to deal with the case justly. This helps to focus the court's mind on the task to be undertaken. Second, and critically, is that the effect of this provision is that standard disclosure is one of six options. Counsel for Husqvarna submitted that this meant that standard disclosure was not the default option any more. I agree. The Chancery Guide (paragraph 17.35) makes the same point. As the Guide states, careful consideration should be given to the alternatives to standard disclosure.

22. In Nichia v Argos Rix LJ was concerned not to introduce a "blanket prima facie rule against standard disclosure". The amended form of CPR r31.5(7) does not go as far as introducing such a rule but in my judgment it does have the effect of providing that there is no longer a blanket prima facie rule in favour of standard disclosure.

23. It is certainly not the case, and counsel did not suggest, that the fact that no disclosure at all is one of the options (at (a)), means that a party seeking an order for some disclosure has a difficult burden to discharge. That would not be appropriate. Nevertheless any court deciding which of the options to adopt has to have some basis for that selection. Rule 31.5 expressly maintains disclosure as a feature of key importance to the interests of justice in civil litigation but the rule replaces a single approach (standard disclosure) with a series of options. The current form of r31.5 means that the selection of the correct approach is something governed not by a prima facie rule but by the overriding objective to deal with cases justly and at proportionate cost.

24. In this case the question of whether to order standard disclosure is in substance the same as the question of whether to order disclosure on obviousness under r31.5(7)(c), i.e. on an issue by issue basis. That is because of the nature of patent disputes. It is common ground that disclosure on infringement should be given. The fact that it is being given by the mechanism of a PPD under PD 63 paragraph 6.1 does not alter that. Stated at a high level of generality there are only two other issues in the case, validity and a point about whether Positec has a sufficient commercial interest to bring the declaratory proceedings. As to the latter, they plainly do. The point was resolved at the CMC and is no longer live. So the only other issue is validity and within that the only issue on which disclosure might matter is obviousness. There is no allegation of prior use, this is not an inevitable result case nor is there any allegation of insufficiency. So issue based disclosure on obviousness would be the same as standard disclosure.

25. In either case, the question is whether it would be in accordance with the overriding objective to make such an order. Proportionality will be important and two major considerations are the likely probative value of the material which could be produced and the cost of doing so, putting that cost into the context of the proceedings as a whole (CPR r1.1 (2)(c) and 1.4(2)(h)). In the end the question comes back to fairness and the interests of justice."

102. One lesson from that case is that the court should consider whether it is right automatically to reach for the tool of standard disclosure rather than considering other options. Another is that proportionality and the interests of justice are central. One of the reasons for limiting disclosure on obviousness is that there are other primary ways for a party to establish its case but the principle applies more generally. It is less unfair to refuse standard disclosure if an opposing party has a range of other options for addressing the issue including, of course, more focussed specific disclosure where a clear need is shown. The court should be on a constant lookout for better less costly ways of conducting litigation and taking approaches consistent with fair resolution of disputes which are least intrusive and least costly.

Other principles relating to disclosure

103. I accept the submission of counsel for MEL that CPR 1.1 and 1.2 require the court to deal with cases justly and at proportionate cost and that this is relevant both to what disclosure should be ordered and how it should be provided. Equally, in the case of disclosure, the effect of the overriding objective is that: "The overriding principle is that disclosure should be restricted to what is necessary in the individual case." (See notes in White Book (2017 Vol 1, page 919, para. 31.0.2). Ensuring that the parties are on an equal footing (CPR 1.1(2)(a)) is also relevant to some extent although it has lesser application if only one side puts in experiments.

Other options for finding out information

104. It is also important to remember that the CPR contains other provisions which enable a party to find out information about an opponent's case. Two examples suffice:

(a) Requesting further information

105. CPR Part 18 provides:

"18.1

(1) The court may at any time order a party to –

(a) clarify any matter which is in dispute in the proceedings; or

(b) give additional information in relation to any such matter,

whether or not the matter is contained or referred to in a statement of case."

106. Information given or ordered pursuant to that provision must be verified by a statement of truth. That general provision can be tailored to the specific requirements of a case and can enable a party to find out, for example, facts about experiments such as whether the other side is relying on experimental evidence to show that a particular protocol is reliable as well as for the results of undertaking the protocol, whether the protocol used has been validated in particular ways and if not why not, what criteria have been used and so forth.

(b) Putting written questions to experts

107. The CPR also contains specific rules enabling a party to put written questions to an opposing party's experts. These are provided in CPR 35 as follows:

"35.6

(1) A party may put written questions about an expert's report (which must be proportionate) to –

(a) an expert instructed by another party; or

(b) a single joint expert appointed under rule 35.7.

(2) Written questions under paragraph (1) –

(a) may be put once only;

(b) must be put within 28 days of service of the expert's report; and

(c) must be for the purpose only of clarification of the report,

unless in any case –

(i) the court gives permission; or

(ii) the other party agrees.

(3) An expert's answers to questions put in accordance with paragraph (1) shall be treated as part of the expert's report.

(4) Where –

(a) a party has put a written question to an expert instructed by another party; and

(b) the expert does not answer that question,

the court may make one or both of the following orders in relation to the party who instructed the expert –

(i) that the party may not rely on the evidence of that expert; or

(ii) that the party may not recover the fees and expenses of that expert from any other party.

108. These rules provide additional ways in which a case supported by experts may be interrogated and challenged prior to trial including situations in which an expert puts forward experiments. There are, of course, other means by which information may be obtained including provisions for statements of case and on the other side, rival experiments. A party served with experiments whether pursuant to a formal notice or otherwise is not bereft of means with which to demonstrate, using information obtained from the party deploying the experiments or otherwise, that they are open to question. In my view that is relevant both to the extent to which fairness requires the court to treat privilege as having been waived following Nea Karteria and to the approach the court should take to requiring disclosure as such.

Summary

(i) Limited approach to waiver of privilege

109. The combination of (i) the general law of privilege, (ii) the approach to scope of waiver generally and (iii) the approach taken in the patent case law relating to experiments suggests that the court should adopt a relatively cautious and restrictive approach to waiver of privilege in material relating to experiments in cases other than clear ones of the kind identified above. Neither the general law nor the specific patent case law provides a warrant for broad and general disclosure of all information about earlier experiments on which the experiments in question may have been based if a later experiment is deployed. To the contrary, waiver of privilege in material not deployed should only be treated as having taken place to a limited additional extent and only in so far as necessary to satisfy the specific objects of the law relating to waiver set out in Nea Karteria set out above.

(ii) Issue-specific information

110. The law relating to disclosure, including the provisions I have set out above, suggests that the court should, in appropriate cases, think about the issue somewhat differently from the way in which it has done in the past. It should consider more whether it is appropriate for an opposing party to provide "issue-specific information", rather than (for example) general or even issue-based disclosure.
111. That may be particularly the case with experiments. If a party putting them forward is invariably required to describe the steps in development in the meticulous detail that characterises the evidence in this case, there is a real risk of over-disclosure and significant work going to waste. Moreover, it can be that the real issues in the case are not resolved by disclosure as such. For example, if a question is why a particular criteria were employed and whether they were justifiable, the real answer may lie in the view of an expert not a disclosure document recording the implementation of that choice. The parties and the court should evaluate whether better ways exist of obtaining the information in question such as one of those identified above.
112. In some cases involving experiments, the issues are clear, cherry picking is obvious and Mayne Pharma disclosure remains the only fair way of addressing this. That is not so in every case.
113. Additionally, the exercise of providing a full account, whether by way of a lengthy list or by way of narrative with exhibits, also risks a party, in giving Mayne Pharma disclosure implicitly waiving privilege still further and consequential upon providing that disclosure when it is said (as it is here) that the disclosure of part of the communications between lawyers and experts in compliance with Birss J's order gives rise to the need to disclose even more. That can lead to a seemingly endless request for the absolutely complete picture even in cases where there is limited reason to suppose that anything of real relevance to the case is to be found in the parts not provided.
114. An issue-based information approach means that a party seeking information can formulate specific questions about what exactly it needs to know and the party relying on experiments can decide how to answer them. In a case between undertakings operating in the same industry which is usually the case in patents, there is unlikely to be an inequality of intellectual arms to such an extent that the opposing party does not know what focused questions to ask. In my view, it is desirable for the court to take a more holistic view as to what the interests of justice require in an individual case rather than determining these questions on the basis of the more contestable foundation of implied or consequential waiver of privilege in large categories of a party's documents.
115. Finally, if the matter is approached in that way, greater and more specific attention can be given to the precise scope of implied or consequential waiver in the materials provided and for what purpose they are deployed. In a situation in which, for example, an expert is asked what informed his or her choice of a particular criterion used in an experiment and gives as one reason some data obtained from earlier experiments, the expert may wish to refer to those earlier experiments. A party could then decide whether to waive privilege in them to enable the expert to give a full opinion. If privilege was nonetheless asserted, even though the expert thought the earlier material was relevant, the court may be justified either in treating it as waived pursuant to Nea Karteria or may be justified in drawing the adverse conclusions referred to in Factortame or may take another course to ensure procedural fairness.
116. Drawing these threads together, the approach this court should take to interpretation of Birss J's order and what, if any, further disclosure should be required in this case must take account of the fact that the law, general and specific:

(1) Does not allow the absolute protection of legal privilege to be easily eroded;

(2) Limits implied or consequential waiver to situations of unfairness and selectivity which has regard both to the material deployed and the purpose for which it is deployed;

(3) Ensures that the scope of implied or consequential waiver is examined and determined at a relatively high level of specificity;

(4) Encourages bespoke use of the range of civil litigation tools, of which disclosure is one, to obtain more information about and enable challenge of proposed evidence, having regard to proportionality, necessity and cost while retaining the ability to test evidence comprehensively;

(5) Encourages the provision of issue-based information, not necessarily disclosure of documents in which privilege is taken to have been waived.

117. I apply that thinking, which is intended to summarise the position that the modern law has reached both on waiver of privilege in material relating to experiments and on the approach to be taken to disclosure, in resolving the issues on this application.

THE CATEGORIES OF DOCUMENTS IN THIS CASE

118. Turning to the categories of documents in issue.

(i) Documents relating to tests in 2013/2014.

119. In 2013/2014, prior to undertaking the Experiments, Profs Boghosian and Efstathiou carried out other tests on third party REMO products, comparing them with samples produced by MEL, for the purpose of some contemplated litigation. The tests included H₂-TIR, TIIE and Raman spectroscopy and other types of tests. This work is mentioned in Ms Bould's evidence. When it came to designing the experiments for this case, the experts started from the protocols used in those earlier tests (see evidence of Ms Bould, 8th statement paras.18-20).
120. MEL contends, first, that these documents are not within the scope of the order because the earlier work is not "directly associated with or underlying" the evidence about the Experiments and that they have a tangential connection with them. Second, it is said that these are not workups of the experiments but altogether separate experiments. Third, it is said that MEL has not deployed any documents relating to the earlier tests so as to waive privilege under the Nea Karteria principles. Fourth, it is said that disclosure would be disproportionate and unnecessary: it would be very costly in terms of time and resources and would not be proportionate. Fifth, it is said that this would include documents which may be confidential to third parties. These points are elaborated in evidence of Ms Bould (8^th statement para. 21).
121. Neo contends, first, that the early experimental work may have involved techniques other than those that comprise the Experiments and that they are entitled to understand why these techniques were chosen and other techniques discounted. It points out that the other techniques identified clear differences between different REMO samples which could have countered any inference of infringement relating to the Neo products. It is said that the information will shed light on the limitations of the Experiments and the reasons for choosing their specific protocols. This information is therefore directly related to or underlying the evidence relating to Experiments. Second, the fact that no infringement proceedings were brought against third parties should, it is submitted, entitle Neo to establish the criteria by which MEL was able to infer infringement. Neo contends that it is important that the basis for selection of the analytical techniques and the parameters on which an inference of infringement were based should be disclosed particularly since the differences between the various products were not invariably very marked. Third it contends that confidentiality is no obstacle to requiring disclosure and that the question of whether privilege has been waived has already been decided by Birss J if indeed this material otherwise falls within para. 10 of his order because that order defines the boundaries of waiver of privilege. Fourth, it is said that in the context of the present proceedings, disclosure would not be disproportionate and that the estimates put forward by MEL of what would be required are unreasonably high. Moreover it is said that given the substantial costs already incurred in this action and the value, such disclosure is proportionate, especially having regard to the significance MEL seeks to attribute to these Experiments.

Discussion

122. It should be emphasised that the terms and the purpose of Birss J's Order required no more than standard disclosure of documents "directly associated with or underlying the evidence adduced under paragraph 6" of that order. In my judgment, the materials underlying the actual evidence have been provided by the extensive materials referred to above. The additional materials referred to under this head are not, in my judgment, "directly" associated with or "underlying" the evidence in question in the sense in which that term was used in Birss J's order and construed in the light of the principles set out above which treats Mayne Pharma disclosure relatively narrowly.
123. It is true that the material may include documents which may explain somewhat more why the particular evidence in question (rather than different Experiments evidence) were adduced. However, having regard to the principles relating to waiver of privilege set out above and, in particular, the restrictive approach to waiver in the general law and in Mayne Pharma, I do not consider that a sufficient case has been made out that the court should treat privilege as having been impliedly waived in that other material sought nor would there be any evident unfairness (in the Nea Karteria sense) if this material was not provided generally. In my view in a case of this kind it is not necessarily unfair for a party not to provide disclosure of all precursor experiments on different products especially when the experiments are likely to be extensively analysed by expert evidence, should the case continue.
124. In the terms of the general law on waiver of privilege, these are not, taken as a whole, documents analogous to those which form part of the same transaction. Nor do any additional principles of fairness require such disclosure having regard to:

(1) The materials which have been provided already;

(2) The ability of Neo to challenge the Experiments in various other ways;

(3) The ability of Neo to find out more information about the Experiments by more focussed requests either for more information or for more specific disclosure.

125. I therefore do not consider that further disclosure should be given on the basis that this is not material contemplated by Birss J's order interpreted in the light of the principles set out above.
126. I am also not satisfied that the further standard disclosure of this material would be necessary or proportionate either given the likely cost and the value to be gained from it, having regard to the matters set out above, including the various other options for obtaining additional specific information.

(ii) Legal advice concerning the Experiments

127. The next category of which disclosure is sought is legal advice given by Pinsent Masons to MEL and, in particular, discussion of the results or protocols relating to the Experiments or the conclusions that may be drawn from them. There are two sub-categories here: first, documents which in principle may be disclosable but where, as a substitute for disclosure, there is a narrative account and second other additional documents of which no information has been provided at all.
128. MEL contends, first, that discussions between MEL and its lawyers about the experiments are not "directly associated with or underlying" the Experiments in the sense that Birss J intended. It points out that such communications were not included in Neo's solicitors' second letter of 7 June 2017 or were sought in the discussion in the second statement of Mr Jenkins at paras. 19-21. Second, it contends that privilege has not been waived because the legal advice has not been deployed under the Nea Karteria principles. Third, it contends that disclosure would be disproportionate, unnecessary and unfair and that Neo has not explained why it would be necessary or proportionate for such advice to be disclosed. Fourth, it contends that it would be unbalanced for MEL to have to disclose its confidential discussions with its lawyers on the subject of the Experiments but for Neo not to have to.
129. Neo contends, first, that any discussions of the results or protocols relating to the Experiments or the conclusions to be reached from them clearly falls within the scope of Birss J's order since it is material directly associated with or underlying the experiments pursuant to Mayne Pharma and that it would not be unduly burdensome to have to provide it, especially having regard to the fact that substantial "narrative" material has been provided. Second, it contends that in the light of Birss J's order the fact that this material may be privileged does not make it non-disclosable and that it would be fair and proportionate to provide disclosure given that lawyers were closely involved in developing the protocols. In relation to the narrative, Neo contends that it is important to have the "raw" documents (attendance notes and file notes and so on) and that this is a situation akin to that in Mayne Pharma where the opposing side should have an opportunity to check that the experts have been fairly instructed. Third, Neo points to the fact that the solicitors were very closely involved in developing the protocols (as can be seen from the documents from them exhibited already and the references to these discussions in the experts reports and Bould 5). Neo contends that it is therefore important to have the full picture, not just in narrative form.

Discussion

130. In my judgment, although, on one view, legal advice relating to the experiments is "directly associated" with them in the sense that it may be directly about them, I do not consider this to be what Birss J meant or was contemplating by his order. Taken in context, and having regard to the principles set out above, that order was directed to requiring disclosure of technical documents directly associated with the Experiments not legal advice relating to their validity.
131. Moreover, for the reasons given above, I do not regard an obligation which requires disclosure of privileged material "pursuant to Mayne Pharma" as requiring disclosure of the legal advice relating to experiments and do not consider that privilege has been waived in them. That would go well beyond what Mayne Pharma or any of the other authorities relating to general disclosure contemplated and I was not referred to any authority in which this had happened. The position of instructions to experts in the "inevitable result" cases is very different and I note that, in the Anan case, legal advice relating to what the experiments showed or their value was not ordered to be disclosed (albeit by agreement).
132. These documents were not therefore within the scope of Birss J's order properly interpreted. I would also have rejected this category on the basis of lack of necessity and disproportionality on the footing that the extended narrative provided gives ample scope for Neo to challenge the Experiments in numerous ways. I am not satisfied that filling in omission of raw documents to which attention was drawn (e.g. in paras 43 and 47 of Neo's skeleton) is likely to lead to sufficient incremental advantage to justify additional disclosure. It is not, to my mind, sufficient simply to say that some of this material may be "potentially relevant" to the question of whether the Experiments are reliable. For the reasons set out above, there should be greater focus as to exactly what additional information is likely to be of real incremental significance for the case. This is a case far removed from a situation, discussed in the cases, in which lawyers put a protocol to an expert and ask "is that all right?". As the evidence shows, the experts have devoted considerable effort and thought to developing and criticising their own thinking about the appropriate protcols to use and have provided extensive explanation of it. I am therefore not satisfied that the views of solicitors on these protocols, even if disclosed would take this aspect of the case further as a matter of generality.

(iii) Draft expert reports

133. The next category is draft expert reports and particularly comments or alterations indicative of the process that led to finalisation of the Experiments. These are said to go to the heart of what they show and their validity. MEL makes much the same points as for the previous category as to why there should be no further disclosure and Neo makes similar points as to why there should be. In particular Neo draws attention to a question mark as to whether the various comments from Professor Boghosian had been adequately appreciated by Professor Martinez Arias.

Discussion

134. I do not consider these draft expert reports to be within the scope of Birss J's order either for reasons similar to those given above. They are not "directly associated with or underlying" the evidence in question, as was implicitly recognised in the evidence before Birss J on the application for Mayne Pharma disclosure by the omission of reference to this category. I consider that, if anything, Mayne Pharma suggests that disclosure of such material is not to be given on this basis. Although in that case the court was not considering specifically drafts of reports relating only to experiments. I do not consider that privilege has been waived in these drafts.
135. This is not a situation in which one of the points in issue is whether the experts went straight to the protocols relied on (as it would be in an inevitable result case). As I understand the position, MEL is not deploying the Experiments for that purpose so there is no reason to provide such disclosure on that basis either. I also agree with counsel for MEL's submission that disclosure would be disproportionate and unnecessary for similar reasons to those set out above.

Overall conclusion on the categories

136. I therefore reject the submission that additional disclosure in these categories should be given which makes it unnecessary to consider whether specifically standard disclosure of this material should be given by list or in some other way.

Manner of giving disclosure

137. Had I concluded that further disclosure was required in principle, I would have considered that an alternative less costly approach to giving disclosure would have been appropriate and that provision of a further narrative concerning the historical documents and the provision of legal advice, exhibiting the documents referred to would have been a suitable alternative and that Birss J's order should be varied to that end. Since I am of the view that this material does not fall within the scope of Birss J's order, that issue does not arise. There is no application before the court by Neo for further disclosure not falling within the scope of that order.

Alternative approaches

138. In saying that, I have some sympathy for Neo's position. In particular the second defendant has been brought into these proceedings in this jurisdiction on the basis of Experiments whose validity may be open to question in a number of ways and, having provided a good deal of information about them, it may be thought inappropriate simply to put the shutters up on any more.
139. However, in my view, this is a case in which if Neo wishes to have further information about MEL's case and the Experiments including some of the matters referred to in its skeleton, it should make specific requests for specific information identifying the precise technical issue to which that information goes. That may be done in one of the ways mentioned above.
140. If that is done, I do not rule out the possibility that a focussed request for further disclosure of specific documents relating to the Experiments may be justified. It would be appropriate to consider any claim to privilege with respect to a specific document in the context of such a request at which proper exploration could be made as to the utility of disclosure and whether the refusal to provide a specific document would constitute unfair selectivity, in the light of the other material provided. That may need to take account of the purpose of deployment. For example, if MEL said that the Experiments should be regarded as valid because of specific information gleaned about the protocols of the earlier experiments but then refused to provide full details of these, relying on privilege, the court may be justified in concluding, at that stage, that privilege should be treated as waived in the relevant parts of the earlier material or it may have one of the other consequences identified above. I therefore do not consider that it would be appropriate to make an order which comprehensively "franked" MEL's disclosure and provided (in effect) a bar on Neo making focussed requests and requesting specific documents relating to specific issues where a real need for them was shown.
141. In summary, I do not regard there as being an implicit or consequential general waiver of privilege in the classes of documents of which further disclosure is sought but without prejudice to any argument that MEL should be taken to have done so with respect to specific documents relating to specific identified issues if it sought to advance particular kinds of case using the documents it has disclosed.

MEL's application for declaratory relief

142. Returning to MEL's application, I consider that this is a case in which, even though disclosure has been provided by way of a narrative statement and there is no formal list, the manner in which further information has been provided was sensible. Indeed, if anything, the detail of the development of the protocols in Bould 5 erred towards over-provision. The material provided gives ample opportunity to Neo to understand what was done and why, criticise it and, if appropriate, formulate focussed requests for further information either of MEL or of the experts. It also enables the court to understand the rationale for the Experiments, how they developed and the extent to which MEL's experts themselves debated different approaches between themselves and lawyers.
143. That said, the manner of giving disclosure was not exactly as required by Birss J's order and it would be inappropriate to make a declaration of compliance with it as it stands. Nonetheless, it seems to me that it would be appropriate to say that I consider this to be immaterial since the disclosure given by Bould 5 (which includes numerous documents and a list of them as well as much more) seems to me to have been a perfectly adequate alternative. I am unpersuaded in this case that this is a situation in which not having all of the original documents matters greatly. The risks of serious partiality as a result of having done the exercise in this way are very low particularly given the ways of challenging the evidence generally.

Conclusion on application 1

144. For the reasons given above, I believe that sufficient information has been provided at this stage but I do not think that it would be appropriate to make the declaration sought. I will however make a declaration that no such further disclosure should be given at this stage of the proceedings. If matters develop in particular ways at a later stage in these proceedings, it is not impossible that a more focussed application for information or disclosure of specific earlier material may become appropriate.

APPLICATION 2 - SAMPLE TESTING AND EXPERTS

145. By the second application made on 7 November 2017, MEL seeks permission to conduct certain tests on samples taken pursuant to the inspection of the second defendant's process in China. It also seeks permission to provide the PPD served by Neo to two named individuals who are experts in the field additional to those already instructed.
146. Evidence relating to the applications was provided by Mr Christopher Sharp, senior associate in Pinsent Masons, solicitors for MEL and Mr Neil Jenkins partner in Bird & Bird solicitors for Neo.
147. It is also said on behalf of MEL that this application required urgent resolution and that, given the history, the court has to resolve the differences now rather than giving time for the parties to try to agree an appropriate regime. I therefore gave my judgment at the conclusion of the argument with reasons to follow. These are the reasons.

Confidentiality of the PPD

148. I take up the procedural history again with the provisions relating to the PPD. The original PPD was served on 21 July 2017 and was said to describe the process used to manufacture the TP1 product the subject of the trap purchase. There followed a further protracted dispute as to which individuals on the MEL side should be permitted to see the PPD and to what confidentiality undertakings they should be subject. During this period, the only individuals with access to the PPD were MEL's external lawyers and it was therefore difficult to move the case forward. That was one of the matters brought back before the court on 14 September 2017. During the course of that hearing, a further, more complete PPD, was provided to Pinsent Masons although no signed PPD had been provided until shortly before this hearing. Arnold J ordered a "two-tier" confidentiality regime with external lawyers and the two experts referred to below admitted to an inner confidentiality club and able to see the whole PPD and while an in-house lawyer and commercial director of MEL were only entitled to see a redacted version of what was by then a "New PPD".

The Inspection

149. In order to understand my reasoning with respect to certain of the samples, it is necessary to say something about the inspection.

Birss J's order relating to the Inspection (April 2017)

150. At the CMC in April 2017, Birss J ordered there should be a process inspection of the second defendant's process to commence as soon as reasonably practicable on the first available date on or after 14 August 2017 as the parties may agree ("the Inspection"). The order provided that the Inspection may be attended by a reasonable number of the Claimant's external solicitors and Counsel and by its independent expert but only in so far as such persons had been admitted into the confidentiality club for the PPD. That order stated:

"Any information and materials obtained by the Claimant's representatives or independent expert at the inspection, or provided by the Defendant during or for the purposes of the inspection (including notes, photographs and video or sound recordings), may be provided to and seen only by persons falling within paragraph 14 of the Order."

151. There was a provision for the parties to apply for further directions relating to the Inspection. Arrangements for the Inspection were not easy to agree. One issue was whether, and on what terms MEL should be entitled to have experts attend the Inspection in the light of the confidentiality of the process. There was correspondence between the parties in which MEL put forward the names of proposed experts to attend and proposed confidentiality terms.
152. In the absence of agreement, the parties invited the court's determination of the issue at a hearing before Arnold J on 14 September 2017 which dealt with this and a number of other procedural issues.

Arnold J's order relating to the Inspection (September 2017)

153. Among the orders made on 14 September 2017, Arnold J ordered that the Inspection commence in the week of 23 October 2017 and gave permission to the claimant to instruct two independent experts to attend it, Dr Alain Rollat and Dr George Salem, in each case subject to the provision of a written undertaking to the defendants in a form to be agreed and, in the absence of agreement, to be settled by the court at a further hearing.
154. In a witness statement served for that hearing before Arnold J, Ms Bould on behalf of MEL provided an extensive justification for the need to have these particular experts attend the inspection including the alleged need for experience in the manufacture of mixed metal oxides or familiarity with zirconia-ceria technology. She made the following points:

(1) MEL would need an expert witness at trial to deal with the issues of infringement and the infringement expert would need to be provided with the PPD and all the materials generated at the inspection.

(2) The identity of the experts to attend the inspection was "very closely related to, if not the same as, the issue of the identity of the Claimant's infringement expert".

(3) It would be necessary for the inspection team to understand the various chemical reactions being carried out at the various stages of the process to understand what samples would need to be taken and why.

(4) The expert "may well need to give an opinion as to whether any difference between the Defendant's method of producing zirconia-ceria REMO and the patented process….was an immaterial variant".

155. Dr Rollat was put forward as having a great deal of experience in the field and it was implicit in that evidence that he would be a suitable expert to address the inspection and act as an infringement expert at trial.

Inspection report

156. Arnold J ordered that, within 14 days of the inspection, MEL should provide Neo with an inspection report describing the inspection and it made provision for identification of any areas of disagreement.

Confidentiality terms are agreed for the inspection experts

157. In the event it proved possible for the parties and the two experts to agree confidentiality terms following that hearing in the light of Arnold J's indications, with which all were content. These terms, which are described in greater detail below, were restrictive as to certain activities in which the experts could engage in future, both as regards the undertakings for which they could work on certain technical matters and as to the time for which they were to operate. Because these stringent terms were agreed, Arnold J did not have to decide whether a lower level of restriction would have offered Neo adequate protection.

A limited confidentiality club for the Inspection

158. The inspection was therefore ready to proceed on the basis of a limited and tightly controlled confidentiality club. At that stage, the number of individuals, other than external lawyers, entitled to see the confidential information in the PPD or to be derived from the inspection was small: two independent experts, subject to stringent undertakings, enforceable by the English court and Dr Moles (who had by then become an independent consultant) and also subject to strong confidentiality undertakings.

Correspondence concerning sampling prior to the Inspection

159. There had been limited discussion previously as to what exactly MEL's representatives wanted to see and sample at the Inspection or what they were proposing to do with it as regards testing. However, on 13 October 2017, Pinsent Masons on behalf of MEL wrote to Bird & Bird on behalf of Neo to give notice of the stages of the process from which they wished to take samples for analysis and put forward an outline of the testing that they intended to carry out on the samples from them. Those categories are set out in confidential Annex 1 to the draft Order. Bird & Bird responded on 17 October 2017 without objecting to sampling in principle but reminded Pinsent Masons that "any laboratory that your client intends (and to which our clients agree) to instruct to carry out compositional analysis of samples of raw materials or intermediates will first need to be approved by our clients and also provide signed confidentiality undertakings (to be agreed) in respect of the samples and the results of the analysis of them". They also raised a question as to the value of certain of the proposed testing. On 20 October 2017 Pinsent Masons said that they were in the process of identifying a suitable laboratory to carry out the analyses and "noted" the requirement for such a laboratory to be approved by Neo and to be subject to confidentiality undertakings - Pinsent Masons did not take any objection to this in principle. The letter said that MEL intended to test the solid samples in categories 2/3 by fingerprint testing to verify that the underlying structure of the product was the same as both TP1 and the final product of the inspection. On 3 October 2017, Bird & Bird requested full details of the fingerprint testing that MEL intended to have carried out on the solid samples before the samples were taken to decide whether it was necessary for Neo to run its own tests on their own samples of such materials. Pinsent Masons responded shortly after by saying that they were not able to do so yet because they were investigating appointing a catalysis chemistry expert to deal with this and other aspects of the case. No laboratory for proposed analysis was identified, no terms were proposed as to confidentiality. So by that stage, MEL and its advisors knew that they would have to obtain approval from Neo for any testing to be undertaken by a third party because of the requirements of Birss J's order set out above.

The Inspection

160. The Inspection was lengthy. It began on 23 October 2017 at the second defendant's manufacturing facility in Zibo City, Shandong Province, PRC. Three representatives of MEL attended: Mr Sharp, Mr Maxwell Keay (MEL's junior counsel) and Dr Rollat. It was completed in the early hours of Saturday 28 October 2017. Mr Jenkins explains, in evidence which has not been countered on behalf of MEL, that the second defendant worked hard to ensure that the MEL inspection team was able to be present at any stage of the manufacturing process should they wish, including the opportunity to be present at the facility 24 hours a day, particularly during the time the material was passing through the kilns.
161. That was continuous from approximately 18.30 on Day 2 to 00.30 on Day 5. MEL's team was also granted unfettered access to witness the mixing of the raw materials and to verify that the precipitation tank only contained those materials.
162. No reason is provided on behalf of MEL for why, if MEL was concerned that the material emerging from the kilns or at the end of the process differed in any relevant respect from the material going into the kilns, a 24 hour watch was not kept on this process. Nor is there any evidence from the Inspection to support the suggestion that this is likely to have happened.
163. It is true that Dr Moles said in his evidence for permission to serve out referred to above that he knew of no other method than the patented one which could produce material with the relevant characteristics but there is limited independent evidence before the court at this stage on this issue. This is not, therefore, a situation in which there is solid evidence showing that the similarity of the final product of Neo's process to MEL's product produced in accordance with the patent is likely only to be explicable by some sort of "switching" during the Inspection. Neo points out, with justification, that no reason at all has been provided to doubt the accuracy of the PPD or that the process inspected did not reflect the actual process of manufacture of the Neo product. MEL's requests for permission to undertake sample testing have to be viewed against that background.

The requests for sample testing and new experts

164. On 1 November 2017, Pinsent Masons wrote to Bird & Bird to say that apparently new legal requirements in China meant that samples exported for testing in the UK would first have to be tested in China to determine whether they were safe to be transported by air. They had also said on 27 October that they were considering instructing Prof Hardacre who, in contrast to Dr Rollat, did have the relevant expertise to opine on the question of equivalence. However, they said that Professor Hardacre was not prepared to sign an undertaking where the restrictive covenant affecting his future conduct affected work with any undertaking (not limited to the MEL group) and that it would have to be limited to 5 years.
165. The upshot is that the position after the Inspection as regards the proposed dissemination of information relating to the second defendant's process was materially different to that before. After the Inspection, MEL (a) was proposing to instruct two new experts, provide the PPDs to them subject to less stringent terms than previously agreed with the Inspection experts a few weeks earlier and (b) was proposing to subject the samples obtained from the Inspection and agreed to be treated as confidential to an incompletely specified set of tests with very limited confidentiality guarantees at least at some stages. Neo was not prepared to agree to this. Hence the application.

The current confidentiality regime

166. Before addressing the orders sought it is necessary to say something more about the confidentiality regime and the basis for it. MEL contends that it is too strict given the nature of the information in question. Neo, on the other hand, contends that it is entirely appropriate given this highly competitive field and given certain critical information which would make it easier for a competitor to produce a rival product to Neo's, if it were disclosed.
167. The confidentiality regime in place for these proceedings established by agreement between the parties has a number of general features:

(1) It applies not just to documents but also to "materials obtained at the Inspection". That includes samples taken, with the effect that it is accepted for present purposes that samples may not be handled by any person not within the regime.

(2) As is usual, non-lawyer members of the confidentiality regime (and the General Counsel of MEL's parent company) have signed confidentiality undertakings. The terms of such undertakings were settled during discussion at the hearings before Arnold J in September 2017.

(3) The confidentiality regime has two tiers relating to (i) a complete and (ii) a redacted PPD. The redacted PPD has the identities of two particulars substances removed – Substances X and Y – that Neo say are particularly sensitive.

(4) The confidentiality undertakings include stringent restrictions on MEL's experts from working in the field of REMO in the future. The hearing before Arnold J involved extensive debate about how wide those restrictions should be. However, Arnold J did not have to settle the terms of the restrictions finally because the proposed experts were prepared to sign compromise restrictions on terms that Neo accepted. Such restrictions were accepted on MEL's part solely on the basis that the experts were prepared to accept them, without MEL conceding that they were necessary, appropriate or precedent-setting. In addition, it was specifically contemplated that MEL could revisit the extent of the confidentiality restrictions.

Issues

168. During the adjournment between hearing the two applications, the parties were able to agree a regime in respect of testing the raw materials and final product samples. By the end of the hearing, only two main issues were left outstanding:

(a) Whether permission should be granted to implement the testing regime proposed for certain intermediate samples taken from the inspection and certain related matters relevant to testing of these samples?

(b) Whether permission should be granted to admit the two academics into the confidentiality club and if so on what terms?

Issue (a) - Inspection and process verification

169. The English courts have had some experience of situations in which a person alleging infringement of a patent is offered inspection of an allegedly infringing process but it turns out that the process of which inspection is offered is or may be a sham (in that it is not or may not be the process actually used to produce the product which is subsequently sold in the United Kingdom but is a process simply shown for the inspection). Birss J referred to one of the well-known cases in the English courts dealing with this issue, Nutrinova v. Scanchem [2001] FSR 42, in his first judgment dealing with permission to serve out as an illustration of the difficulties that can arise in trying to prove that an infringing process has been used (see [2015] EWHC 3596).
170. In general therefore, an order providing for inspection of a process has two purposes: (i) to demonstrate the defendant's process in operation and (ii) to provide verification that the process which the defendant contends that it is using is, in fact, the process it is or has been using to produce the product imported into the United Kingdom. There is no better evidence to counter the assertion of a patentee that it is impossible to produce a product with given characteristics without infringement than to show such product being produced regularly on an industrial scale in a way that does not involve use of that process.
171. It is in the public interest that, if an infringing process is being used, that this is discovered. That points towards a defendant being as open as possible on an inspection. It is equally in the public interest that a defendant which has discovered a way of producing a competing product without infringement and in which it has its own intellectual property rights does not have those rights put at unnecessary risk as a result of litigation of an unjustified claim.

Principles

172. Although this case is concerned with provision of confidential information to a potential expert and for sample testing rather than provision of information to a representative of a party, the general principles set out in Warner-Lambert Co v. Glaxo Laboratories Limited [1975] RPC 354 and Roussel Uclaf v. Imperial Chemical Industries plc [1990] RPC 45 are, in my view, equally applicable to the present situation, albeit with some minor qualification.
173. In Warner Lambert the Court of Appeal said at 360:

"If in a particular case it is right that disclosure of any facts should be made by one party to his opponent's advisers before trial, it must normally follow as a matter of course that the opponent should be entitled to know the facts so disclosed. His advisers are his agents in the matter, and strong grounds must be required for excluding the principal from knowledge, which his agents properly acquire on his behalf. But this principle must be subject to some modification if trade secrets are to be protected from disclosure to possible competitors. Suppose, for example, that in a case such as the Coloured Asphalt case the court makes an order for inspection of the kind which was made in that case and the plaintiff's advisers, having inspected the defendant's process, report that in fact no infringement has taken place. The whole object of the procedure would be defeated if the plaintiff could insist as of right on knowing the details of the defendant's process, and no useful purpose would be served by his doing so. Where a matter in question in an action, begin that matter in question in an action, being the matter upon which inspection or disclosure will throw light, is of a technical nature, the party seeking discovery may well require inspection by, or disclosure to, technical and professional advisers. If the matter be of a kind on which the party will be likely to be able with the aid of those advisers to form some kind of view of his own, it seems to me that he should normally be allowed to know as much about the facts as his advisers. If, however, the case were one of so esoterically technical a character that even with the help of his expert advisers the party himself could really form no view of his own upon the matter in question but would be bound to act merely upon advice on the technical aspects, disclosure to him of the facts underlying the advice might serve little or no useful purpose. In such a case a court might well be justified in directing disclosure of allegedly secret material only to expert or professional agents of the party seeking discovery on terms that they should not, without further order, pass on any information so obtained to the party himself or anyone else, but should merely advise him in the light of the information so obtained. Even so, if the action were to go to trial, it would seem that sooner or later the party would be bound to learn the facts, unintelligible though they might be to him, unless the very exceptional course were taken of excluding him from part of the hearing. Even where the information is of a kind the significance of which the party would himself be able to understand, it may nevertheless be just to exclude him, at any rate during the interlocutory stages of the action, from knowing it if he is a trade competitor of his opponent."

174. In Roussel Uclaf v Imperial Chemical Industries Plc [1990] RPC 45, Nourse LJ cited with approval the following passage of Aldous J at p54.

"Each case has to be decided on its own facts and the broad principle must be that the court has the task of deciding how justice can be achieved taking into account the rights and needs of the parties. The object to be achieved is that the applicant should have as full a degree of disclosure as will be consistent with adequate protection of the secret. In so doing, the court will be careful not to expose a party to any unnecessary risk of its trade secrets leaking to or being used by competitors. What is necessary or unnecessary will depend upon the nature of the secret, the position of the parties and the extent of the disclosure ordered. However, it would be exceptional to prevent a party from access to information which would play a substantial part in the case as such would mean that the party would be unable to hear a substantial part of the case, would be unable to understand the reasons for the advice given to him and, in some cases, the reasons for the judgment. Thus what disclosure is necessary entails not only practical matters arising in the conduct of a case, but also the general position that a party should know the case he has to meet, should hear matters given in evidence and understand the reasons for the judgment."

175. Secret processes merit particularly strong protection in this context (see IPCom GmbH & Co Kg v HTC Europe Co Ltd & Ors [2013] EWHC 52 (Pat) (24 January 2013) where the court drew a distinction with cases "at the high end of the scale represented by secret process cases"). There is no unique way of balancing the interests. Birss J summarised the position as follows in Smith & Nephew Plc v Convatec Technologies Inc & Anor [2014] EWHC 146 (Pat) (31 January 2014) at [4]:

"….At the stage of disclosure it is well established that in a proper case a confidentiality scheme or "club" can be set up. See three Court of Appeal authorities Warner Lambert v Glaxo [1975] RPC 354, Roussel Uclaf v ICI [1990] RPC 45 and Lilly ICOS (No 2) [2002] EWCA Civ 2. The scheme may be arranged by order of the court but is often arranged by an agreement between the parties, albeit always subject to the Court's jurisdiction…. The system is flexible and there are many variations…."

Disclosure to a party cf. disclosure to a third party

176. The majority of the case law has concerned disclosure to an individual or individuals employed by a party. In those cases, the rationale for at least one individual knowing the information is particularly strong since otherwise a party may be subject to a judgment without any internal representative understanding the reasons why that happened. This is a situation in which there is often no choice but to provide limited disclosure in order to avoid unfairness. The position may be slightly different where disclosure to a third party is concerned since, in such cases, it may be possible to advance the litigation in different ways. However, usually in a patent case disclosure to at least one external expert is likely to be required to address a serious issue at trial. Those cases did not explicitly refer to the overriding objective but in my view the court should also have regard to that general principle in deciding whether confidential information should be provided to a third party for the purpose of conducting litigation subject to the CPR.

Summary of principles

177. For the court to give permission to a party to disclose an opposing party's confidential information (including a product in which it is embodied) to a third party (including a potential witness, expert or factual, or a laboratory for testing), it should be satisfied that such disclosure is:

(1) Reasonably necessary to address a serious issue at trial;

(2) Protected by confidentiality provisions which are sufficient to minimise and/or mitigate the risks of the confidential information becoming known, disclosed or used in an unauthorised manner having regard to all the circumstances including the nature of the information in question, the identity of the proposed recipient and the proposed use of the information for the litigation; and

(3) Consistent with the overriding objective of the CPR.

178. Those principles should be applied with the case law cited above in mind and must adopt a flexible approach tailored to the particular facts. In particular, where a party's trade secrets are involved, the court should be particularly astute to ensure that no unnecessary risks are taken with them and that any necessary risks of unauthorised disclosure are kept to a minimum.

Confidentiality

179. MEL advances the following arguments in support of its contention that its proposed confidentiality regime is appropriate. I have summarised them at a general level for the purpose of the judgment to avoid the need to have a specific confidential schedule going into greater detail.
180. First, there is no dispute that Neo's PPD is confidential and deserving of the protection of a confidentiality regime of the sort usually found in patent cases in this Court.
181. Second, among other confidential information Neo's PPD includes the name of an additive which is referred to as X in Arnold J's order so as to preserve its confidentiality. Substance X is central to the case. Neo contends, first, that the identity of substance X and the knowledge that it can be used in an industrial process of this kind is sufficiently confidential that it should bar a recipient of the information from working in the specific field not just for MEL but for any competing company. MEL disputes this contending that the identity of this substance is not so secret or valuable and is no more so than aspects of other PPDs regularly considered which are not subject to such a regime. It contends that there is no evidence that (for example) Neo's own employees or own expert would be under such a stringent regime of confidentiality and there is no basis for such an obligation to be imposed on any expert.
182. Third, MEL contends that X is not as confidential as Neo contends. Since Arnold J's order new information has come to light. The details are given in Mr Sharp's first statement at paras. 9.7-9.10 and I will not recite them here. Neo argues that it would be impossible for a recipient of information about X to put that out of his or her mind and would inevitably have it in mind if working with any other competitor.
183. Fourth, it is said that the final product cannot possibly be confidential given that the final product is available on the open market. The whole point of the Inspection is that this final product is said by Neo to be the same as the trap purchase. Neo responds to Mr Sharp's evidence with evidence from Mr Jenkins showing how confidentiality remains notwithstanding the matters referred to by Mr Sharp.

Discussion

184. I consider that, notwithstanding MEL's arguments, that, at this stage of the case, a sufficient basis exists for seeking to impose strong protection for Neo's confidentiality in its product and process so far as possible. It is not possible to resolve on this application the precise ambit of Neo's trade secrets. It seems unquestionable that there is an element of trade secrecy about aspects of its process and, even if other aspects may not be confidential or as confidential, determining at what level of generality confidentiality exists is impossible on this application. It may extend to X as such or it may only extend (for example) to the specific proportions or conditions in which X is used. It may be that because the process as a whole is confidential, it is not possible to separate out components of it.
185. However, in determining what confidentiality restrictions are appropriate I have borne in mind the fact that the evidence, taken as a whole, including that which is in the public domain suggests that the real value of Neo's confidential information lies not just in the use of X but in the details of the process and that reduces the risk that disclosure of information at a general level will compromise Neo's position. These were matters not known to Arnold J when he was considering appropriate confidentiality terms. MEL does not dispute that the PPD as a whole contains confidential information which merits protection in principle. All the evidence as to the absence of confidentiality in this case comes from an external lawyer for MEL instructed in the case and I am not satisfied that this is a sufficiently firm basis on which to conclude that the information sought to be protected is not confidential.
186. Taking the materials together, this is case for requiring strong albeit not necessarily the strongest possible confidentiality protection.

Inspection sample testing

Agreed testing – starting materials and finished product (Categories 1 and 4)

187. There was no dispute between the parties that some element of verification of the process shown at the Inspection was appropriate and there has been no fundamental objection to reasonable testing to this end.
188. By the start of the hearing, the parties had moved significantly closer as regards how the starting materials and end product samples (Category 1 and Category 4) should be dealt with. I suggested that there should be a further attempt to agree an approach for these. That was done after conclusion of argument on this part of the case. A framework was agreed which involves testing of some of the raw materials by a laboratory in China and X by a laboratory in the United Kingdom. Some of the precise details of the confidentiality agreements and the way in which the samples should be dealt with in getting them to the UK remained to be tidied up although this has now been done.

Disputed testing – intermediate samples (Categories 2 and 3)

189. The main dispute concerns the intermediate samples from two stages of the process witnessed at the Inspection. The problem arises from the fact that the testing contemplated by MEL would either involve a risk that the details of the Neo formulation would be revealed by that testing itself (Category 2 and 3) and/or that further confidential information would be required to put certain samples into a condition in which they could be subject to comparable testing (Category 3).

Category 2 samples

190. This is the slurry of 6 batches following the initial step. It includes X. As Mr Jenkins says, "The Category 2 Samples also contain information about the identities of the individual components used in the Defendant's process (including substance X and the salts of the rare earth used) and the proportions in which they are used which is confidential information". Laboratory testing may therefore reveal X and the proportions of X. There is no specific test only for sulphate ions which is practicable for this slurry.

Reasonably necessary to address a real issue in the case

191. I am satisfied that this testing is reasonably necessary to verify that the raw materials used were actually used in the actual mix and that there was no sulphate in the claimed proportion in the mix.

Confidentiality protection

192. The real question is whether that there is a risk which Neo should not have to bear that the undertaking which it is proposed should undertake that analysis may (a) discover the identity of X during the course of its testing or (b) disclose that X is present.
193. MEL has proposed two laboratories in China which may be able to undertake appropriate testing. One has signed confidentiality undertakings. However, these laboratories have only been investigated comparatively recently. One was contacted just over a week before the hearing. There is limited information about these laboratories on this application. Although I have no reason to doubt their integrity, the proposed laboratories are not within the jurisdiction of this court. The confidentiality undertakings provided so far are not particularly stringent. If, even accidentally, they were to disclose information about the product tested to a third party, that could cause irreparable damage to Neo.
194. Moreover, it is unclear exactly what testing it is proposed to invite the laboratories to undertake. No draft protocol for such testing has been prepared and it is unclear precisely what has been discussed with these laboratories.
195. In my view, before requiring Neo to have samples of its confidential product provided to a laboratory outside the jurisdiction for testing on terms considerably less protective than Arnold J regarded as appropriate it would be right to enable further communications about this issue to take place to determine whether a different approach less likely to compromise Neo's confidential information can be devised.
196. Shortly before I gave my decision on this issue on the second day of the hearing, the parties indicated that they were, in any event, exploring alternative possibilities for testing of this material possibly by Neo itself under appropriate supervision. That is sensible and I will therefore make no order in respect of these samples but give permission to apply should these alternatives prove impossible.

Category 3 sample testing by Professors Boghosian and Efstathiou and disclosure of confidential calcination conditions

197. That leaves Category 3 samples about which there was the most extensive dispute.
198. In addition to subjecting samples of the finished product (Category 4 samples) from the inspection to Fingerprint Testing, MEL wishes to perform the same kind of testing on Category 3 samples (after they have been calcined by an independent laboratory) as was done on the TP1 samples. It had originally been suggested that it may wish to do the same for Category 2 samples as well but that has been dropped. That testing is proposed to be done by Professor Boghosian for Raman spectroscopy and by Professor Efstathiou for kinetic testing but after the Category 3 samples have gone through a laboratory calcination step.

Reasonably necessary to address a real issue in the case

199. The justification put forward for doing this testing in Mr Sharp's evidence is that because he and the rest of the team attending the Inspection "were unable to be present at the Second Defendant's manufacturing facility 24 hours a day, we have no way of verifying that the Samples handed to us at the end of the inspection were those which we had seen at the various stages of the Inspection". However, there is equally no evidence that it is likely that these samples were or may have been different. As noted above, Neo did not place any obstacles in the way of the inspection team attending throughout the production process. So this testing is intended, in some sense, to be a substitute for incomplete observation of the whole process.
200. The English court will not (for example) order pre-action disclosure of a defendant's confidential process unless there is some basis for considering that infringement of rights may well have occurred, even if confidentiality of the process can be preserved by a tight confidentiality club. Although the situation was not the same as here, by way of illustration, Patten J said in BSW Ltd v Balltec Ltd [2006] EWHC 822 (Ch) at [83]:

"Disclosure of commercially sensitive material is a necessary part of litigation in this Court and there are commonly used procedures in the form of confidentiality clubs and the like which are employed to limit the recipients of such disclosure and its analysis for the purposes of resolving the issues in dispute. Some disclosure to the parties themselves is necessary and inevitable for the litigation process to operate. Similarly, evidence may be available to support design right and copyright infringement on the basis of which wider disclosure of other sensitive material may be required. But to allow CPR 31.16 to become a means of examining a competitor's otherwise secret designs on the basis that some kind of infringement might have occurred cannot in my view be permissible unless there is at least a clear and convincing evidential basis for the belief that acts of infringement may have taken place and the Court can be satisfied that the pre-action disclosure sought is highly focused. Otherwise, the potential for abuse is obvious. To say that one can establish the usual confidentiality club is not an answer in itself to the objection that there is no basis for being required to offer up the confidential material in the first place".

201. In my view, the position is similar where confidential information is sought to be obtained to determine whether a defendant has "switched" samples during the course of an inspection. There has to be some reasonably secure ground for believing that this might have been the case before disclosure of trade secrets to test that matter is provided, especially in a situation where the party seeking disclosure has been given 24 hour access to the process being operated.

Nature and probative value of the proposed Category 3 sample testing

202. The testing proposed is different from that proposed for the other categories. Category 1 and 2 samples are proposed to be tested to determine a simple fact: are sulphate ions present at all in the raw materials and initial mix (or at the levels required by the claims)? Category 4 samples are to be tested for comparability to the TP1 Samples. That is more complex but, in principle, simply involves a rerun of existing tests on the samples from the Inspection.
203. The proposed Category 3 sample testing involves a more involved experiment with an additional laboratory calcination step. In my view, the court must exercise some control over experiments that it permits and only allow those which are likely to have real incremental probative value having regard to all of the other material in the case.
204. I am not satisfied that these proposed tests have such value. In that connection, I have had regard to the following.
205. First, extensive data was generated by the fingerprint testing of TP1 samples. At least in some cases, the differences between the various materials tested which are said to have been produced by different processes do not appear to be particularly striking having regard to the range of other data and how the differences are described. There may be a dispute as to the significance of this at trial, should the case continue, and to what aspects of the production process they are attributable. Category 3 samples involve the addition of a further variable (namely the precise calcination conditions) and add an extra layer of contestability.
206. Second, there is no direct evidence from Dr Rollat, who was put forward as an expert in this field and who attended the Inspection, suggesting that this testing was essential in order for him to verify that the Inspection was of the genuine process and that the samples provided were of material produced by the process shown. Nor is there any other convincing evidence that this testing will provide material assistance to the court in resolving the central issues at trial. Counsel for MEL submitted that the trial judge would not thank this court for not permitting this testing to be done at this stage. I take a different view of the likely reaction of the trial judge in circumstances where the material to emerge from this proposed suite of lengthy and doubtless costly tests may be inconclusive and there is no dispute that 24 hour access was provided for the process. Moreover, even with the timetable to trial, should it turn out that this testing really is necessary, there will be sufficient time to undertake it somewhat later.
207. Third, the Inspection was supposed to provide a definitive answer, inter alia, to whether the fingerprint testing was a reliable guide to whether the process of the patent had been used in this sense: if the Inspection showed that the process employed by Neo produced product with the same or very similar fingerprint to that of product using the patented process, it would show that it was not necessary to use that process to obtain those results. It does not seem appropriate in those circumstances to use the contested fingerprint testing to validate the integrity of the Inspection. It is common ground that calcination conditions can affect the finished product to some extent and the expert evidence submitted indicates a general sensitivity to process conditions. Neo says that the proposed laboratory calcination of the Category 3 samples may not replicate its actual production conditions and, although I cannot resolve that issue on this application, it is likely that there will be a real dispute over this, if this testing is permitted.
208. It follows that the best that can be expected to emerge from this proposed testing is inconclusive results which prompt further debate as to their probative value. There is no evidence that this testing is able to provide a conclusive (in the sense of indisputable) answer to whether or not the finished product at the end of the process was the product of calcination of the earlier product from which the Category 1 and 2 samples were taken.
209. Fourth, the raw ingredients for the process are, in any event, to be subject to testing and the final product of the process is to be subject to the fingerprint testing all by agreement achieved during the course of the hearing. It appears likely that there will be some agreement as to testing of one of the intermediates. That leaves one of the other intermediates. It is unlikely that if the combination of (a) a five day inspection with an offer of 24 hour attendance by two lawyers and an expert chemist, (b) sampling and composition testing of the ingredients and the finished product do not reveal that the patented process has been used, that the analysis of the intermediate samples (one of which is likely to be tested by agreement as well) will do so. So the incremental value is likely to be modest.

Confidentiality protection

210. In order for this proposed testing to be carried out, it will be necessary to calcine the Category 3 samples first. MEL therefore additionally seeks details of the second defendant's confidential calcination conditions so as to produce a final product in laboratory conditions as similar as possible to the industrial process. Information of this kind is said to be highly confidential by MEL itself and it is common ground that this is the second defendant's confidential information.
211. It is true that Professors Boghosian and Efstathiou and their respective universities have entered into confidentiality agreements with MEL. However, in order for the testing to take place there would have to be a further disclosure of this further confidential information to another expert or laboratory which may be Professor Hardacre's laboratory but may be a different laboratory.
212. Every time further information is disclosed to another undertaking, the risk increases of some of Neo's confidential information becoming known to an undertaking which should not know it. Moreover the current proposal would mean that not only was the PPD disclosed to Prof Hardacre including what are said to be highly confidential details of the first part of the process but also that the precise conditions of calcination, namely the latter part of the process. It is likely that confidentiality issues will also arise in exporting the Category 3 samples from China without that material being subject to testing by or on behalf of the shippers to ensure that it is safe to transport. That gives rise to a further risk that confidential information may be revealed, which has not so far been adequately addressed. It may be that there are other practical alternatives for testing of the Category 3 samples of the kind currently being explored with respect to the Category 2 samples which would cut through these problems but these have not yet been explored.
213. I note that there is no other issue in the action to which the precise conditions of calcination is relevant. Neo admits that there is calcination. There is no claim of the patent which specifies precisely the calcination conditions. There is, in my view, no need for this information to be provided for the case in other respects.

Overriding objective

214. Cases must be resolved justly and there are times when the risks of damage done by the litigation process itself is disproportionate to the benefit to undertaking the task in question. In such cases, the overriding objective would suggest that these risks should not be run.
215. For the above reasons, I do not think it would be just to require provision of these samples and the further confidential information.

Further application

216. I do not rule out the possibility of a further application with respect to testing of the Category 3 samples. But before that is done there should be evidence that such was necessary and proportionate to resolve a real issue in the case. If, for example, it is suggested (on the basis of evidence commensurate with the seriousness of such an allegation) that the process shown at the inspection and the products produced as a result were not fairly representative of the process used by the second defendant and this proposed testing was determinative of whether it was, the court would need to be satisfied (i) that there were good reasons to believe that there had been some unfairness about the Inspection (ii) that this proposed testing would be decisive as to whether there had been (iii) there were no better means of determining the issue and (iv) appropriate confidentiality or other arrangements were in place both for the testing itself and the means for transporting the samples for testing.
217. At present, none of those conditions are satisfied and I therefore make no order in respect of testing of the Category 3 samples and provision of confidential calcination conditions.

Issue (b) - Admission of Professor Hardacre and Dr Chansai to the inner confidentiality club

218. As noted above, until very recently, MEL had two experts in the field, one of whom attended the inspection for five days and who was put forward as having appropriate expertise. Having seen the new PPD, those experts have recently informed MEL that they do not have the necessary expertise to assist with issues relating to equivalence. This, it is said, only became apparent following service of the new PPD on 14 September 2017 and the admission of MEL's experts to the confidentiality regime thereafter.

Professor Hardacre and Dr Chansai

219. MEL has identified Professor Christopher Hardacre of the University of Manchester as an expert with the necessary expertise in the field of catalysis chemistry. MEL therefore wishes to instruct Professor Hardacre to assist with issues relating to infringement and to carry out the calcination of the Category 3 samples referred to above to prepare it for the fingerprint testing. Professor Hardacre has suggested that a member of his team, Dr Sarayute Chansai, assists him with laboratory testing.

Two new experts?

220. I have already ruled out fingerprint testing of the Category 3 samples and that makes a further individual, Dr Chansai, to assist Professor Hardacre unnecessary. The only issue is whether MEL should be able to engage Professor Hardacre as an expert to address the infringement case and therefore should receive the PPD and the information from the Inspection.

Reasonably necessary to address a real issue in the case

221. MEL submits that since Dr Rollat has said he is unable to assist with the issue of whether there is infringement on the basis of an argument of equivalence, MEL now has no choice but to see a new expert to deal with that issue if it wishes to pursue the case at all. MEL has yet to serve a statement of case on infringement and it contends that it needs expert assistance in two respects:

(1) Determining whether there is a credible case of infringement by equivalence at all;

(2) Formulating such a case, should such exist.

The need for expert evidence for the potential equivalence case

222. MEL submits that this case cannot be conducted without MEL's advisers, including its testifying experts, knowing, for example, what X is. This is because it is said that an issue at trial will be whether X, or a particle of which X is comprised, is "equivalent" to the sulphate ions called for by claim 1 of the patent. An inability to discuss this issue with a suitable expert would, it is said, seriously hamper MEL's ability to conduct the case at trial and no potential new expert on these matters has been prepared to agree to an undertaking in the terms previously agreed by Dr Rollat.
223. The courts have had to grapple on several occasions with the question of infringement by equivalents where the patent in question was prima facie specific as to chemical reagents to be used. Expert evidence is regularly adduced on this issue. Actavis v. Lilly has if anything increased the potential role that expert evidence may play in the question of scope of protection in at least some such cases. Such evidence may legitimately be brought to bear on at least some aspects of the test for equivalence, even in cases where it might be thought that the language of the claims clearly precluded such an argument.
224. I note that the Patent claims in the present case are specific as to reagents, molar ratios (in certain respects) and reaction temperature. Moreover, the patent refers to extensive prior art processes (see Patent paras. [0003]-[0012]) and includes a number of comparative examples, illustrating the sensitivity of the process to relatively modest changes in the molar ratio of sulphate anions to zirconium cations and to the use of other alternative reagents or different temperatures (see Patent paras. [0057]-[0067]). These are said to produce no usable product, take too long or produce product which, contrary to the stated object of the invention, consists of two crystal phases. However, it was not contended on behalf of Neo on this application that an argument of equivalence was so hopeless that, on that basis alone, MEL did not require any expert evidence at all to address it, although that point was not fully canvassed in argument.
225. That said, the potential difficulties in an equivalence argument were, in my view, realistically recognized in the submissions before Arnold J concerning the PPD where counsel for MEL said "in order to have an intelligent conversation with my lay client about whether this is a case that is worthy of going forward, after the experts have looked at it and considered whether it should be settled, whether it should be withdrawn completely, the client needs to understand at least the parameters of the debate" (Transcript of 21 September 2017, p2).
226. It might be said that where a party has foreshadowed the possibility that the claim may not proceed, the court may be legitimately more cautious in permitting disclosure of the other side's confidential information. I take that into account but it does not, in my view remove the need for expert evidence, indeed, it may increase the desirability of it, namely to enable early settlement. There is a powerful public interest in enabling litigants to advance seriously arguable patent cases. There is an equally powerful public interest in enabling weak cases to be appreciated, resolved or withdrawn without potentially further significant cost or adverse impact on a defendant.
227. There are therefore two good reasons why high quality expert evidence on this issue is required. To assist the court with evidence relating to it and if there is no credible case to assist MEL in appreciating that fact and provide a sensible technical platform for resolving the dispute.
228. I am therefore satisfied that there is a real need for such expert evidence.

The need for a new expert

229. The next question is whether there is a need for a new expert, Professor Hardacre. I have referred to the evidence above concerning Dr Rollat who might, on that basis, be considered to be a suitable expert requiring no additional supplementing. However, I have concluded that since Dr Rollat has disavowed expertise on the relevant issue, there is no real option but to involve a new expert. There is no serious challenge to Professor Hardacre's expertise at this stage nor is it said that he is a fundamentally unsuitable choice for the case. Professor Hardacre is a senior academic at a well known English university. He appears suitable. In principle MEL should be able to instruct him as its infringement expert (by which I include such issues of validity as may arise already and may additionally arise should a case of infringement by equivalence be advanced).
230. I conclude that it is reasonably necessary to provide the confidential information in question to Professor Hardacre.

Confidentiality

231. The issue which has caused me greatest difficulty on this part of the case is the question of the terms upon which Professor Hardacre should be permitted access to Neo's confidential information (including the PPD) for the purpose of discharging his duties as MEL's expert including as to considering with MEL whether there is a serious case of equivalence in the light of the PPD, the patent specification and the claims.
232. Neo submits that it would be wrong to admit Prof Hardacre into the inner confidentiality club without his being prepared to enter into confidentiality restrictions similar to those entered into by Dr Rollat. On the one hand, Neo is justified in saying that the reasons put forward for wanting to change experts are not very compelling and the disclosure of its confidential information to Professor Hardacre is not completely risk free. On the other hand, Mr Sharp's evidence that there is, for practical purposes no alternative, if there is to be a new expert, is hard to go behind and that the risks are, taken as a whole, very limited.

Discussion

233. The court has to balance the interests of enabling litigants to pursue genuine cases and instructing appropriately qualified experts to do so against the interests in preserving confidential information of a party unwillingly required to disclose it for the purpose of litigation. There are risks on both sides. On one side there is the risk of hampering the pursuit of a serious case and/or disabling a claimant from rapidly concluding that the case should not be pursued further. On the other side there is the risk of confidential information leaking out. Both those risks have to be minimised as much as possible, while recognising that it may not be possible to remove them altogether.

The proposed confidentiality terms

234. Clause 1 (i) – (ix) of the proposed terms for Professor Hardacre contain a set of provisions relating to confidentiality including an obligation to treat all confidential information in the PPD (and all materials obtained at the inspection) as strictly confidential and an obligation not to use the information for any purpose other than these proceedings. Professor Hardacre is prepared to sign a confidentiality undertaking in substantially the same terms as Dr Rollat but with two important exceptions. These differ from the confidentiality undertakings of Dr Rollat in that, instead of a general restriction on working in the field of REMO in the future, the undertakings contain a more limited restriction on working with MEL or its affiliated companies in the field of REMO for 5 years.
235. Professor Hardacre is understandably unwilling to enter into broader undertakings given his work with others in the industry. As Mr Sharp says, the terms which prevent him from working in the field of REMO in the future, may prevent Professor Hardacre even from continuing with projects he is currently working on.
236. In this case, I take particular account of the following in settling the confidentiality terms:

(a) the nature of the information in question as a trade secret, what is now said about the various respects in which the information is said to be confidential (including the level of detail);

(b) the importance of ensuring that the parties and the court are able to have the benefit of expert evidence on important issues;

(c) the risk, which I assess as low in the light of the further evidence about confidentiality, of dissemination of sufficiently specific confidential information to be of real incremental assistance to others over and above that which is not confidential as a result of what is proposed to be disclosed to Professor Hardacre (on the assumption that the confidentiality terms are complied with by him);

(d) the nature of the intended recipient, the fact that he is independent of the parties and his primarily academic role;

(e) the fact that enquiries have not identified any other expert with relevant expertise who is prepared to act on stricter confidentiality terms;

(f) whether the additional lifetime, industry-wide restrictions of the kind sought by Neo would risk imposing unreasonable fetters on the ability of the expert in question to undertake his academic and other work in future including work in which he is currently engaged.

Conclusion on confidentiality terms for Professor Hardacre

237. In my judgment, having regard to all of the circumstances, the balance comes out, albeit marginally, in favour of admitting Prof Hardacre to the confidentiality club on the terms proposed by MEL (namely without the full additional industry-wide life-time restrictions agreed to by Dr Rollat), on condition that Professor Hardacre is also provided with and confirms that he has understood the consequences of receipt of such information which should be done by giving him a copy of the relevant extracts of Glaxo and inviting confirmation that he has understood them. It is important for him and MEL to understand that the possession of confidential information is a real burden, regardless of the precise terms of the confidentiality agreement. If he considers that possession of this information may cause difficulties in his current or future work with other undertakings in this field, he will need to consider very carefully how to address it.

Conclusion on application 2

238. In the result, there will be:

(1) An order as to testing of Category 1 and 4 samples in the form agreed between the parties;

(2) No order as to the testing of Category 2 and 3 samples or the provision of the second defendant's confidential calcination conditions but with permission to reapply;

(3) An order admitting Professor Hardacre to the inner confidentiality club on the terms indicted in the judgment above.

Postscript

239. I am grateful to all counsel for their considerable assistance on these substantial applications and will hear further argument on the consequences. Part of the reason for giving longer reasons than may ordinarily be appropriate on what are, essentially, issues of case management is that, in this instance, the court is being invited to do acts which, if done by a party, would be actionable, namely to authorise provision of particularly strongly protected confidential information (privileged material and trade secrets) to a party or third party. In this case, the parties are entitled to know in greater detail, why the court has thought it appropriate to act – or decline to act – in a particular way since that may assist in resolving future disputes on these issues.