England and Wales High Court (Patents Court) Decisions

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		5 March 2003

Peter Prescott QC:

Introduction

1. This is a very unusual case. It is an appeal from a decision of the Patent Office. It was not a formal, reasoned decision of the usual sort. But it was a decision all the same. It was a refusal to furnish certain information to the appellant under section 118 of the Patents Act 1977: namely, on what date was patent application GB 9107381 withdrawn? The ground for the refusal was that the Patent Office no longer had any record of the information that was sought. The appellant contends that it is for the Office to find out. The Office contends that it is under no such duty, and says that nobody is entitled to this information in the first place.

To See Ourselves As Others See Us

2. Mrs Mandy Haberman is a well known inventor. She is best known for her 'Anywayup' trainer cup (baby feeding cup). It will not dribble Ribena on the carpet when a child holds it upside down. It may not sound like rocket science but for parents of small children it is far more important. For inventing this cup she received the British Female Inventor of the Year award (2000) and a gold medal at the Salon International des Inventions. From small beginnings, she has become a successful entrepreneur. Millions of trainer cups made to her design have been sold. See, for example, the website of the World Intellectual Property Organisation, which has chosen Mrs Haberman as a case study, precisely in order to show what a private inventor can achieve.
3. Mrs Haberman is also recognised for her efforts as a campaigner. She is in favour of a patent system which should be truly accessible to persons such as herself, that is so say, the small entrepreneur and innovator. Her opinions have achieved respect. So much so, that she has been invited to sit on various government committees which address that topic. (I take judicial notice of the foregoing facts since they are well known or readily accessible to those who practise in this branch of the law.)
4. A truly accessible patent system would be, amongst other things, easy for ordinary people to understand. And it would be devoid of unnecessary technicalities and pitfalls. Robert Burns wrote some lines which are very apposite to the present case. They are better in his native Scots, but perhaps clearer in English translation:-

O would some Power the gift could give us

To see ourselves as others see us!

It would from many a blunder free us.

If we – the international patent system – were granted the gift to see ourselves through the eyes of a Mrs Mandy Haberman, what might we learn? Would we be pleased with ourselves? Going by this case: I do not think so. I am bound to say that Mrs Haberman would seem to have every right to complain.

European Patent 0634922

5. For her trainer cup invention Mrs Haberman was granted a British patent. After much trouble and expense, it was held valid by Mr Justice Laddie: Haberman v. Jackel International Ltd [1999] EWHC 269. She was also granted a similar patent (EP 0634922) by the European Patent Office. However, it has run into a snag. It has been revoked by the Opposition Division. The decision can be reversed by the Board of Appeal; but that is how matters presently stand. Let me now recount how this came to pass.
6. In order to do so, I must begin by describing a topic which is known to all professionals in this branch of the law. But I want to see ourselves as others see us. So I shall imagine that I am trying to explain it to an intelligent person who knows nothing about it. The topic is called "Regenerating your priority date".

Regenerating Your Priority Date

7. For a patent to be valid, your invention must be new. It will not be considered to be new if the same idea was already part of the state of the art. There are several ways in which this could have happened. Another inventor might have come up with the same idea, and published that. Or you might have put your product on the market already. Or someone might have published your invention with your permission. Then it would be too late for you to obtain valid patent protection.
8. Therefore, it is crucial to file for patent protection before your idea is disclosed to the public. By filing an application for a patent, you establish your priority date. Provided you can keep your priority date, nothing that is disclosed after that date can hurt your chances of obtaining a valid patent on your idea. (There is a minor exception, but I need not describe it because it is irrelevant for present purposes.)
9. What I have said is true not only in this country, but in most of the countries of the world. If you file for patent protection in a foreign country, the Patent Office of that country will recognise your British priority date, provided you comply with certain formalities. The countries of the world have agreed that it should be so, and have agreed to be bound by the same rules. They are laid down in Article 4 of the Paris Convention for the Protection of Industrial Property.
10. First, you must satisfy them that you really did file for protection in the U.K. on the date that you claim. This is easily achieved: you just produce a certificate from the U.K. Patent Office.
11. Secondly, you cannot hang on to your U.K. priority date indefinitely. People in other countries are entitled to know where they stand, and you are not allowed to keep them in a state of suspense. So, if you want to preserve your priority date, you must file for protection in the foreign country not more than 12 months later. If you miss the deadline, you lose your priority date there. You can still apply for foreign patent protection. But you run the risk that your invention may have been published in the meantime. In which case, you will have lost the opportunity to obtain a valid patent for that country.
12. So far, the rules are easy to understand. But now I must describe a third rule. Its justification is not so easy to understand – not in some respects.
13. In order for your claim to priority to be valid, it must be the case that the date you are claiming was the first such date. In other words, if it can be proved that you had already applied for patent protection for the same idea, before the date you are claiming, your claim will be disallowed. You will lose your priority date. This is so unless your earlier application has been abandoned, without being published, and leaving no rights to priority outstanding. What is more, it is necessary that your earlier application was abandoned no later than the date you filed your second application. It is this last bit which is not so easy to understand.
14. Consider what this means by way of an example. Suppose you file for patent protection in London on the first working day of the year: 2 January. Now you have a priority date of 2 January. After nearly a year has elapsed, you are not happy with your application. For example, you need more time to perfect your invention, or to raise finance. You do not want your invention to be published by the Patent Office, for then your competitors will get to know about it. Furthermore, if the Patent Office did publish your invention, it might prejudice your chances of getting proper patent protection for your idea later on – when you hope to have perfected your idea, or to have raised enough capital. You will anyway need enough capital to finance your foreign filing campaign, since this is very expensive. Anyway, you need to buy more time: you need a breathing space.
15. So, you decide to abandon your current application, without prejudice to your right to come back later. On (say) 31 December you inform the British Patent Office accordingly. They act on your wishes. Your patent application ceases to exist. It is never published. Your competitors do not get to know about it, except for the fact that you had filed some sort of patent application. Now you can come back anytime you like, for example the next day, and start all over again. You can claim a new priority date – the date when you came back and filed again. You have your right under the Paris Convention to establish your new priority date, provided you act within the 12 months, of course. Naturally, you have run the risk that somebody else has come up with the same idea in the meantime, and applied to patent it. But you may judge that the risk is small, and you may turn out to be right.
16. What I have just described happens all the time. Patent attorneys call it "regenerating your priority date". It is perfectly legitimate. Based on my experience, I would say that it must happen dozens of times every week, if not hundreds.
17. But note what I said: you must abandon your old application no later than the date you file your new application. Suppose, in the example I have just given, you abandoned your old application on 31 December, but filed your new application before doing so: say, on 30 December. Now Article 4 of the Convention will no longer work in your favour. On the facts being proved, you will lose your claim to priority. You might protest –: "But I am not claiming my old priority date (2 January). I just want my new priority date (31 December). What harm can there be in that?" Your protests will be of no avail. They will disallow your new priority date anyway.
18. Speaking for myself, I have never been able to see the merit in that aspect of the rule. If a person has abandoned a patent application, if it never has been published, if nothing has come of it which might affect the rights of other persons, how can it matter? What makes it particularly bizarre is this. In the ordinary way, nobody else will ever get to know about the situation – precisely because the application was never published. So, for it to become a live problem, would be sheer happenstance. In other words, the rule is capricious. At any rate, according to the way in which it normally interpreted. I can think of an alternative interpretation of Article 4 of the Convention which, to my mind, makes better sense. But it is not the commonly accepted interpretation, and the correct interpretation of Article 4 is not a matter which is before me now.
19. However, there is not a patent attorney in the land who does not know about this pitfall. They were taught about when they were students and they keep in constant practice. Therefore, if you employ a patent attorney to prosecute your application, you can be pretty sure that he will not allow you to fall into it. Of course, it is always possible that somebody can make a mistake. But it is such an elementary mistake that I would need fairly convincing proof before I believed it myself.

Mrs Haberman's Two Applications

20. A person who comes up with an invention and wishes to obtain patent protection will normally employ a skilled professional for that purpose. In 1991 and 1992 such professionals were known as Patent Agents. Mrs Haberman selected the well known and reputable firm of Lloyd Wise, Tregear & Co. I shall refer to them as Lloyd Wise for short. The partner in charge of her case was Mr Michael Deans. He had been a partner in the firm for more than 10 years. He was assisted by a trainee. No letters were allowed to go out unless read and signed by a partner.
21. Using the services of Lloyd Wise, Mrs Haberman filed an application for a patent at the Patent Office, London, on 8 April 1991. It was allocated the serial number 9107381. I shall refer to it as "GB-1".
22. It is relevant to note that GB-1 was filed under section 15 of the Patents Act 1977, that is, it did not have any claims defining the invention it was sought to protect. It is very common to use this procedure, and its purpose is to establish priority. It is also relevant to note that, according to section 15(5), the filing of claims at the Patent Office must be accomplished within the period prescribed by law, that is, within 12 months. If not, it is taken to be withdrawn. Thus the deadline was 8 April 1992. The surviving documents make it quite clear that Lloyd Wise were perfectly well aware of that fact (see the letter I shall quote below), which is only what I would expect.
23. Patent applications are normally published by the Patent Office about 18 months after the date of filing. When this happens, it is too late to recall them. Their contents become part of the state of the art. But this can be prevented by withdrawing the application in good time.
24. By March 1992 the situation was as follows. Mrs Haberman was still not happy with the final form of the design for a commercial embodiment of her invention. She had not found a suitable manufacturer for her feeding cup. She had no idea whether her invention would be of interest in other countries. She had no funds for foreign applications. In short, she needed more time – a breathing space.
25. On 11 March 1992 Mr Deans wrote her a letter of advice. So far as I can tell it was perfectly correct advice, standard in the circumstances. I shall quote portions of it:

You indicated that you are not yet happy with the final form of design for the trainer cup.

Meanwhile, we are faced with the approaching date of 8^th April 1992 by which we would normally have to complete the present application by adding claims and an abstract and requesting the standard statutory novelty search …

At this same stage we can add to the disclosure, albeit by actually filing a new application claiming priority from the original application. If your evaluation of the modified design is complete by that time, this would be the sensible course to adopt.

(i) So that was option No. 1: if you have perfected your design before 8 April 1992, file a further application describing your refinements, but keep your priority date (8 April 1991) for the basic idea you have already described in GB-1.

(ii) Then Mr Deans proceeded to advise Mrs Haberman about an alternative option, No.2. He wrote:-

If you are still not satisfied that you have what is effectively the final form of the trainer cup as we approach 8^th April 1992, you may prefer instead to withdraw it and to file a new application when your evaluation of the final form of design is complete. The danger, of course, is that such new application would only take its actual filing date so that in theory someone else might have had the same or a closely similar idea in the interval and so in effect have leapfrogged over you. There is nothing we can do about that possibility.

26. In other words: if you have not succeeded in perfecting your design, abandon your existing application GB-1 (sacrificing your priority date of 8 April 1991) and, when you do perfect your design, file a new application describing that. This will take a later priority date, but the risk will be worth running provided nobody else has come up with the same idea in the meantime. In short: regenerate your priority date.
27. He went on to advise her about a third possibility (using the current application as a device for flushing out the prior art.) I need not describe it further because, as is clear, Mrs Haberman did not avail herself of that option.
28. What is clear is that Mrs Haberman decided to choose option No. 2. We know for several reasons, including a file note in the handwriting of Mr Deans' assistant. Transcribed for present purposes, it reads:

(a) Mrs Haberman

We will be dropping [GB-1] and filing a new GB application as soon as possible not claiming priority.

29. The emphasis is in the original, and the only common sense interpretation is that Mrs Haberman had decided to abandon GB-1 and regenerate her priority date. The hand-written note is undated, but it must have been written after Mrs Haberman had read Mr Deans' letter of 11 March 1992. This is corroborated by a file copy of a letter from Lloyd Wise to Mrs Haberman which bears the date 3 April 1992; it reads:

Please find enclosed a draft patent application. This has been based on the original application [GB-1], but includes the additional figures and references thereto, and also a set of claims.

…

This substitute application [my emphasis] can be filed as soon as we have your approval by telephone.

30. Mrs Haberman must have given her approval, presumably by telephone, because on 7 April 1992, Lloyd Wise, on her behalf, filed another application for a patent at the Patent Office in London. It did not claim priority from GB-1. It was allocated serial number 9207766. I shall call it "GB-2". As foreshadowed, it contained the same description and drawings as GB-1, plus certain further description and drawings. I infer that the extra material comprised such further refinements as she, Mrs Haberman, had been able to come up with at that point in time.
31. Notice that, by 7 April 1992, it was too late for all practical purposes for Mrs Haberman to embark on a programme of filing for foreign patent protection while keeping the priority date of GB-1. In any case there is no suggestion that she did anything of the sort. The facts, as recounted so far, are entirely consistent with a decision to abandon GB-1 and regenerate her priority date: the date of GB-2.
32. GB-1 was indeed withdrawn. It was never published. It never served as a claim to priority. Henceforth, Mrs Haberman's claims to priority were based on GB-2.
33. The point underlying this case is: precisely when was GB-1 withdrawn? I shall return to that below.

Mrs Haberman's Patents

34. In due course Mrs Haberman did file a number of applications outside the U.K., claiming the priority date of GB-2 (i.e. 7 April 1992). They eventually matured into granted patents.
35. One of those was EP 0634922, granted by the European Patent Office. It is the one all the trouble is about. It covered a number of European countries, but not the U.K. (To get patent protection for the U.K. Mrs Haberman persisted in her application GB-2, which was granted by the British Patent Office. It was that British patent which was upheld by Mr Justice Laddie.)
36. Meantime, two American inventors had come up with an idea for a trainer cup which was – I use a neutral term – similar to Mrs Haberman's. They applied to patent it in the U.S.A., and their patent was granted on 16 February 1993: USA 5,186,347. So, provided Mrs Haberman could hold on to her priority date of GB-2 (7 April 1992), the publication of this American patent in 1993 would not hurt her chances of getting valid protection in most of the countries of the world.

The European Opposition

37. After a European patent is granted there is a period during which it can be opposed by anyone who disagrees with it. That is what happened to Mrs Haberman's EP 0634922. It was opposed by three companies: I shall call them Jackel, Playtex and OMSA. They opposed her patent on various grounds, but mainly on the ground that her invention was not new, or was obvious. They cited a plethora of prior art: prior proposals for making various sorts of feeding cups. In the end, the Opposition Division of the European Patent Office did not base their decision on those.
38. What the Opposition Division did do was to revoke Mrs Haberman's European Patent 0634922 on one single ground. They held that her invention was not new because the same idea had already been published in the American patent I have referred to: USA 5,186,347.
39. The reader might ask: how was this possible, seeing that the American patent was not published until 1993 – after the priority date of GB-2? The answer is this: the Opposition Division thought that GB-2 was no good as a claim to priority. They considered that her first application for patent protection was really GB-1. Even though GB-1 had been withdrawn, had not been published, and had not served as a basis for a claim to priority. The reason was because they thought that GB-1 had not been withdrawn until after GB-2 was filed. GB-2 was filed on 7 April 1992, but they thought that GB-1 was not withdrawn until 9 April 1992: two days too late.
40. Such decisions are quite rare. As I have already explained, once an earlier patent application has been withdrawn, without being published, the rest of the world can seldom know what were its contents. Unless those contents can be known, it is not possible to say that the earlier application was for the same invention, or a similar invention. In most countries of the world there is no procedural machinery for compelling disclosure of the contents of an abandoned and unpublished patent application. So, in practice, the point does not arise all that often. In English-speaking countries, however, there is such a procedure: it is called discovery of documents. In this case, Mr Deans suspects that Jackel (one of the opponents of Mrs Haberman's European patent) got hold of the information by having been the defendants in the case before Mr Justice Laddie: through the discovery process. If that were true it would have been a serious abuse, punishable by the laws of England, to use the information for the purpose of the European opposition proceedings without the permission of the English court. I shall say no more about it because Jackel are not parties to the case before me and they have had no opportunity to defend themselves against the accusation.
41. Jackel, having raised the point, withdrew their opposition on 4 April 2000. Leaving the opposition to be continued by Playtex and OMSA. As I have explained, the opposition was successful and Mrs Haberman's European patent was revoked. Mrs Haberman has appealed. I am told that the oral proceedings are going to take place on 3 June 2003.

The Twist in the Case

42. When was GB-1 withdrawn? The short, if surprising answer is: nobody knows for certain.
43. The intelligent lay person would say: "Why can't we find out by asking the British Patent Office? Surely they must have a record. It will put the matter beyond all doubt." However, he or she would be mistaken. The Patent Office has destroyed its pertinent records for the relevant period.
44. The Patent Office cannot keep all the papers that have ever been filed there. If they did they would have custody of hundreds of millions of pages. That would not be practical. So, over the years, the Office has adopted the practice of destroying papers it considers to be obsolete. There is no secret about it, and the practice has been described and defined from time to time in the Official Journal.
45. So the crucial information – exactly when was GB-1 withdrawn – is not available from the official records. Yet the European Patent Office needs to act on this information. And not only the European Patent Office, but any court or tribunal in the world which may have to decide the same question.
46. Notice this obvious absurdity in the international patent system. Even supposing it really mattered, in the great scheme of human affairs, precisely when an unpublished patent application was withdrawn (and I cannot see why), you would think that there would be a reliable method of finding out. There is none. Or rather, there was none at the time that concerns us in this case. Since 1 June 2001, the Patent Office database has recorded the precise date when patent applications have been withdrawn. Even so, there is no obligation to do so under the Paris Convention.

When Was GB-1 withdrawn?

47. As I have said, nobody knows for certain on what precise date GB-1 was withdrawn. The only information that we have comes from Lloyd Wise's files, which are incomplete, plus our knowledge of what patent agents normally did and of Lloyd Wise's office practice. As I believe I have made clear, we have pretty definite knowledge of intent to withdraw GB-1 in such a way as to receive the benefits of Article 4 of the Convention. What is not so clear is whether they implemented that intent.
48. I must preface the following remarks by noting that this was not a case of sending documents to the Patent Office by post. Lloyd Wise's office was in central London, as was the branch of the Patent Office where documents could be and were filed. There was a system for delivering documents to the Patent Office by hand. Thus, in general, documents could be delivered on a "same day" basis if so required.
49. GB-2 was filed on 7 April 1992. So, for the intent to be made effective, GB-1 had to be withdrawn on that same day, or sooner. Now, in Lloyd Wise's files is an unsigned copy of a letter to the Patent Office. It says:

The applicant is no longer interested in pursuing the abovementioned application, and therefore the application is hereby withdrawn.

50. Unfortunately, the document bears the date "9^th April 1992". If this truly reflects the date on which it was sent, they were too late. However, Mr Deans believes this dating is mistaken.
51. He explains why in his witness statement, which was made in 1999. Of course, this was 7 years after the events in question, and after such a lapse of time Mr Deans could not be expected to have an actual recollection of all those events, nor does he so claim.
52. However, what Mr Deans can remember are the standard practices of his office at the time. It is in the nature of the profession of patent attorney that he or she lives against a constant background of crucial dates which must not be missed. To miss even one is a nightmare. (For that reason I am glad I did not choose that branch of the profession.) Thus, in my experience, all firms of patent attorneys have standard procedures to try to ensure that it does not happen.
53. Another thing that Mr Deans can remember is his knowledge of the workings of Article 4 of the Paris Convention (which is similar to Article 87 of the European Patent Convention). He knew that – according to the usual understanding, anyway – you have to withdraw your client's earlier application no later than the date on which you file her new application, if you wish to regenerate her priority date and preserve her rights under the Conventions. As he says in his witness statement: "Indeed, that requirement was trite law and practice to a patent agent". That would accord with my experience. He testifies that the practice was well known to his assistant.
54. Here are Mr Deans' reasons for disbelieving the date "9^th April 1992" on the copy of the letter of withdrawal.

1. It would have been pointless to write such a letter on 9 April because GB-1 would have died a natural death anyway on 8 April, the day before. This is because GB-1 was filed without any claims defining the scope of the invention; as is allowed by section 15 of the Patents Act 1977. However, it is allowed on condition that claims are filed within the prescribed 12 month period. In other words, by 8 April 1992. As I have shown above, Lloyd Wise were well aware of that requirement and were planning what to do about it in March. I would add this. To dispose of an unwanted application by letting it die its natural death (and not by writing an unnecessary letter) is standard procedure in the world of patent prosecutions.

2. Like any other firm of patent agents, Lloyd Wise had a Records Department. It is the job of such a department to make sure that crucial dates are not missed. Lloyd Wise's records showed that GB-1 had to be actioned by or before 8 April 1992, certainly before 9 April, or the case would be lost. As Mr Deans puts it in his witness statement:

[Our] records department would have been constantly reminding us that action had to be taken on this application by 8 April 1992. This is, of course, standard procedure in a patent agents' practice because of the risk of dates being missed. At [Lloyd Wise], no matter how senior the individual, the records section were empowered to badger that individual until such time as they have been satisfied that the appropriate action has been taken. Thus Sarah Roques [his assistant] and myself would have been fully conscious of this date.

I have stressed some words in the above text because they fully reflect what I would have expected to happen based on my own experience. The word "badger" is a strong one in the English language: it means pester or harass, only more so. That records staff are empowered to badger anyone, even the Senior Partner, is because the alternative is even worse – at best, constant anxiety and loss of sleep. That is the reality of the situation.

3. The above is corroborated by the witness statement of Mr John Miller, the head of Lloyd Wise's records department, and by the lady who was Mr Deans' secretary at the time, Ms Teri Collins. Ms. Collins had good reason to be acquainted with the practice at that office. She testifies that:

We would have been receiving from our Records Department reminders that action had to be taken in relation to [GB-1]. Those reminders would have increased in frequency and urgency as 8 April approached.

Mr Miller testifies that by 1992 Lloyd Wise's records system was computerised. He says that under this system:

All partners and their secretaries are provided at the beginning of each week with a detailed list of all pending dates and the actions which need to be taken with the earliest dates at the top of the list. That will include a list of all actions which need to be taken that week. The latter element (i.e. actions to be taken that week) is then updated by means of lists circulated to each partner responsible and their secretaries on a daily basis.

He explains that there was a firm rule by which the Records Department had to be satisfied that all deadlines had been complied with by the date in question or a satisfactory explanation obtained from a partner. There was a system in place to make sure this rule was obeyed. Like annoying barking dogs, the staff of the Records Department are not allowed to give up unless and until a partner calls them off. He must give a satisfactory explanation for doing nothing. Mr Miller goes so far as to say that "there was no possibility of the date simply being missed or overlooked by default". It should be noted that Mr Deans was himself the partner with overall responsibility for the Records Department: as the reader will have gathered by now, such responsibility is not allocated lightly.

4. As a further incentive to memory, the files at Lloyd Wise always bore the filing date in large letters on the cover. The file for GB-1 was no exception: see exhibit "MJPD12". It was standard practice that any partner who signed a letter – and none but partners could do so – would be supplied with the file. In my opinion, the date (8 April 1991), when presented to any partner in early April 1992, would burn in his mind like a hot wire.

5. During the litigation against Jackel certain papers of Lloyd Wise were lost, possibly by a photocopying agency. Mr Deans remembers this because the events happened in November 1998, less than a year before he made his witness statement. One of the missing papers that Mr Deans remembers is a hand-written note from him to his assistant instructing her to withdraw GB-1 and then file GB-2.

55. All those facts strongly suggest, to my mind, that Lloyd Wise intended to withdraw GB-1 on 7 April 1992 and had a system in place whose purpose was to ensure that such intentions were respected. It is a presumption employed by the laws of many countries that Omnia praesumuntur rite esse acta. That is to say, formalities are presumed to have been complied with until the contrary is shown. This maxim is based on common sense and reasonable probability. Once it is established that a person intended to do a formal act – as Lloyd Wise intended here – it is reasonable to suppose that he in fact did so, unless the contrary is shown. This is reinforced if the evidence shows that he had a system in place whose object it was to ensure that such formalities were indeed complied with. Again this is common sense and probability. Such facts do not prove that the formality certainly was complied with. For it may be possible to show, on the balance of probabilities, that an exception occurred.
56. That a letter was written on the date it bears is itself an illustration of this principle. Here the surviving copy of the letter to the Patent Office, withdrawing GB-1, bears the date "9^th April 1992". If that was the only fact that we knew, it would establish, on the balance of probabilities, that the letter was written on that date, and so could not have been delivered to the Patent Office on 7 April. But we all know that sometimes letters are not written on the date they bear. Thus it is itself only a presumption, and may be rebutted by more compelling evidence.
57. In this case there is evidence that could explain why the surviving copy of the letter is dated "9^th April 1992". There are reasons for supposing that Mr Deans was out of the office on 8 April because he was buying a house at the time. In such case, if his secretary was typing the letter in the afternoon of 7 April 1992, she might well have dated it "9^th April 1992" in the belief that he would sign it on his return to the office on 9 April. This is a clear example of what we all know from our experience of the world: it is commonplace for a letter to be typed on a day that is earlier than the date it bears. Having typed the letter, Mr Deans' secretary would photocopy it and put it in the file. It is that document which we now have. But it does not necessarily prove that the letter was, in fact, sent to the Patent Office on 9 April 1992. Mr Deans clearly intended the letter to be sent no later than 7 April, and it is perfectly possible that he re-dated it by hand, or sent it on 7 April without noticing it bore the wrong date. This particular evidence, of itself, does not rebut the presumption raised by the date borne by the surviving copy letter. But it serves to weaken it. Then, the rest of the evidence, which strongly suggests that Lloyd Wise intended to withdraw GB-1 on 7 April and had a system for implementing such intentions, may show, on a balance of probabilities, that GB-1 was withdrawn on 7 April 1992, and not 9 April 1992. If the Patent Office records had survived we could know for certain. But, through no fault of Mrs Haberman's, they do not. All we have to go on are the probabilities of the situation.
58. In the decision of the Opposition Division of the European Patent Office it is said that Mrs Haberman had failed to provide "concrete proof" that Mr Deans withdrew GB-1 in time. It is not clear to me why the onus of proof should necessarily lie on Mrs Haberman. And if it does, it is not clear to me why, according to the practice of the EPO, the standard of proof should be "concrete". See Case Law of the Boards of Appeal of the European Patent Office, 4^th edition, pages 357-359. However, that is not a matter for me, but for the Board of Appeal of the European Patent Office.
59. Mr Colin Birss, who appeared for Mrs Haberman before me, invited me to decide the point myself if need be. I have some sympathy with that request. If we try to "see ourselves as others see us", we shall surely appreciate that the present position is somewhat absurd. Every time the priority date of GB-2 is brought into question, not only in the European Patent Office, but perhaps in other jurisdictions where they are not so familiar with the practices of European patent attorneys, it must be considered afresh. Would it not be far better if there was some machinery for deciding it centrally? For example, from the records of the British Patent Office? True; but the machinery does not exist; and I cannot create one by a process of judicial invention. In any case, I have not heard one of the parties to the EPO case, namely OMSA, because they are not before me. Playtex were parties before me, but quite properly their counsel, Mr Michael Hicks, pointed out that the matter was not for me to decide. I must therefore decline Mr Birss' request. Mrs Haberman must trust to the good sense of the Board of Appeal of the European Patent Office, and of such other tribunals as they question may come before. I cannot usurp their functions.

Mrs Haberman's Request for Information

60. In an attempt to procure a central decision of this international question, Mrs Haberman's advisers have seized on Section 118 of the Patents Act. This is a section which requires the Comptroller of the Patent Office to provide certain information on request. In practice the information concerns the dates on which certain events happened, and a form must be filed for the purpose: see Rule 92 of the Patents Rules 1995.
61. Mr Birss argued that the Comptroller is obliged to provide information as to when an application for a patent was withdrawn, see Rule 92(1)(c). He said that normally the Comptroller should be able to provide it from the information in her files but, if the file no longer exists, she should take steps to fill the gap, receiving evidence if necessary. Mr Birss conceded that his request might appear bizarre at first sight, seeing that it would be a case of the Comptroller furnishing information to the party who is the source of that information in the first place. But he said that the section is worded quite generally, and the boot might have been on the other foot. What if it was Playtex who was requesting this information? Could the Comptroller decline to provide it, in breach of her statutory duty? Surely the Comptroller should, if necessary, require witnesses to attend at the Patent Office in order to arrive at the right answer? (See Patents Act 1977, section 123(2)(d), and Rule 103(3), which empower the Comptroller to require the attendance of witnesses.)
62. I do not believe this submission is well founded. The basic mistake that people have been making in the case before me is to use the Rules to try to interpret the Act. In my judgement, this is not permissible. The Act was made by the legislature of this country, but the Rules were made by the executive, and only because the legislature empowered them to do so. The executive cannot, by making rules, alter the meaning of the Act. The Act was passed by Parliament on 29 July 1977 and, whatever its wording meant then, it means the same now. One should start by looking to see what words Parliament used and what they meant, without having one's judgement obscured by what the rule-makers did later on.
63. So, although it is true that Rule 92 is confined to the dates when various events happened, section 118 is not. Its language is general. Thus section 118(1) says that the Comptroller shall give the person making the request "such information … relating to the application as may be specified in the request, subject, however, to any prescribed restrictions". Purely as wording, this is ambiguous. Ignoring the "restrictions" for the moment – for they are to be prescribed in rules that, as of 29 July 1977, have not yet been made – it could mean that the Comptroller must provide any information whatsoever, provided it relates to the application. Or it could just mean that the Comptroller must provide such information as she can gather from her own official records.
64. Both meanings are possible as a matter of sheer semantics. But it is rather hard to believe that Parliament was authorising the rule-makers to require the Comptroller to provide any information whatsoever, so long as it was connected with a patent application. As an absurd example: how much money did GB-2 make for its proprietor? A much more rational theory, therefore, is that section 118 is concerned with information which is available from the Patent Office's official records. That is what I hold.
65. I do not mean to say that the Comptroller is confined to providing information stored in this or that register or computer database. If it were shown that the "official" version must be incorrect – say, the date "29 February 2003" – or open to reasonable doubt, I expect that her officers would arrive at the right version, using the internal resources of the Office. A failure to do so would not be beyond challenge. But that is not the point before me now.
66. This interpretation avoids some of the difficulties which would arise if I had adopted Mr Birss' submissions. Thus his interpretation would require the Patent Office to supply information which it did not have. He did not shrink from the logical conclusion: that this could require obtaining information from an outside person, precisely in order to re-supply that information to that very person. This very case.
67. Before there can be any question of requiring witnesses to attend, there must be some proceeding on foot or in contemplation in which their evidence might be received. Such proceedings are mentioned in several places in the Act and Rules, and they all fall into one of two classes. First, there are what are commonly known as ex parte proceedings: matters entirely between a citizen and the Patent Office, concerning which no others have a right to attend or be notified. An example of such a proceeding is a complaint by an applicant for a patent that his application has been turned down. Secondly, there are what are called inter partes proceedings: in which outsiders can take part. An example of that is an opposition to a patent. In the case of this second class, the Act provides machinery by which outsiders can be notified, so that they can take part if they want to and protect their positions. Or, at least, whereby they can have the decision reversed, as in section 34. Section 118 says nothing about such matters.
68. In order to allow them to be heard, Mrs Haberman's advisers have named Playtex as a party to this appeal. They have not named OMSA, who are still parties to the proceedings before the European Patent Office. Furthermore, in my judgement, if section 118 means what Mr Birss contends it means, her advisers need not have notified even Playtex. The Patent Office would still be obliged to furnish the information to Mrs Haberman.
69. What Mr Birss is really doing, in my judgement, is trying to make section 118 into what Parliament did not intend it to be, namely, a procedure for ascertaining facts, not merely for giving them. "Give" is the word that Parliament actually used.
70. I believe that Mr Birss' contention fails for another reason. I hold that section 118 does not entitle persons to demand the dates of the withdrawal of unpublished applications at all. Thus section 118 says:

(1) After publication of an application for a patent in accordance with section 16 above the comptroller shall on a request being made to him in the prescribed manner and on payment of the prescribed fee (if any) give the person making the request such information, and permit him to inspect such documents, relating to the application or to any patent granted in pursuance of the application as may be specified in the request, subject, however, to any prescribed restrictions.

(2) Subject to the following provisions of this section, until an application for a patent is so published documents or information constituting or relating to the application shall not, without the consent of the applicant, be published or communicated to any person by the comptroller.

(3) Subsection (2) above shall not prevent the comptroller from –

(a) sending the European Patent Office information which it is his duty to send that office in accordance with any provision of the European Patent Convention; or

(b) publishing or communicating to others any prescribed bibliographic information about an unpublished application for a patent;

nor shall the subsection prevent the Secretary of State from inspecting or authorising the inspection of an application for a patent or any connected document under section 22(6) above.

(1) Where a person is notified that an application for a patent has been made, but not published in accordance with section 16 above, and that the applicant will, if the patent is granted, bring proceedings against that person in the event of his doing an act specified in the notification after the application is so published, that person may make a request under subsection (1) above, notwithstanding that the application has not been published, and that subsection shall apply accordingly.

(2) Where an application for a patent is filed, but not published, and a new application is filed in respect of any part of the subject-matter of the earlier application (either in accordance with rules or in pursuance of an order under section 8 above) and is published, any person may make a request under subsection (1) above relating to the earlier application and on payment of the prescribed fee the comptroller shall give him such information and permit him to inspect such documents as could have been given or inspected if the earlier application had been published.

71. Notice how the section is arranged. Subsection (1) requires the Comptroller to furnish information concerning published applications. GB-1 is not a published application. Subsection (2) affirms the general principle that information concerning an unpublished application cannot be revealed without the permission of its proprietor. The remaining subsections concern exceptions to that general principle. Notice the most pertinent exceptions.
72. Subsection (3)(b) permits, but does not require, the Comptroller to furnish 'bibliographic' information. Although I do not mean to use the Rules to interpret the Act, examples of what the rule-makers regarded as 'bibliographic' are to be found in Rule 95. Incidentally, although it includes the fact that a given application has been withdrawn, it does not include the date when. I mean to derive nothing from the narrowness of Rule 95 for the purposes of this judgement. The point is that subsection (3)(b) is permissive, not mandatory.
73. Subsection (4) concerns the case of the proprietor of an unpublished application who has threatened someone with legal proceedings. The Act intends to discourage groundless threats of patent proceedings, see section 70, and so section 118(4), in order to protect the person who is threatened, enables him to find out what he is being threatened with. Here Parliament has chosen to override the applicant's interest in the secrecy of his application, in order to protect the wider public. It was him who chose to threaten in the first place.
74. Subsection (5) concerns unpublished applications. Mr Birss says it fits this case. Thus a 'new application' has been filed (GB-2); it is about a 'part of the subject-matter of the earlier application' (GB-1); it (GB-2) has been published; and a request has been made for information about GB-1. His opponents, Mr Vanhegan for the Comptroller and Mr Hicks for Playtex, ask: what about the words in brackets? What about "either in accordance with rules or in pursuance of an order under section 8 above"? We know the case has nothing to do with section 8 (which concerns disputes about ownership of patent applications). What are the "rules"?
75. To which Mr Birss says: all patent applications are filed in accordance with rules. So subsection (5) concerns any case where there is an unpublished application, followed by an application which is published, overlapping subject-matter between them, and a request by anyone to know what is in the unpublished application.
76. I have no hesitation in rejecting this submission. The subsection says nothing about the owners of the unpublished and published applications needing to be the same person. So if Mr Birss is right, they could be anyone. This would mean that anyone who has a published patent application could demand to see the contents of his business rival's unpublished and abandoned applications, provided they had some subject-matter in common – say, they were both about brewing beer. I can discern no rational motive for such a law. On the contrary, if it were so the general principle that people are allowed to keep their unpublished applications to themselves, see subsection (2), would be almost completely undermined. Mr Birss says, not so, because this would be prevented by the Rules. That is again using the Rules to interpret the Act, an exercise which I believe is forbidden. In any case, the construction is arbitrary and bizarre.
77. In my judgement, "in accordance with rules" is a reference to people making what are called divisional applications, see section 15(4). This says:

(4) 'Where, after an application for a patent has been filed and before the patent is granted, a new application is filed by the original applicant … in accordance with rules in respect of any part of the matter contained in the earlier application … the new application shall be treated as having, as its date of filing, the date of filing of the earlier application.'

A typical instance would be a person who finds that his original application disclosed two different inventions and that he is not allowed to cover both in one patent. So he "divides out" so as to get two. Since both have a common origin, both are allowed to retain the priority date of the original.

78. So, in my judgement, and putting it shortly, the true purpose of section 118(5) is to deal with those special cases where, in the course of patent prosecutions under the Act, what was originally a single application is split into two or more applications. This can arise for example under section 8 (as where two inventors are in dispute about ownership and the Comptroller, like Solomon, decides to split the subject matter in two, allowing each to keep its original priority date), or under section 15(4). In those cases it is of course desirable, in order to protect the interests of third parties, that they should have access to the original application, so that they can check to see that the new claim to priority is well founded. Their interests are considered to override the interests of secrecy.

Result

79. The appeal must be dismissed. The Respondents are entitled to their costs. They claim £2,561.15p for the Comptroller and £3,078.65p for Playtex. I find them to be reasonable, and assess them as claimed, payable in 14 days.

Conclusion

80. It will be fairly obvious that I have a great deal of sympathy for Mrs Haberman. Her predicament is a disgrace. If I could help her, I would. But I cannot bend the law. As I have said earlier in this judgement, her recourse is to trust to the good sense of the Board of Appeal and those other tribunals before whom the question of her priority date may come.