Oxonica Energy Ltd v Neuftec Ltd

How do we interpret a formal commercial agreement if it is ambiguous and we have reason to believe that its draftsman did not have a deep understanding of the relevant law? I think that is what this case is about.

The secret of drafting legal documents was best described by Nicolas Boileau, who was not only a literary critic but a qualified lawyer: “Ce que l'on conçoit bien s'énonce clairement et les mots pour le dire arrivent aisément”. What one conceives well can be stated with clarity and the words to say it come easily. We should all have that framed and displayed on our desks. But too often the opposite precept is followed. Bits of legal boilerplate are bolted together so that it is the words that are allowed to shape the concept instead of the other way round. In that regard the invention of the word processor has worked wonders. Sometimes, I fear, it has dispensed with the ‘concept’ altogether. Misfortune not infrequently follows.

Those who draft intellectual property licences may learn something from the misfortune that has befallen the parties to this case. They have entered into a licence agreement that contains a crucial phrase which is exceedingly hard to interpret. I have changed my mind several times about its meaning – it is a veritable Necker cube of licence agreements. The result is business uncertainty and costly litigation.

I shall call the parties “Neuftec” and “Oxonica”. On 7 December 2001 they signed a Licence Deed. Under this, Neuftec granted to Oxonica an exclusive patent and knowhow licence. Amongst other remuneration Neuftec is entitled to a royalty on sales of what are called Licensed Products. The question I have to decide is: what is meant by “Licensed Products” in that context?

Oxonica contends that it does not have to pay any royalties unless it sells a product in a country where Neuftec has a pending patent application or a granted patent and the claims thereof cover the product in question. Only such a product is a Licensed Product. On this approach, the answer to the question “Does Oxonica have to pay a royalty to Neuftec on this sale?” can vary from country to country. It can also change in time within any particular country e.g. if a patent application leads to a granted patent with narrower claims.

Neuftec ripostes that that is to treat the Deed as a pure patent licence, which it is not. It is also a knowhow licence. The opportunity to use the knowhow exists irrespective of any local patent privileges. It is true that the Deed defines “Licensed Products” in terms of patent claims, but that is just a convenient way of defining the technology in respect of which royalties are payable. Thus the scope of the obligation to pay royalties is uniform and does not depend on the country in which the sale takes place.

Each of those cases is plausible in its way; but each has its own problems.

There is quite a lot of evidence – including copious exhibits – about the events leading up to the signing of the Licence Deed. Quite a bit of it is irrelevant or inadmissible, or concerns propositions that are no longer advanced by counsel.

Needless to say – and absent an application for the rectification of the document, which is not made here – it does not matter what a party thought an agreement meant when he signed it. Nor does it matter what he was negotiating to get. What matters is what the document would convey to a neutral and reasonable person. But no document, nor any other statement made by human beings for that matter, can properly be understood except in its context. Therefore what I have to decide is what the Licence Deed would convey to a reasonable person who has all the background knowledge that would reasonably have been available to the parties in the situation in which they were at the time of the agreement. This knowledge includes everything which would have affected the way our reasonable man would have understood the language of the document. However as already stated it does not include any knowledge of the parties’ subjective intentions, or of their previous negotiating positions. (A more flexible approach has been argued for (Footnote: ¹), and the boundaries of the latter principle are unclear (Footnote: ²); but I do not consider anything turns on this that is going to help me to decide this case).

That is the modern approach to the interpretation of commercial documents and it was explained by Lord Hoffmann in Investors Compensation Scheme v. West Bromwich Building Society [1997] UKHL 2; [1998] 1 WLR 896, 912-913. I shall revert to that later.

The background knowledge that the neutral, reasonable person employs when understanding a commercial document can include a knowledge of the relevant law (BCCI v. Ali [2001] UKHL 8 at §39, [2002] AC 251 at 269, H.L.). Indeed a person could not properly understand the Licence Deed in this case unless he had a good general knowledge of patent law and international practice in that regard. The problem is that the more of that knowledge he had the more he might be baffled by the document. I shall deal with that later.

I must say something about proof of foreign patent law. For the limited purpose I have in mind the parties do not have to prove it. Increasingly the Patents Court takes judicial notice of patent law in other countries. In this day and age it would be insular not to. So much so, that in Celltech Chiroscience Ltd v. MedImmune Inc [2002] EWHC 2167 (Pat) Jacob J with the consent of the parties tried two questions of US patent law – file wrapper estoppel and argument estoppel, neither of which exists under our domestic law – without needing to receive expert evidence of US law:

Although the history of the previous negotiations is not relevant, as I have said, in the following narrative I shall mention some of those matters all the same. That is because I shall need to deal with a particular argument advanced on behalf of Neuftec.

In 1996 Mr Bryan Morgan started a company called Celox Limited. I should explain at once that Celox was the predecessor in title of Neuftec, the defendant in our case. When I write Celox, please think Neuftec. Celox was interested in fuel catalysts. In December 1999 Mr Ronendra Hazarika became a director of Celox. Mr Hazarika had a background in the motor racing industry. Celox decided to produce an additive for diesel fuel: a catalyst that could be dispersed in the fuel and would reduce fuel consumption and engine emissions.

Celox started to do research. In the periodic table there is a series we used to call the rare earth elements (although we are supposed to call them the lanthanoids now). Anyway, Celox thought the lanthanide oxides might be good for the job. Of those, cerium oxide is the most abundant and therefore, other things being equal, the most attractive. The use of lanthanides as fuel additives had been suggested before but they were known to clog the filters and cause abrasion and blockages in engines.

Celox thought that the way forward would be to use extremely small particles of cerium oxide coated with a lipophilic (oil-loving) substance to make them disperse in the fuel. How to manufacture such particles? It occurred to Celox to approach a nanotechnology company. They had read about one in a trade journal: Oxonica Materials Limited. It was one of those spin-off companies at Oxford. As we shall see, it is now the parent company of the Oxonica in our case. To avoid confusion I shall call it “Oxonica Parent”.

Celox did approach Oxonica Parent and thereafter the parties dealt together on a confidential basis. At that stage Oxonica Parent employed or had access to scientists who knew about nanotechnology. But it lacked general commercial experience (see below) and it has not been suggested that it had any experience of fuel additives or that it had contemplated entering that field before. A joint venture was proposed. Celox’s knowledge about fuel additives (and, I dare say, the markets for such additives) would be married to Oxonica Parent’s knowledge about nanoparticles. A joint venture company was incorporated. Celox Limited and Oxford Parent were to be 50:50 owners and to share the benefits accordingly. Oxonica Parent’s scientists started to do research, adding to the information which Celox had brought.

Celox Limited had applied for patent protection in this country on 29 June and 13 September 2000. Those patent applications were withdrawn and were not published but they served as a basis to claim international priority. On 29 June 2001 Celox made an international patent application pursuant to the provisions of the Patent Cooperation Treaty.

As we shall see, ‘international patent application’ is the official phrase and everybody uses it; but it is somewhat misleading. For the moment I shall call it “the PCT”. All we need bear in mind for the moment is that a PCT, as such, is not an actual request that a patent be granted; that it is not converted into one unless and until it enters the ‘national phase’; that it is customary to delay entering the national phase for as long as the law allows; and that the deadline for doing so at that time was 30 months (Footnote: ³) from the priority date.

Claim 1 of Celox Limited’s PCT as filed reads as follows:

That, then, was the broad idea. The document also taught certain refinements on that idea. For instance, that the particles size should be in the range 1 to 50 nanometres. That is very small indeed. A nanometre is a millionth of a millimetre. The document also taught that the particles should be coated with a lipophilic (oil-loving) substance – to make them disperse better in the fuel oil. It gave various examples of such substances.

In June 2001 Oxonica Parent was advised that a 50:50 joint venture arrangement could prejudice certain advantages it would otherwise enjoy under the Enterprise and Investment Scheme. It probably had to do with tax, but that is of no present consequence. Anyway, it was decided to adopt a different arrangement. Instead of a joint venture company, Oxonica Parent would set up a wholly-owned subsidiary and that subsidiary would exploit the technology. (In the upshot, that subsidiary is the Claimant in this case and I shall sometimes call it “Oxonica Subsidiary” in order to avoid confusion.) Celox would licence its intellectual property to Oxonica Subsidiary which would pay Celox some money by way of an upfront payment, royalties, a proportion of profits and a milestone payment.

In essence, that was what was eventually agreed; except that, for a reason I need not describe, Neuftec Limited was put in the place of Celox. (Mr Morgan and Mr Hazarika were also directors of Neuftec.) The arrangement was effected by two documents dated 7 December 2001: (1) a Main Agreement covering the operation of the business and (2) an exclusive licence (the Licence Deed). Oxonica Subsidiary (then bearing the unmemorable name “RMBKNE5 Limited”) was made a party to the Main Agreement and was the licensee under the Licence Deed. Shortly before that date, the PCT was assigned by Celox to Neuftec.

Mr Hacon, who appeared for Neuftec, contended that any ambiguity in the Licence Deed should be resolved in his client’s favour, seeing that the revised arrangements were substituted for the earlier joint venture proposal in order to accommodate Oxonica Parent’s tax position, and for no other reason. I think his point really is this: that on Oxonica’s interpretation of the Licence Deed the benefits to Neuftec would be inferior to those they would have got under the joint venture, yet we know that Oxonica Parent would have been willing to undertake the joint venture but for the tax problem; therefore Oxonica’s interpretation is not plausible. I do not believe Mr Hacon is right on this point. The joint venture arrangement was never entered into and it is just part of the previous negotiations. We want to know what the parties did eventually agree, and not what they might have been going to agree, but did not: see paragraph 9 above. As a matter of fact, we do not know that Oxonica Parent would have been willing to undertake the joint venture – they were merely thinking about it and in the end they did not do it. And when the joint venture proposal was abandoned there was no binding agreement to negotiate something of equal value to Neuftec. In any case the joint venture proposal could not be a useful comparator. There was little detail and the agreement the parties did eventually make is different in kind. If there had been a joint venture it might or might not have involved outside equity investors and I do not know what effect the dilution might have had.

At the date of the Licence Deed Celox Limited had not developed a fuel additive to a commercial standard. It was Oxonica’s scientists who took over Celox’s work and did develop it to that standard. In the evidence there is a dispute concerning the importance of Celox’s knowhow. Oxonica says it was limited. In these proceedings I cannot weigh the relative merits of the parties’ respective contributions; nor do I have to. For it is not disputed that when Celox approached Oxonica Parent it was possessed of a body of knowhow about lanthanide oxide fuel additives – including test results on motor vehicles – whose significance cannot be dismissed. Nor, at the hearing before me, did Oxonica contend that Neuftec failed to impart that knowhow to Oxonica (Footnote: ⁴).

In Coco v. A.N. Clark (Engineers) Ltd [1969] RPC 41, 48 Megarry J said that

I therefore hold that a prudent company in the position of Oxonica at the date of the Licence Deed would anyway have wanted a knowhow licence if only to act as a safety shield.

The PCT was published on 3 January 2002. At some date in 2002 with the consent of Neuftec the prosecution of patent applications was taken over by Oxonica and paid for by them, but they kept Neuftec informed. There is no complaint that Oxonica deliberately or negligently caused narrow patent protection to be obtained.

After the documents had been signed, Oxonica exploited the knowhow passed to it under the Licence Deed and developed a commercial fuel additive called “Envirox”. Envirox was manufactured on Oxonica’s behalf by another company. Sales were made and royalties were paid to Neuftec under the Licence Deed. Eventually Oxonica Parent was floated on the London Alternative Investment Market.

Patents were eventually granted to Neuftec in a number of countries. Some applications are still pending. However, there are important countries where the granted patents all have claims that are substantially narrower than those of the PCT. That is because the examiners found some relevant prior art. European Patent 1,299,508 which was granted on 12 January 2005 is said to be illustrative. Its broadest claim is for a method of improving the efficiency with which fuel is burnt or of reducing emissions

In June 2006 Oxonica thought it was going to secure a lucrative contract with a Turkish oil company. This potentially involved the supply of a large amount of Envirox and correspondingly large payments to Neuftec under the Licence Deed. However, at a meeting in London on 5 October 2006 officers of Oxonica informed Neuftec that Oxonica had decided to obtain the Envirox to be supplied under the Turkish contract from a different source. Moreover, Neuftec was told, the new Envirox – referred to as “Envirox 2” in the evidence – fell outside the claims of Neuftec’s patents – so no royalties would be paid. Envirox 2 from that source is said to be better and cheaper than the original Envirox.

It is now common ground that Envirox 2 falls inside the broadest claim of the PCT, but at least in some important countries it falls outside the broadest claims of the granted patents. It falls outside the latter because the oil-loving substance with which the particles of rare earth oxide are coated does not belong to the delimited class i.e. alkyl carboxylic anhydrides.

Oxonica now sues for a declaration that Envirox 2 is not a Licensed Product as defined and that sales of it in the United Kingdom or in any other territory to which a granted patent relates with claims in the same or substantially the same form as European Patent (UK) 1,299,508 do not attract royalties under the Licensed Deed. Neuftec counterclaims for an audit in respect of sales of Envirox 2 and payment of all sums due.

In the Licence Deed (and I shall come to it in detail) the expression Licensed Products, upon the sale of which royalties are payable, is defined as “any product, process or use falling within the scope of claims in the Licensed Application or Licensed Patent”. That is why Neuftec says “Either will do: it is enough that Envirox 2 falls within claim 1 of the PCT application”, whereas Oxonica says “Not any more; it now needs to fall within claim 1 of the patent”.

Neuftec, a company that was in possession of knowhow of substantial value and of an international patent application concerning fuel additives, entered into a licence agreement with Oxonica. The fuel additive that Oxonica now sells is covered by claim 1 of the international patent application, but not by the claims of patents that have been granted in a number of important countries. Oxonica refuses to pay royalties on this product in respect of sales in those countries because it says that it is now the granted patents that matter, and not the international patent application.

In this case I must interpret a formal commercial agreement that purports to specify the position in great detail, that contains definition clauses, that uses terms of art known to experienced patent lawyers, and that was professionally drafted. It used to be thought that courts were obliged to interpret formal commercial agreements of that sort according to the literal meaning of the wording, unless they were ambiguous, in which case the interpretation that best accorded with business common sense was to be preferred; but if not ambiguous, the literal meaning had to be taken even if it defied common sense. This approach, which did have some advantages (Footnote: ⁵) but demanded a high standard of draftsmanship and sometimes led to injustices, was gradually abandoned in the 1970s and 1980s; but to my knowledge the first to restate the law on this topic in a principled way was Lord Hoffmann. He did so in two cases decided in 1997, Mannai Investment Co Ltd v. Eagle Star Assurance [1997] UKHL 19, [1997] AC 749 and the Investors Compensation Scheme case already cited in this judgment.

In Mannai Lord Hoffmann pointed out that the old approach was founded on a mistaken and obsolete theory of human languages. Thus at 778D he referred to:

In that case a tenant wanted to terminate his office lease by giving notice to the landlord. Such notices, to be effective, must be unambiguous and must comply strictly with the terms of the lease; and what had to be done to comply with this lease was to give at least 6 months notice in writing such notice to expire on the third anniversary of the date of commencement, which was 13 January. The tenant did give more than 6 months notice, but by mistake his letter said “We hereby give notice to determine the lease on 12 January”. Anyone – including the landlord – could have seen he must have meant 13 January. The Court of Appeal following the old-fashioned approach held that his notice was ineffective because, as a matter of language, 12 cannot possibly mean 13. The fallacy there, as we now know, is that when we seek to interpret an utterance we are interested not so much in the meaning of language but in the meaning that is conveyed by a person’s discourse. Therefore the House of Lords held that in the context of that discourse 12 could mean 13 – could mean nothing else – and it allowed the appeal.

Thus the function of language is to convey meaning to the person to whom it is addressed; and in the case of a commercial document, the one to whom it is taken to be addressed is the reasonable person with an adequate knowledge of the background. But he knows that people sometimes use incorrect language and from his knowledge of the background he infers the correct meaning. As Lord Hoffmann said at 774D-F:

Although we have no difficulty in understanding Mrs Malaprop in that instance, her use of language puts us on notice that here is a person who is none too well acquainted with the aquatic fauna of the Nile, a river where not only are there no allegories to be found, but no alligators either. In somewhat the same vein, we quite often come across agreements that concern specialised legal subjects – and intellectual property licences fall into that category – which have been drafted by non-specialists. We can tell this from the way that technical expressions have been employed in the document. Sometimes those documents contain crucial ambiguities, not so much because of the misuse of the technical expressions, but because the draftsman had not enough understanding of the subject-matter to beware of the ambiguities that ought to be avoided. Such draftsmen, inspired by their word processors, are sometimes apt to cut and paste choice phrases and expressions where it was not necessary to do so. The result was foretold by Lord Hoffmann in the Investors Compensation Scheme case at 917A:

When a serious ambiguity occurs a court cannot resolve the difficulty as the House of Lords did in Mannai Investments, where the reasonable and informed reader could see that there was but one interpretation that could make any sense. Two meanings are possible, perhaps even plausible; but the court cannot just give up. All it can do is to find the interpretation that best accords with business common sense. And, in doing that, the court will not, in my judgment, be overly swayed by the consideration that it is “doing violence to the natural meaning of the language” if, as Lord Hoffmann pointed out in the later case, it would appear that it was the author of the document who did the violence. It was there held by the majority that in writing “Any claim (whether sounding in rescission for undue influence or otherwise)” the draftsman ought to be taken to have meant “Any claim sounding in rescission (whether for undue influence or otherwise)”.

I have said that the reader of the Licence Deed would need to have a good general knowledge of international patent law in order properly to understand it. Indeed that knowledge was requisite in order to draft it properly in the first place but, as I shall show, seems to have been somewhat lacking. At the risk of expounding what is well known in our field, here are some matters that are relevant.

Patent ‘Application’ Is Ambiguous

When one speaks of a patent application one may have to be careful to distinguish between two different meanings which that expression can bear. It can mean:-

It is amazing how often we employ this expression – ‘application’ – without being conscious of its ambiguity. Here are some examples.

International Patent Applications

There is no such thing as an international patent nor can you apply for one. Patents are national, and have force in the territories of the countries to which they pertain, and nowhere else. Therefore whoever wishes to be granted a patent must, sooner or later, request it from the authorities who are competent to grant patents for that State.

What, then, is an ‘international patent application’? It is a preliminary procedure conducted on the international plane which, should you choose to use it, will give you extra time before you need commit yourself to the expense of national patent applications and will reduce that expense if you do. This is how it works. A description of your invention in a single language is sent to an International Searching Authority. They search your invention against the prior art, and prepare a search report. The search report can help you decide whether it would be worth your while to go ahead and seek national protection – what is called ‘entering the national phase’ – and if so, in how many countries, for each one of those will cost you money. (Think of the translation costs alone.) A still further advantage is that many national patent Offices will trust the international search report – although the Convention does not oblige them to do so – instead of searching the prior art themselves; and so you may be able to save quite a lot in search fees.

An international patent application, as such, cannot be refused (Footnote: ¹⁰) and confers no patent monopoly. It does not convert automatically into a national application: for example you may have to pay the prescribed national fees and furnish translations: see e.g. Patents Act 1977 section 89A(3) and Patents (Fees) Rules 2007 rule 3(2). Unless and until a national patent is granted you cannot take anyone to court for using your invention. If eventually you do succeed in obtaining national patents, however, the laws of most countries will backdate your rights to the date when the international patent application was published e.g. you could claim back-damages against infringers (see e.g. Patents Act 1977 sections 69 and 89B(3), also Article 29 of the Patent Cooperation Treaty).

When making a PCT application it is common to define the invention somewhat more broadly than you guess the national authorities will eventually allow, because it is good practice to start off wide and narrow down later in the light of the prior art that is cited against you – seeing that you cannot really know that prior art in advance. Even so, it is possible for the claim that is eventually granted to cover matter that does not fall within the claim of the PCT application.

Mixed Patent and Knowhow Licences

For present purposes ‘knowhow’ may be defined as all the information possessed by a licensor which a licensee would like to have too, in order to help him get started. Maybe the licensee could collate the information for himself by going to public domain sources; but he would lose time, and that is why knowhow is valuable in business. In practice the information will consist of a miscellaneous rag-bag of items, some of which may seem pretty trivial in isolation. For example, the fact that there are such things as fuel additives for improving engine efficiency: nobody would agree to pay royalties merely for using that bit of information on its own. But conversely, nobody would grant a knowhow licence under which the licensee did not have to pay royalties unless he used all of the knowhow, for then the licensee might abstain from using one bit of the knowhow and so avoid having to pay royalties on the rest.

Therefore it is often difficult to come up with an acceptable royalty clause. There is one method, however, that is not uncommonly used in mixed patent and knowhow licences (sometimes called technology transfer agreements). As the late Mr T.A. Blanco White wrote in The Conveyancer and Property Lawyer many years ago (Footnote: ¹¹)

Having made those regrettably lengthy observations I can now turn to the language of the documents that the parties signed.

The Licence Deed must be read together with the Main Agreement of even date. Here are some of the provisions of the Main Agreement that reflect back into the Licence Deed.

The Recitals of the Main Agreement say as follows:

Instead of the 50:50 joint venture as originally contemplated, the exploitation vehicle – Oxonica Subsidiary – is a wholly-owned subsidiary of Oxonica Parent. Mr Ronendra Hazarika becomes a director of Oxonica Subsidiary.

Amongst expressions that are defined in the Main Agreement and are relevant to what I have to decide are the following:

It is expected (but, so far as I can see, not mandatory) that Oxonica Parent shall raise equity finance of £4 million: this is described as the Investment Round. If this is completed within 6 months Oxonica Subsidiary must pay £250,000 to Neuftec (Clause 7.2).

There is a worldwide non-competition covenant by Neuftec (Clause 10). Amongst other things, Neuftec must not ‘solicit … a customer … of the [Oxonica Subsidiary] Business for the purpose of offering [to him] goods or services similar to or competing with those of the Business’. Note that this is wider than the scope of the exclusive patent rights. For instance, it would prevent Neuftec from offering the customer a rival fuel additive that contained no lanthanide oxides at all.

I shall now set out the provisions of the Licence Deed which are most relevant for present purposes.

Pausing there, the language of these recitals cannot trump the operative parts of the Deed. But it can help us to interpret them and to impart a flavour. In these recitals the entire focus is on the PCT. Granted patents are not mentioned, which is technically inapposite. But a recital is often a bird’s eye view and its very brevity may help us to understand what was the main focus. As we have seen, recital (B) of the Main Agreement is about the exclusive licensing of Neuftec’s technology and knowhow. The recitals of both documents have to be read together.

Turning now to the operative clauses of the Licence Deed, these begin with a series of definitions. The most important of those for present purposes are as follows.

Again pausing: note that the Knowhow does not need to be secret knowhow. But in any case a substantial amount of it would be likely to bear the quality of confidentiality qua the Oxonica companies, because of the so-called ‘springboard’ cases on the law of confidential information.

Pausing again there, I do not interpret this reference in Clause 2.4 to the termination of the Deed to include the termination of the licence when the last of the Licensed Patents has expired. If I am wrong about that Clause 2.4 would be drastic in its effect. It would signify that Oxonica might build up an important business in these additives over nearly 20 years – with plant, laboratories, numerous employees and a valuable goodwill in its trade mark Envirox – and then have to give it up even though it had done nothing wrong. It would probably have to give it up in practice because Licensed Knowhow is so broadly defined. It means just about everything that Neuftec knew as of 7 December 2001 about lanthanide oxide additives, whether known to others or not, and whether patentable or not. The more Oxonica exerted themselves over the last few years of the licence, the more they would be throwing their efforts away: which is why they would not be likely to exert themselves, and why I say that interpretation might well be in breach of the competition laws, and unenforceable. I am inclined to think, without actually deciding, that Clause 2.4 refers to termination for some repudiatory breach e.g. under the immediately preceding Clause 2.3 for failure to pay the £250,000.

I can now tabulate the main benefits for the Oxonica and Neuftec parties, respectively.

The more I read the document, the more I think: “A little learning is a dangerous thing”. It contains a number of malapropisms: they cause me to believe that the draftsman was not very familiar with patent practice and terminology. It conveys an impression that he was not always clear in his own mind about what he was doing and, when not clear, allowed his word processor to do his thinking for him. Bits of legal phraseology have been lifted from I know not what precedents and assembled in a strange way.

I shall give a few examples. The first of these shows unfamiliarity with patent terminology. Clause 2.2 says: “This Deed … shall continue in force until either the date on which (a) the last of the Licensed Patents has expired or (b) the last of the Licensed Applications has been revoked”. A patent application is never “revoked”: it may be refused. It is the Licensed Patents that might be revoked, but the language the draftsman has chosen does not cover that eventuality. It is as if he supposes they will last until they expire. For that matter he cannot really have meant that the Deed shall cease to be in force when one of those events does eventually occur. Existing obligations e.g. to pay royalties already accrued will continue to subsist. The terminology does not matter much in itself. But, just like Mrs Malaprop’s “allegory”, it puts us on notice that this draftsman was not very familiar with the fauna of the patents world.

Consider next the definition of “Licensed Application”. It is stated to mean the PCT application annexed to the Deed “and any continuation, continuation-in-part or divisional applications thereof as well as foreign counterparts and re-issues thereof”. This is a string of technical expressions used in American patent law, yet this is an agreement governed by English law, between two parties neither of which is American, and signed at a time when the only patent application that existed, so far as I know, was a PCT claiming priority from two former British applications, which PCT designated numerous States. The draftsman has employed language that you would expect to find in an agreement between two American companies where the starting point and focus was an American patent application. The only patent application mentioned in the Deed is the PCT, and strictly speaking there could never be any continuation or continuation-in-part “thereof”, nor any re-issue. (It is true that the PCT application founded an opportunity to apply for patent protection in the USA and then those things might arise; but that would be in one country out of many.) And in the context of a PCT application that designates numerous countries of the world in alphabetical order, what does “foreign counterparts” mean? The same can be asked of the definition of “Licensed Patent”.

Consider now the consequences. Continuations, continuations-in-part and divisionals are collectively known as continuing applications. US patent procedure can allow the filing of an unlimited chain of continuing applications. “The term ‘parent’ is often used to refer to the immediately preceding application upon which a continuing application claims priority; the term ‘original’ is used to refer to the first application in a chain of continuing applications.” See Chisum on Patents, vol. 4A, chapter 13. The following is a simplified explanation.

It follows that at the date of the Licence Deed it was possible that, by afterwards filing chains of continuing applications (USA) or divisional applications (Europe), Neuftec could have had pending patent applications for many years. Those pending applications could have had wide claims, as wide as they pleased – even though the original application was refused, or was allowed with but narrow claims. So far as I can see this process could have been kept going almost indefinitely. What, then, of the proposition that royalties are payable in respect of products falling within claims of “any continuation, continuation-in-part or divisional applications”?

Although I must stress that I do not found my judgment upon it, I think I can deduce quite a lot about the drafting history of this document. Its draftsman must have started off with the general idea that an exclusive licence was to be granted under the PCT. The Recitals tell us as much. That would have been the genesis of the expression “Licensed Application” (note the bizarre use the singular). From that concept “Licensed Patent” (also in the singular) would have followed. There must have come a time when the draftsman appreciated that there could be more than one application and more than one patent and in order to flesh out those concepts he probably looked for some convenient boilerplate language. I am quite certain he must have used a precedent. I do not know what precedent that was, but it is likely to have been derived, in its turn, from a series of earlier precedents. These things are handed down or copied through generations. (One is reminded of the old-time lecturer’s notes – transferred to the student’s notebook without going through the mind of either.) They are often influenced by historical considerations that are now obsolete.

One such consideration – and I am convinced that our draftsman was blithely unaware of it (Footnote: ¹²) – was that, years ago, the competition authorities in America and Europe treated intellectual property licences with considerable suspicion. They feared that they might be part of a disguised restraint upon trade. The consequences for the parties might be severe, and at any rate inconvenient. Therefore some draftsmen took to using a device known as a limited licence. According to this device any restraint was imposed by the intellectual property right itself – which existed anyhow – and not by the law of contract (Valentine Korah, ‘Technology Transfer Agreements and the EC Competition Rules’, Clarendon Press, 1996, pages 71-77). Sometimes even an obligation to pay royalties was treated as a restraint because it might be a burden that affected the licensee’s ability to compete (pages 186-190). It was thought that if you used the same phrase to describe the intellectual property right and the products upon which royalties were payable that would get rid of the problem or, at any rate, put the competition authorities off the scent. The expression “Licensed Products” is typical of that era and was used by our draftsman in the Licence Deed. Unfortunately he failed to appreciate that it could not do double duty in a straightforward way in the document he drafted, and I shall return to this later.

I conclude, thus far, that the drafting of this commercial contract warns us very strongly not to put too much weight on “detailed semantic and syntactical analysis of words … [if it] is going to lead to a conclusion that flouts business commonsense” (per Lord Diplock in Antaios Compania Naviera SA v. Salen Rederierna AB [1985] AC 191, 201, cited by Lord Hoffmann in Investors Compensation Scheme). Hoffmann LJ (as he then was) said in Co-operative Wholesale Society Ltd v National Westminster Bank plc [1995] 1 EGLR 97 at 99F as follows:

The key expression is “Any product, process or use falling within the scope of the claims in the Licensed Application or Licensed Patent”. The ambiguity starts with the little word ‘or’. In natural English it is used to connect two or more conditions, the satisfaction of any one of which will suffice. Sometimes the context causes us to understand that the alternatives are not mutually exclusive (as in “Working knowledge of German or French an advantage”). At other times it causes us to understand that they are (as in “Double or quits”). In formal logic the symbols (inclusive or) and (exclusive or) are used to specify the difference. In natural English the word “or” cannot do the work by itself and context is vital.

Oxonica interprets it to mean “the claims in the Licensed Application or Licensed Patent, as the case may be”, and denies that both can exist at the same time and place. It says that the procedure from application to grant is linear and sequential. For each territory at any given point in time, the inventor is either in possession of an application or a granted patent – one or the other but never both. At the date of the Licence Deed, the PCT had not been subjected to any of the rigours of examination for patentability and there was no guarantee that the claims as filed would survive through to grant. It makes little sense to describe a product or process as “Licensed” when no licence is required. Why should Oxonica have to pay royalties on a product which the rest of the world can make and sell for nothing? If Envirox 2 is not a Licensed Product, why must Oxonica cease and desist from exploiting it upon termination of the Deed?

Oxonica does not contend that ‘application’ means an application that ought to be granted. It accepts that it is liable to pay royalties on products falling within the claims of an application that are subsequently narrowed down and that the monies cannot be recouped. If asked why it should have been willing to pay those, I think that Oxonica would say that the application would be pending for a relatively short time; that during that time its business would be relatively new and the volume (and hence the royalties) would be comparatively small; and so it would be reasonable to pay this price in case the claims of the application survived the examination without modification.

Oxonica’s argument partly assumes what it sets out to prove. It assumes that ‘application’ is intended in Sense 1 (see paragraph 43 above): a live request that a patent be granted. But if the proper interpretation is that ‘application’ is to be taken in Sense 2 – the content of the document (in this instance, the contents of its claims) – then it is not true that an ‘application’ and a patent cannot exist simultaneously in the same country. Furthermore it is quite possible for an ‘application’ even in Sense 1 to co-exist with a granted patent. I have already shown how: what about a divisional application or, in the USA, a continuation or a continuation-in-part? On the evidence there exists a European divisional application right now, and Neuftec intends to file yet another one of those.

The answer to the question, why should Oxonica have to pay royalties on a product that the rest of the world is free to use, may be that Oxonica also has a licence – indeed an exclusive licence – to use Neuftec’s knowhow. And the answer to the question, why should Oxonica have to cease and desist from exploiting Envirox 2 upon termination of the Deed is, I think, that it does not have to, unless it was terminated for fault: see paragraph 62 above.

Thus although Oxonica’s arguments are plausible, they do not compel the conclusion that it desires.

Oxonica’s interpretation would also generate an obligation to account and keep records pursuant to Clause 4.3 that might be quite formidable, and sometimes impossible to perform. Because this is not so obvious, I ought to explain why. These products are likely to be high value, low bulk goods which are eminently transportable. Typical customers might include the operator of a fleet of long distance lorries or even an international oil company. By their nature, then, these products are eminently likely to cross international frontiers. Whether a product is a Licensed Product, or not, is said to depend on the patent position and this varies from country to country. It may sound easy to identify the relevant country; but patent law shows that it is not infrequently a controversial question.

Royalties are payable on “revenue received by the Licensee from the sale of the Licensed Products” (Clause 4.4). And Licensed Products means “any product, process or using falling within the scope of the claims in the Licensed Application or Licensed Patent”, as I have said. Take, first, sales of a product. In British patent law a product is not considered to be sold or disposed of in this country unless the title passes to the buyer while the goods are physically present in this jurisdiction. That is established by a line of well-known authorities including Badische Anilin und Soda Fabrik v. Johnson [1898] AC 200, H.L.; Badische Anilin und Soda Fabrik v. Hickson [1906] AC 419, H.L.; Morton-Norwich Products Inc v. Intercen Ltd [1978] RPC 501; and under the current Act, Kalman v. PCL Packaging (UK) Ltd [1982] FSR 406. In each of those cases it was held the British patent was not infringed by sale (Footnote: ¹³) because the property in the goods had already passed before they arrived in this country. In order to determine when and where the property passes it is first necessary to determine the proper law of the sales contract. For example if the proper law is English the property will pass where and when the vendor and purchaser have agreed that it shall; failing such agreement, it passes according to the rules prescribed by the Sale of Goods Act 1979. Thus the result may depend on such specialties as ‘the battle of the forms’, whether it was a classical c.i.f. contract or f.o.b., whether there was a Romalpa clause, and so on. Other legal systems may have different approaches. American patent law seems to treat the goods as sold in the state where the buyer has his business: North American Philips Corp v. American Vending Sales Inc 35 F.3d 1576, 1994. All this would be difficult to work out; and although it would not be impossible in itself, it could be a constant source of friction and uncertainty.

Now take the case where the thing sold is a process or use: namely, the method of improving fuel efficiency or reducing emissions that comprises dispersing the coated lanthanide oxide in the fuel. One envisages an international lorry driver buying the product and putting it in his tank when the vehicle is filled with fuel; this could be repeated at each refuelling stop. In what countries the method would be used could not normally be known to Oxonica and proper accounting would not be possible.

During the course of argument I asked Mr Acland (who appeared for Oxonica) to consider what would happen if his clients were to arrange for all their output to be sold in a country where there was no application and no patents, their customer to take delivery there. (The customer might be an international oil company, say, or their exclusive worldwide distributor.) Could his clients do that without paying any royalties to Neuftec? Mr Acland replied that they could (Footnote: ¹⁴): such products would not be Licensed Products.

It would also be a consequence of Oxonica’s position that royalties are payable according to the breadth of Neuftec’s current patent position in each country, and nothing else, even though the document is also a knowhow licence under which the knowhow was imparted for use in all countries, and even though Neuftec has given up its right to compete in all countries for the whole term of the licence. Thus in a relatively short space of time Oxonica would be free to sell a product like Envirox 2 in many important countries of the world without having to pay any royalties, even if Oxonica were making use of Neuftec’s knowhow.

Neuftec says the phrase means “the claims in the Licensed Application or Licensed Patent, or any of those”. Thus it is enough if a product falls within a claim of the PCT, even if it should not fall within any claim of the granted patents. It makes sense (says Neuftec) because the Deed is also a knowhow licence and the claims of the PCT may have been chosen as a convenient yardstick to denote royalty-bearing products. The phrase “Licensed Application or Licensed Patent” was put in to cover the possible case where a granted patent had even wider claims than the PCT.

Licensed Application is defined not only as the PCT itself but also “any continuation, continuation-in-part or divisional applications thereof as well as foreign counterparts and re-issues thereof”. During the course of the argument I asked Mr Hacon what would happen if the scope of the claims of one of those applications was different from (e.g. wider than) those of the PCT. What then would define the royalty-bearing products? Mr Hacon replied that a product would bear royalties if it fell within the scope of a claim of any application – in whatever country. That would govern the position worldwide.

I then asked Mr Hacon what was to stop his clients filing an application somewhere with claims of enormous scope, then charging Oxonica royalties on their worldwide sales. Likewise, what was to stop his clients from being granted an enormously wide patent (for there are still countries which do not examine patent applications for validity: they grant whatever claim is requested and leave it to the proprietor to take his chances in the courts). Mr Hacon felt the force of that, and he felt constrained to reply that it would not do unless the claim was properly founded on the teachings of the PCT.

I am not persuaded by that answer. It presupposes that the matter would be governed, not by a claim that was actually filed, but by a claim of justifiable width – what ought to have been filed. Indeed it might then be argued that that is precisely what Oxonica are complaining about here, namely that claim 1 of the PCT was not allowed because of novelty-destroying prior art (Footnote: ¹⁵).

Mr Hacon felt constrained to adopt the argument I have criticised above in order, as he said, to avoid “doing violence to the language of the Deed”. In sum, Mr Hacon’s argument seems sound in its outline, but becomes implausible if he is compelled to adopt the conventional meaning of some of the language in the Deed. I might have felt obliged to give effect to that if I were convinced that the author of this document knew precisely what he was about.

Mr Hacon also argued that because the document was drafted by Oxonica’s then solicitors it should be interpreted contra proferentem.

In Tam Wing Chuen v. Bank of Credit and Commerce Hong Kong Ltd [1996] 2 BCLC 69, 77, P.C. Lord Mustill said that

The maxim verba fortius accipiuntur contra proferentem was introduced into English law by writers such as Sir Edward Coke (Co. Litt. 36a, 183a), Lord Bacon (Maxims of the Law, Regula III), and Sir Henry Blackstone (2 Commentaries 23, 4). But they discussed it, not in relation to what we would now call commercial agreements, but to unilateral documents, such as grants under deed or pleadings. Thus Blackstone denied that it could apply to indentures, “for the words of an indenture, executed by both parties, are to be considered as the words of them both”. They must have derived the maxim from the civilian writers. There is a useful comparative law paper by Cserne (Footnote: ¹⁶), to which I am indebted. In the classical period of Roman law the principle seems to have been applied, if at all, to situations where in practice one side alone had the opportunity to formulate the wording (Footnote: ¹⁷). In English law it seems that the maxim first came to be applied to commercial agreements out of a desire to tame exclusion clauses – an early form of consumer protection.

The maxim operates most comfortably in standard form contracts, where one side puts forward the document on take-it-or-leave-it terms. If, for example, an insurance policy contains a clause excluding losses caused by ‘floods’, and damage is caused by water escaping from a burst domestic pipe, we would say: “Well, they should have put it more clearly, if they wanted to exclude ‘floods’ of that sort” (Footnote: ¹⁸). It is harder to apply the maxim to the case of contracts individually negotiated.

The contra proferentem maxim is an old one and there is no doubt that it still survives, although it is not to be used save as a last resort. In the present case the main drafting effort was undertaken by Oxonica’s then solicitors, but it is clear that Neuftec was not obliged to take it or leave it: indeed Neuftec admits that it too had something to do with some of the language of the document. The royalty clause (4.2 and 4.4) may be said to be for the benefit of Neuftec, and it also could be said that it was up to Neuftec to specify its rights with greater clarity. As against that, it is not so much the royalty clause that is obscure, but the definition of Licensed Products which occurs in Clause 1; and this is for the benefit of Oxonica as much as for Neuftec.

I therefore hold that the maxim is, in this instance, not adequate enough to supply the answer to the case, even as a last resort.

It will be helpful to repeat the definition of “Licensed Products”. It means “any product, process or use falling within the scope of claims in the Licensed Application or Licensed Patent” (two alternatives). I reject Oxonica’s contention that the second alternative naturally displaces the first when a patent is granted in a particular country. The document does not say so. To my mind, a telling point is that, if and when the second alternative did displace the first, Oxonica would not be licensed to use the Knowhow “to make, manufacture, use, sell and/or otherwise exploit the Licensed Products throughout the Territory [i.e. the world]” (Clause 2.2). It would merely be licensed to do so in those countries where the product fell within the scope of claims of a granted patent. Put a little differently, the more that Oxonica worked outside the claims, the more evident it would be that it lacked permission to use such information as was divulged to it in confidence. That is not sensible. Furthermore Oxonica’s construction would produce sundry inconveniences and anomalies – as previously discussed.

In that respect, I uphold Neuftec’s construction. In my judgment royalties are attracted by any product falling within at least one claim of the document appended to the Licence Deed: the PCT as filed.

However Neuftec’s main problem – at first sight, a serious one – is that the definition of Licensed Application covers not only the PCT but also “any continuation, continuation-in-part or divisional applications thereof”, which seems to prove too much. See paragraphs 84 to 87 above. In my judgment the problem is to be resolved as follows.

I have already explained that the definition of Licensed Products is ambiguous and that the ambiguity starts from the fact that we cannot tell – just by looking at the definition itself – whether the two alternatives are meant to be mutually exclusive. The ambiguity continues in the word “application”, which has two different meanings as already explained. That produces a number of possible permutations and the resulting meaning must depend on the context. In my judgment the key to the matter is that, first, there are two contexts in which “Licensed Products” is employed in this document. They are (1) Clause 2.1, which confers upon Oxonica the exclusive licence to exploit the Licensed Products; and (2) Clause 4, which imposes an obligation to pay royalties based on sales of Licensed Products. Those contexts are not the same. Therefore it cannot be excluded a priori that the wording operates differently in each case.

The purpose of the first context is to license Oxonica under Neuftec’s intellectual property rights. A licence makes lawful that which otherwise had been unlawful. Oxonica wanted Neuftec’s permission to use the Knowhow and to operate within the claims of whatever patents might be granted in various parts of the world. Because patent protection, when granted, dates back to the date when the antecedent application was published (subject to a minor possible exception (Footnote: ¹⁹) which is not material here) the applications had to be included too. Oxonica also wanted the licence to be exclusive i.e. it wanted to be in the position of being able to assert those rights against third parties. In short, the purpose of Clause 2.1 was to confer upon Oxonica the exclusive licence to exploit Neuftec’s intellectual property and that included the Knowhow and any technology falling within the claims of the PCT as filed and any other patent applications and patents claiming priority from the PCT. For that purpose the boilerplate language employed by the draftsman, although decidedly long-winded and certainly not very well chosen, is nevertheless intelligible and should give rise to no real difficulty.

The purpose of the second context is different. It is to define the class of royalty-bearing products. Of course, in some licences there is intended to be an exact correlation between the two, in the sense that the licensee must pay if he does something that is covered by the intellectual property right, but not otherwise. That is likely to be so e.g. where there is a pure patent licence negotiated between hostile parties, for instance in settlement of litigation actual or threatened. But that is by no means invariably the aim, and in this day and age there is usually no reason why it should have to be the aim. Indeed, in some mixed licences (knowhow and patents) it may not be practicable to achieve it or, in the eyes of the signatories, desirable.

In my judgment it was not the aim here, either. Take first the actual wording. Royalties are payable on things “falling within the scope of claims in the Licensed Application …” i.e. the PCT “and any continuation, continuation-in-part or divisional applications”, etc. On a literal reading no royalties would be payable unless the thing fell within the scope of claims of all of those aforementioned things. If a thing fell within the scope of one or the other, but not both, no royalties would be payable. One odd man out – say, a divisional application claiming a different aspect of the invention – could be enough to do it. Likewise, if all applications had been disposed of, no royalties would be payable unless the thing fell within the scope of claims “in any patent issuing from the Licensed Application as well as foreign counterparts [whatever that may mean, if anything] and re-issues thereof”. The paradoxical and absurd result would be that the more applications and patents there were, the more likely it would be that Neuftec would get no royalties.

In my judgment the proper way to interpret this Licence Deed is to say that “Licensed Products” means things falling within the scope of claims of the Licensed Application or Licensed Patent as the context may require. Likewise, the definitions of “Licensed Application” and “Licensed Patent” should be understood with the qualification as the context may require. Those are not glosses on the language. They must be implied, in order to make sense, even quite formally, of the English.

Having regard to the context, therefore, I hold that royalties are payable on things falling within the scope of claims of the PCT as filed, and nothing else. It is an application in Sense II above, and not sense I. That was the convenient yardstick that measured out the scope of Neuftec’s technology imparted to Oxonica. The Recitals to the Licence Deed – read together with the Recitals to the Main Agreement – make it tolerably clear. Claim 1 of the PCT corresponds essentially to “any product, development or invention containing lanthanide oxides … for use in connection with the combustion of fuels” (Recital (B) of the Main Agreement).

This construction has a number of advantages. It does not create the difficulty that, by signing this document, Oxonica exposed themselves to the risk of proceedings for breach of confidence for using the knowhow outside the scope of granted patent claims. It does not create the difficulty that it would be troublesome, and sometimes impossible, to account for royalties on country-by-country basis. It does not create the difficulty that, on a contrary view, Oxonica would be exposed to a liability to pay royalties in any country where Neuftec had a divisional application (Europe) or a continuing application (America) with claims of such width as Neuftec pleased to choose. You can always make an application. It avoids the unattractive and unbusinesslike result that Oxonica could evade payment of royalties altogether by selling and delivering all of its products to an international distributor or oil company in a country where no patent or application existed. It is most improbable that that could have been intended.

There is yet a further advantage to which I have not yet alluded. So far, the only remuneration to which I have referred is the 5% royalty mentioned in Clause 4.2 and 4.4. But Neuftec is entitled to other remuneration, namely a 25% share of the profits (Clause 4.6) and a revenue milestone lump sum payment of £1 million (Clause 4.5). But those are, respectively, the profits of the Business “relating solely to the exploitation of the IP Rights” (see the Main Agreement, Clauses 1.1.3, 1.1.9, 6.1.1 and 6.1.3) and a lump sum payment conditional on a turnover of £10 million attributable to the “Business” as equally defined. Now, on Oxonica’s interpretation of this Licence Deed, what would happen if Oxonica chose to work entirely outside the claims of the Licensed Patents e.g. by making and selling nothing but Envirox 2? In order to account for the 25% share of the profits it would be necessary to determine what part of Oxonica’s business was solely attributable to the exploitation of the Licensed Knowhow – there would be no other relevant IP Right. In the same way, in order to determine whether and when the £1 million was due it would be necessary to determine at what point Oxonica’s business, being that part of its business solely attributable to the exploitation of the Licensed Knowhow, had achieved a turnover of £10 million. This would not be easy to do. There would be no convenient yardstick. A fact-intensive investigation would be necessary at every point. It would be one thing to do it in the course of litigation. It would be quite another to do it as a routine book-keeping exercise. But if my interpretation of these documents is correct, the difficulty disappears.

That one and the same defined expression may bear different meanings in different parts of the Deed will, I own, seem surprising to some. Had the draftsman of this document been a latter-day Sir Mackenzie Chalmers the point might have been compelling. But we are not dealing with drafting of a high standard and according to the modern approach to the interpretation of commercial documents due allowance should be made. It might also have been compelling if this case had had to be decided under the old regime – before Mannai Investments and Investors Compensation Scheme cases. Above all, words and phrases do not have an inherent meaning. Emphatically they do not when they are ambiguous even according to conventional usage. After long reflection I can see no reason why it is necessary to attribute a meaning to an expression derived from one context and then to plug it in blindly into a different context.

Royalties are payable in respect of any product, process or use falling within the scope of any claim of the PCT application as appended to the Licence Deed, and nothing else. Envirox 2 is a Licensed Product as defined, and attracts royalties accordingly. The claim fails and the counterclaim succeeds.

If my interpretation of the Licence Deed is wrong, it still would not be appropriate to grant the declaratory relief sought by Oxonica. The answer might then depend on the existence and status of divisional applications and the like.

I shall hear the parties about the form of relief or they can, if they wish, send their submissions by email to my clerk by some convenient date to be arranged. I am presently minded to say that (1) costs should follow the event and (2) permission to appeal, if sought, is almost inevitable.