The patent owner's rights


Term of protection

The basic term of protection under a UK patent, and under a European Patent, is 20 years. Protection runs from the date on which the application is filed, or the priority date if priority is claimed.1 Annual renewal fees are payable from the fourth year after the application is filed (so renewal fees could become payable before the patent is even granted). A six-month grace period is allowed, on pain of payment of an additional fee, for late renewals.

Under the 1949 Act it was only 16 years, and patents in existence when the 1977 Act came into operation received an extra period of protection. To make things a little fairer, this additional windfall period was made subject to licences of right – that is, anyone wishing to work the invention in that time would be able to get a licence on request provided that he paid the licence fee demanded by the patentee.

Supplementary protection certificates

Some products, particularly pharmaceuticals, take a long time to get to market. Extensive tests and trials are required, so that by the time the patentee begins selling products covered by the patent there is often little protection left. Sometimes as much as 12 years of the patent term may already have gone. In the Copyright, Designs and Patents Act 1988 pharmaceutical patentees were given the possibility of extending the term of protection of their pre-1977 patents by being relieved of the licensing of right requirements referred to in the preceding paragraph. Later this concession was also extended to pesticides.

The EC has also recognised the need for additional protection for proprietors of pharmaceutical inventions, and Council Regulation 1768/92 as amended2 provides for the grant of supplementary protection certificates (SPCs)3. SPCs will have the effect of giving an extra period of protection for pharmaceutical products, though strictly speaking they do not extend patents.

The actual period of supplementary protection is the period between the application date and the date of fist authorisation to place the product on the market, less five years, subject to a maximum of five years.

Plant protection products are brought within the SPC scheme by Council Regulation 1610/96, which is implemented in the UK by the Patents (Supplementary Protection Certificates for Plant Protection Products) Regulations 19964.


The owner of a patent may surrender it before it has expired by giving notice to the Comptroller under section 29.


A patent, once granted, gives its owner the exclusive right to work the invention it protects. The law on infringements is the same for patents granted under the 1977 Act and for European Patents (UK). Although the information about how to work the invention is made public in the specification of the patent, and is available for others to draw upon, no-one may do precisely what is set out in the patent without the permission of the owner.

There are two questions to deal with. First, has the alleged infringer done one the acts set out in section 60? And second, is the alleged infringing act within the scope of the monopoly claimed in the patent? This question, involving the interpretation of the claims set out in the patent, is not an easy one to answer. The two parts of the infringement equation are sometimes referred to respectively as ‘infringement in law’ and ‘infringement in fact’, though these expressions do not appear in the Act.

The defendant may also be able to take advantage of defences, either provided in the Act or in the common law, and may take the opportunity to challenge the patent’s validity in a counterclaim seeking revocation. The examination process carried out before a patent is granted is bound to be imperfect, and the examiner’s resources, including time, are obviously limited. In litigation, the parties are already committed to enormous expenditure and repeating, in greater detail, the examination process is probably a minor expense. Given that patents frequently build on existing technology and may claim only a marginal development of what is already in the state of the art, it is quite easy to find prior art on which to base a challenge to the novelty or inventiveness of the patented invention. Even the Dyson vacuum cleaner patent faced challenges (in Dyson v Hoover), with the defendant arguing that it was not novel or inventive over an old US patent for a device to extract dust from mines. The judge rejected the argument, but it was not without merit. A successful revocation counterclaim means that the defendant has not infringed at any time, because there was nothing to infringe.

The current attitude of the courts to patents can be judged by monitoring the success of infringement and revocation claims. A pro-patent court might be more inclined hold patents valid and infringed, while a court which is less well-disposed to the whole idea of patents might hold more of them invalid.

General points

Before the 1977 Act there was nothing in the statutes about what constituted an infringement. It was a matter of looking at the claims and working out whether the defendant was within them, rather like working out whether a trespass has taken place by looking at the title to a property. The courts always required that there be some sort of commercial activity,5 but this was not written into the Act.

The law applies equally to deliberate infringements and to innocent ones, although the remedies available against an innocent infringer will not be the same as those which can be obtained against an infringer who knew full well that the patent existed (section 62(1)).

No infringement action can be started before the patent is granted. Infringements taking place after publication of the application may be the subject of a claim for damages (section 69) but the action cannot be started before the patent is granted.


Infringement is defined in section 60. The first sub-section deals with direct infringements and the second with indirect ones. The section is based on Article 25 of the Community Patent Convention (now Regulation 7 of the EU patent regulation). It says that an infringement is an act done without the permission of the patent owner, in the UK6, while the patent is in force, and it specifies these acts:

  • making, disposing of, offering to dispose of7, using or importing a patented product, or keeping it, whether for disposal or otherwise8;

  • using a patented process or offering it for use in the UK in the knowledge (or where it would be obvious to a reasonable man) that such use would be without the consent of the proprietor and would be an infringement of the patent9;

  • disposing or offering to dispose of, using or importing or keeping, whether for disposal or otherwise, any product obtained directly from a patented process10;

  • supplying or offering to supply in the UK a person (other than a licensee or other person entitled to work the invention) with any of the means, relating to an essential element of the invention, for putting the invention into effect. This form of indirect infringement requires some element of guilty knowledge on the part of the infringer. The infringer must know, or it must be obvious to a reasonable man in the circumstances, that those means are suitable for putting and are intended to put the invention into effect in the UK. Section 60(3) provides a defence where the supply is of a staple commercial product readily available from a number of different sources, not made by the defendant to the order of the primary infringer: see below, paragraph .

Professor Cornish has made the point that these acts can neatly be categorised as taking place before, during and after manufacture (though not in the order in which they appear in the Act).

The items in this list demand some elaboration.

Disposing, offering, using and keeping products

Wide interpretations are in order. Disposing includes buying an infringing product, intending to export it.11 Offering includes displaying at an exhibition12 and advertising in a catalogue13. Using means more than just having equipment which has not even been assembled14 but that might amount to keeping it which could be an infringement. However, Article 7 of the Regulation indicates that this means keeping it preparatory to selling it, so a warehouse keeper or carrier will not be liable.15 It requires an intention to trade in the item in order to make a profit.16

Obtained directly from a patented process

A narrow interpretation is given to this phrase, and the court will ask whether the end product retains the essential characteristics of the patented process.17


In SABAF v MFI and Meneghetti18 the House of Lords considered, obiter, when a person could be said to import products for the purposes of section 60(1). At first instance, Laddie J had held that (but for the important fact that the patent was invalid) Meneghetti would have infringed by importation. On appeal, that decision was reversed because Meneghetti had taken no steps within the jurisdiction, so could not have infringed for the simple reason that they had imported nothing. Their Lordships decided that the word ‘import’ has its ordinary meaning, and the importer was the person who brings the goods into the UK. This turns on who has title to and legal possession of the goods, and the person who makes arrangements to ship the goods is not the importer unless they also happen to own them. On this basis, the Court of Appeal’s finding on this point had been correct.

The requirement for knowledge

Where the infringement consists of offering a patented process for use in the UK, guilty knowledge is required and the test is an objective one. In the copyright case, LA Gear v Hi Tec Sports19, the standard was that of a reasonable trader in that area of commerce, and applied to patents this implies a need to keep abreast of what others in the field are doing.20

When is an act committed in the UK?

The infringing act must be committed in the UK. In the Internet age, this assumes a new importance. In Menashe Business Mercantile Ltd v William Hill Organisation Ltd,21 the Court of Appeal held that simply because a host computer is outside the UK does not mean that an invention implemented using that computer is not put into effect in the UK, the essential test for whether the court will have jurisdiction to hear an infringement claim. It is not however sufficient merely that the invention has an effect in the UK: the means for putting it into effect must be in the UK. Old ideas about location have little application in a world where inventions can be implemented using computers in other jurisdictions, and in the case before the court the user used the invention in the UK (therefore also using the remote host computer in the UK, notwithstanding that it was elsewhere). The defendant’s supply of software to users within the UK resulted in the invention being put into effect in the UK so the defendant could not raise a defence under section 60(2).

Other types of infringing conduct: joint tortfeasors

It is possible for other conduct to amount to an infringement, even though it is not mentioned in section 60(1) or (2). In particular, this might be relevant where there are joint tortfeasors – the defendant might have induced the infringing act, or was party to a common design. Given that the patentee can only sue an infringer who has done something in the UK, this could be useful if the products were actually made overseas. It requires more than just supplying the goods to a purchaser in the UK, though: as the court said in Sabaf, ‘[e]ach person must make the infringing acts his own.’

Groundless threats

Section 70 of the 1977 Act makes groundless threats of infringement proceedings actionable by any person aggrieved by the threats – so, not only the person against whom the threat is made. There is no remedy, however, where the alleged infringement giving rise to the threat of proceedings consists of making articles according to the patent, or importing such articles. The protection is effectively given to people lower down the supply chain, for example retailers and distributors, who cannot be expected to satisfy themselves about the status of the goods they purchase, and the remedy is given to protect the manufacturers and importers who might otherwise find their distribution channels closed because of a patentee’s threats. The possibility of being sued for infringement is often enough to secure compliance, but public policy demands that dealing with an infringement must involve exposing the patent to the possibility of revocation (as a counterclaim for invalidity of the patent is almost an essential complement to a defence in an infringement action, even in a case such as Dyson v Hoover where the patent looked pretty safe).

The groundless threats provisions are watered down by the 2004 Act to reduce the likelihood that patent owners will be discouraged from trying to settle disputes. Because of the possibility that a patent will be found invalid in legal proceedings, patentees may refrain from opening negotiations with alleged infringers for fear of the initial approach being found to be a groundless threat. Instead, they will sue first.

The new law provides that a patentee will not be liable for making threats if he has correctly assessed that the infringing acts are taking place and he is able to demonstrate that at the time he made the threat he had no reason to suspect that the patent was invalid.

A further problem is that a patentee may negotiate with an infringer where the alleged infringing acts are manufacturing or importing, but not otherwise, so must take care which infringing acts they mention. A letter before action that refers to manufacturing or importing only will not be actionable but if it is accompanied by a form of undertaking that mentions retailing as well the patentee may have issued an actionable threat.

The 2004 Act deals with this problem by allowing the patent holder to discuss any infringing act with the alleged infringer.

Finally, there is a problem if the patentee cannot identify the primary infringer (the manufacturer or importer). He may feel disinclined to approach a retailer or stockist for fear of being sued, however carefully the approach is framed. the retailer may be the only person who can identify the importer, so being able to speak freely to him is clearly important for the patent holder.

In Kooltrade v XTS Limited22 the High Court took the view that without prejudice correspondence may be admitted to prove an unjustified threat of patent infringement proceedings. There are also relevant cases on groundless threats in trade mark law and unregistered design right.

Interpretation of patent specifications

The construction of patent claims is a matter that has caused huge controversy. The basic principle is simple enough: to amount to an infringement, the act complained of must be within the scope of the specification and the claims (which in turn must be supported by the description) in the patent application, as amended before grant if this has been necessary to satisfy the examiner. An enquiry into the scope of an invention involves a two-step process:

  • To determine the technical meaning of the words used, with expert evidence being given about the meaning of technical terms if needed; and

  • To determine the effect of the words used. This is for the judge, and it is this stage at which problems often arise.23

Traditionally, unless it would produce an absurd result and deprive the patentee of all protection24, a literal interpretation was used: what is not claimed is disclaimed25 This treats the claims as ‘fenceposts’ marking the absolute extent of the monopoly (which makes the analogy with trespassing work well).

The intention has always been to strike a fair balance between the interests of the patentee (whose interest lies in the widest possible interpretation, regardless of the words used in the specification) and of the user (whose requirement for legal certainty would be best served by limiting protection to precisely what has been claimed by the words of the patent). That the latter approach would lead to unfairness has long been accepted, although in some circumstances it may still be appropriate.

The problem lies with variants of the claimed invention. If the infringer has done something that is clearly within one of the claims, there should be no argument about infringement, but problems arise when the infringer’s product or process is slightly different from the patented one. The history of interpretation of patent claims is one of the courts’ attempts to ensure that patentees enjoyed the protection they deserved, and unmeritorious attempts to avoid infringing by relying on technicalities failed. The past history is now largely academic, for reasons that will become apparent, though it is still important as the context in which the judges have reached the position that pertains today.

In his excellent book, A Practical Guide to Patent Law26 Brian Reid identifies no fewer than nine approaches to interpretation. More have come along since. What follows here simplifies matters somewhat, and does not deal exhaustively with Mr Reid’s list.

Drawbacks of literal interpretation

Textual or literal infringements are rare, and perhaps have become rarer as the patent system has matured. The courts have to deal rather with variants on the claimed invention, and they recognised a long time ago that a patent specification is a document written for scientists or engineers skilled in the art to which it relates. In 1968, in Rodi & WeinbergerAG v Henry Showell Ltd27 Lord Reid disapproved a literal approach to interpreting patent specifications:

... claims are not addressed to conveyancers: they are addressed to practical men skilled in the prior art, and I do not think that they ought to be construed with that meticulousness which was once thought appropriate for conveyancing documents.

Which is all very well, but it ignores the fact that many patents are in fact drawn up at considerable expense by skilled patent agents, to be read not only by technical people (and patent specifications are a rich source of technical knowledge that is mined to considerable effect by researchers28) but also by other patent agents, solicitors, barristers and judges: the notion of the ordinary inventor writing a description of his invention for his peers was never representative of more than a minority of cases.

The ‘Pith and Marrow’ Approach

Where the issue of infringement is not clear cut, the question may be to consider whether the differences between the defendant’s variant and the patented invention are material. Do they differ in essential or inessential respects? This involves the so-called ‘pith and marrow’ approach, or doctrine of mechanical equivalents, where those integers29 or components essential to the invention are separated from those which are not. Even though there might be variations in the inessential elements, if the infringer has taken all the essential ones there can be no doubt that the infringement has taken place. The fact that inessential ones, those with mechanical equivalents, were not taken would be immaterial.30

However, it can be difficult to identify the substance of the invention. It is easy so to frame the claims that some slight modification suffices to remove the variant from the scope of the patent: in van der Lely v Bamfords 31 a patent for a hayrake with demountable rear wheels was not infringed by one demountable front wheels. In Rodi & Weinberger v Showell32 an expandable watch bracelet with U-shaped links was not infringed by one with C-shaped links. In both cases (which were decided under the old law, of course, and have no value now as precedents) the accused product worked in the same way as the patented one. Both were majority decisions of the House of Lords, and in the earlier case Viscount Radcliffe said:

When, therefore, one speaks of theft or piracy of another’s invention ... and this ‘pith and marrow’ principle is invoked to support the accusation, I think that one must be very careful to see that the inventor has not by the actual form of his claim left open to the world the appropriation of just that property that he says has been filched from him ...

Note that in any case this does not deal with the basic problem of interpretation, to determine what is within and what is not within the claims.

The Purposive Approach

In Catnic Components v Hill & Smith Ltd33 the House of Lords adopted a purposive approach to interpretation, which replaced the ‘pith and marrow’ approach34. The case concerned steel lintels, building components that support the masonry over a doorway or window. The plaintiff had a patent for lintels with the ‘second rigid support member extending vertically from or near the rear edge of the first horizontal plane or part adjacent its rear edge’. The defendant’s lintel had the rear support member inclined at an angle of some 6 to 8 degrees to the vertical, and the case hinged on whether when the patent said vertical it meant it exactly.

Looking at what the applicant had in mind rather than the precise words used, in the specification, their Lordships (Lord Diplock giving the leading judgment) held that the verticality of the support member plaintiff’s lintel was inessential as the invention worked just as well with a slight incline (as in the defendants’ product). The defendants had argued that the verticality was essential and therefore there was no infringement of the pith and marrow of the patent. A literal interpretation would have deprived the patentee of all protection: ‘vertically’ was treated as a descriptive word.

The approach to construing patent specifications adopted in that case is described as ‘purposive’. Lord Diplock said that ‘the kind of meticulous verbal analysis in which lawyers are too often tempted by their training to indulge’ was inappropriate in construing a patent specification: the real issue was whether practical persons, skilled in the art, would understand that the patentee intended strict compliance with a particular word or phrase to be an essential requirement of the invention. However, paradoxically, he stressed that there was no separate question of ‘non-textual infringement’: the only thing that could be infringed was the patent as it stood, the only uncertainty lying in what the text meant. His Lordship said that the important thing was the understanding of ‘persons with practical knowledge and experience of the kind of work in which the invention was intended to be used.’ The key question was whether such a person would read the language of the patent as demanding strict compliance, excluding any variant even though it might have no material effect on the way the invention worked. To a builder, the defendant’s lintel was as good as the patentee’s: he or she would treat the rear support member as vertical for his or her purposes: and it was to the builder (or the builder’s merchant) that the specification was addressed.35 Which is all very well, but it is hard to interpret something purposively unless the purpose is clear – meaning, in this context, that the author of the patent documents must describe the invention clearly to enable the reader to understand the purpose of the claims.

The Improver case

Catnic was a 1949 Act case, and in Improver Corp v Remington Consumer Products Ltd36 Hoffman J (as he then was) had the opportunity to apply the same approach to a 1977 Act patent. In interlocutory proceedings the Court of Appeal said that the Catnic test remained appropriate under the 1977 Act, and in the substantive proceedings Hoffmann J (constrained by decisions of higher courts) reformulated Lord Diplock’s test, putting it into three slightly more complex questions which became known as the Improver questions:

  • Does the variant have, in fact, a material effect on the way the invention works? If yes, it falls outside the claim.

  • If no, would it have been obvious to the informed reader at the date of publication of the patent specification that the variant had no material effect? If no, the variant falls outside the claim: the patentee is assumed to have limited the claim deliberately.

  • If yes, is it apparent to any reader skilled in the art that strict compliance with the primary meaning of the claim was intended to be an essential requirement? If yes, the variant is outside the claim, but if no it is an infringement.

Hoffmann J in that decision observed that a patentee may have intended a word or phrase to have not a literal but rather a figurative meaning, the figure being a form of synecdoche - (a form of the metaphor in which the part mentioned signifies the whole); or metonymy (a form of metaphor denoting the relation between two objects. Metonymy is to synecdoche what a metaphor is to a simile). That aside probably hinders students of patent law more than it helps them: the important thing about the Improver case is that it set the interpretation of claims on a course from which it could not be deflected – not much, anyway – until the same judge, now with much greater authority, was able to have another go at solving the problem. In the meantime the Improver questions were applied without question in many cases,37 and notwithstanding that they were never intended to be more than a tool to assist in applying the rules of claim interpretation they came arguably to be viewed as a set of rules in their own right.

The Protocol on the Interpretation of Article 69

Patent interpretation is not to be carried on in a vacuum under the 1977 Act and the European Patent Convention. The Convention has associated with it a Protocol on the interpretation of Article 69, the equivalent of section 60 of the 1977 Act. The Protocol dates from 5 October 1973, contemporaneous with the Convention itself, and was revised in 2000 with the addition of a paragraph about equivalents to which we will return later. It requires that the patent be neither interpreted literally nor used merely as a guideline. The Article is to be ‘interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties.’

To confuse the issue, the Improver case was paralleled by other cases in other European countries – the patent in suit being a European one. In Germany and The Netherlands the plaintiffs won, where they had lost in the English courts. The situation was further complicated by the fact that the German and Dutch courts had followed precedents which were rather out of line with the Protocol, but the point that practice in this country was out of step with practice on the Continent had been made.

PLG Research Ltd v Ardon International Ltd38 is the one case in this period which throws doubt on the Questions. The Court of Appeal, taking an approach that it felt was in line with the Protocol, asked whether the variant was one that a skilled person could deduce from the claims as interpreted by the description and drawings. Lord Diplock’s formulation, as revised in Improver, should, it thought, no longer be followed. Millett LJ preferred to follow the approach then being taken by the German courts, which he referred to as a modified version of the Improver approach: the first two questions were the same, while question 3 asked what variants the skilled reader would contemplate falling within the claim.

However, in Assidoman Multipack Ltd v The Mead Corporation39 Aldous J, declaring that the Court of Appeal’s remarks in PLG had been obiter, preferred the purposive approach to one limited to the words of the patent. He considered that Lord Diplock’s test still held good: and in any event it probably amounted to the same thing as the Protocol test. He soon had the chance to repeat this statement with the authority of the Court of Appeal, following his appointment (as Aldous LJ) to that court: see Kastner v Rizla40.

The Protocol Questions

The Improver Questions were therefore widely considered to provide a useful tool to help with applying the Protocol. In Wheatley v Drillsafe Ltd41 the finding of Ferris J at first instance, that the claimants’ European patent for a method of making threaded holes in tanks and for equipment to do the job was not infringed, was reversed by a divided Court of Appeal, with Aldous LJ in the minority. (The trial judge had, however, found that the patent was invalid, making his comments about infringement obiter: the Court of Appeal upheld the appeal on validity too, but unanimously). Aldous LJ, in the main judgement, used the Improver questions, now referred to as ‘the Protocol Questions’, to help interpret in accordance with the Protocol the appellant’s claim for their ‘centreless hole cutter’, which the patent itself defined as ‘a hole cutter not having a drill for forming a pilot hole’. The accused hole-cutter did not use a drill to form a pilot hole (the technical problem lay in avoiding piercing the tank before the hole-cutter had done its business) but did have a retractable probe which located the cutter and prevented wandering. Sedley and Mance LJJ took perhaps a more literal approach, and considered the probe to be equivalent to a drill and the accused device certainly not to be ‘centreless’. But it is not the outcome of the case that is important for posterity, it is the endorsement of the Protocol Questions.

Not that every case turned on the answers to the Questions. There could still be literal infringement, although the patent would have to be quite a simple one42. There could also be circumstances, especially where the patent was for a pharmaceutical, in which the Questions were not approriate and only the third question need be answered43.

Kirin Amgen and the demise of the Questions

Whatever the Questions were called, they had almost come to have the status of legal rules. Mr Justice Hoffmann had created a monster. It fell to Lord Hoffmann to kill it off. Kirin Amgen v Hoechst Marion Roussel and others44 is in many ways a classic case: the patent dealt with the introduction of an exogenous DNA coding sequence into a mammalian cell to express the hormone EPO. The defendants used an endogenous process to produce recombinant DNA, expressing it in a human cell that already contained the EPO gene. So the question was whether the defendant’s variant was within the scope of the claims in the patent which had been written before the endogenous gene activation method was known.

Endogenous and exogenous being even at a purely semantic level obviously very different matters, the answer might appear clear, but one must never underestimate the ability of words to take on a new and extraordinary meaning in patent-land.45

Lord Hoffman’s judgment – with which the other judges concurred – takes the view that what is in the claim is what matters. A claim could cover uses of technology unknown at the time of drafting, but the court had to ask whether the person skilled in the art would understand the description of the invention in a sufficiently general way to include the new technology, There is no doctrine of equivalents in Europe (as there is in the US): the claim means what it says: judges cannot read into it matter that is not there: what the patentee has not claimed is disclaimed.

The court cannot try to read the patentee’s mind or look at the prosecution file, and this Lord Hoffman observed was the approach taken in Germany and The Netherlands as well as the UK. Only one question had to be answered: what would the person skilled in the art have understood the language used by the patentee to mean? Everything else, including the Protocol, is mere guidance. The Protocol Questions might even turn out to be unhelpful in cases involving new technology.

Post Kirin Amgen

So patents are no longer to be construed by taking the old English approach of narrow forensic construction, or the old continental one of using the claims as a point of departure in determining the inventive achievement disclosed by the specification as a whole. Rather, they are to be determined by reference to the words used, giving them the widest meaning they are reasonably capable of having but no more. A purposive construction of the words will still be appropriate, but there is nothing unusual in that: a purposive approach is adopted in construing all documents, including statutes and contracts. In Virgin Atlantic v Premium Aircraft Interiors UK Ltd46 Jacob LJ summarised all the issues on claim interpretation, making that judgment essential reading.

Kirin Amgen makes clear that the question for the court is what the claims mean to the skilled addressee. Normally, the answer will be that they mean exactly what they say, but sometimes the words will define an approximation. This is particularly the case where numerical or geometrical limits are used, such as the word ‘vertically’ in Catnic, the meaning of which to a skilled builder might well differ from the meaning it would convey to a mathematician.

Most of the cases where the strict language of the claims has not been followed have involved mathematical or quasi-mathematical limits, where it may be correct to interpolate the word ‘about’. In the Kirin Amgen case, the skilled person would not have understood the claim to cover the activation of an endogenous gene: it was limited to the expression of an exogenous DNA sequence coded for EPO, so the infringement claim failed.

The case stresses the importance of good patent drafting. Fewer wide patents are better than many narrow ones (except, perhaps, for the financial well-being of those who draw them up), and wide claims have advantages tempered by the possibility that if they are too wide they may be revoked. The judgment encourages drafters to produce two sets of claims, a narrow series, confined strictly to the invention, and a wider series which push at the outside of the envelope. Certainly a patentee who has taken insufficient care about drafting the claims can no longer expect the courts to come to his assistance.

The Protocol amended

The doctrine of equivalents is a judicial doctrine which is used to assess whether an alleged infringement is indeed an infringement. It seeks to exclude any trivial variations that have no bearing on the working of an invention. The United States has the best-known and most complicated such doctrine, but other countries such as China and Japan also have them. The UK does not – or does it?

The problem is that once the scope of the monopoly has been allowed to exceed the bounds of the claims, as it is acknowledged has happened in the USA, it is difficult to know where to stop. In the leading U.S. case of Graver Tank & Manufacturing Co Inc v Linde Air Products Company47, Jackson J said that the American courts had recognised:

that to permit imitation of a patented invention which does not copy every literal detail would be to convert the protection of the patent grant into a hollow and useless thing. Such a limitation would leave room for - indeed encourage - the unscrupulous copyist to make unimportant and insubstantial changes and substitutions in the patent which, though adding nothing, would be enough to take the copied matter outside the claim, and hence outside the reach of law.

In Royal Typewriter Co v Remington Rand Inc48 Learned Hand J said that the purpose of the doctrine of equivalents was ‘to temper unsparing logic and prevent an infringer from stealing the benefit of the invention’. The doctrine deliberately extends protection to matters outside the claims which perform substantially the same function in substantially the same way to obtain the same result.

In Kirin-Amgen, Lord Hoffmann (at para 36ff) reviewed the development of the doctrine and made clear that there was no such thing in the UK. The English courts had not gone outside the claims to catch equivalents: rather, they had approached the problem identified by Learned Hand J by abandoning literality. The EPO 2000 introduced a doctrine of equivalents in Europe, the Protocol on the Interpretation of Article 69 being amended by the addition of a second article:

For the purpose of determining the extent of protection conferred by a European patent, due account shall be taken of any element which is equivalent to an element specified in the claims.

That does not amount to extending protection beyond the claims – something which Article 69 itself clearly prohibits. But it does contradict the assertion that there is no doctrine of equivalents in UK or European patent law: clearly, expressly, unambiguously, there is, although it has yet to make itself felt in the case-law.

Non Infringement

Specific defences are provided for in section 60(5), which are derived from the Community Patent Convention (now Article 7 of the Regulation) or from the Biotech directive. They are designed to strike a fair balance between the interests of the patentee and the legitimate business interests of third parties, a theme which also runs through the law on the interpretation of patent claims. In outline, section 60(5) covers (adopting the numbering of the sub-sub-sections):

  1. Private, non-commercial use;

  2. Experimental use;

  3. Dispensing medicines;

  4. Temporary presence in UK;

  5. Prior use; and

  6. Use between lapse and restoration.

Private, non-commercial use

Even under the pre-1977 law, it was always a requirement for an infringement to take place that the act complained of be performed for commercial purposes. Nothing done privately and for purposes which are not commercial will constitute an infringement49.

Use for experimental purposes

Section 60(5)(b) excludes acts done for experimental purposes relating to the subject matter of the invention. This covers experiments aimed at modifying or improving the invention but not trials to test the viability of commercial production or field trials preparatory to obtaining a licence for a product or in anticipation of delivery of a patent50.


Private medical dispensing is excluded from being an infringement by section 60(5)(c) provided it is extemporaneous preparation in a pharmacy of a medicine for an individual in accordance with a prescription from a medical or dental practitioner51.

Temporary presence in UK

Certain uses of patented technology on aircraft, hovercraft and ships temporarily (or accidentally) in the UK (or UK territorial waters) are excluded by section 60(5)(d)-(f). The Court of Appeal says that ‘temporary’ means for a limited time – frequency, persistency and regularity are irrelevant.52

Farmer’s privilege

Section 60(5)(g) and (h)

Use before patent application was filed

Certain uses begun before the patent application was filed may be continued in prescribed circumstances53.

Use between lapse and restoration

Nothing done after a patent has been allowed to lapse and before it is restored will constitute an infringement: in effect, there were no exclusive rights in being at the material time. However, section 28A does provide that if the patent could still have been renewed at the time, or if the infringing act was a continuation of an earlier infringing act, this exception will not apply.

Staple commercial product (section 60(3))

Section 60(3) provides a defence where the supply is of a staple commercial product readily available from a number of different sources, not made by the defendant to the order of the primary infringer

Common law defences


Repair of patented products

The right to repair a patented product is a matter of implied licence: in fact, it is not so much a defence as a negation of the ‘making’ infringement. The leading case is Solar Thompson v Barton54, which concerned making a new pulley wheel for an apparatus that required replacement wheels from time to time. The defence was first recognised in Sirdar v Wallington (1907) 24 RPC 539.

In United Wire Ltd v Screen Repair Services55 the House of Lords held that a patent was infringed where repairs to a product go as far as the making of a replacement product. The claimant is a manufacturer of sifting and filtering screens for use in the oil exploration industry. The former consisted of a frame retaining a mesh which is of such a size that fluid passing through it would leave behind solids. The defendants repaired these by stripping down the frame and adding two new layers of mesh to it. The claimants argued that this infringed their patents, and the defendants argued that there was an implied licence to repair, the claimants’ rights were exhausted by the marketing of the screens, and a person who repaired a screen was not making it within the meaning of section 60 of the Patents Act 1977.

In the first instance the judge found that there was an implied right of repair, but the Court of Appeal reversed that decision on the basis that the key question was whether the patented product had itself been made.

The House of Lords held that the doctrines of implied licensing and exhaustion of rights did not apply where the alleged infringement consisted of making the patented product. The right to repair was not created by licence but was a residual right belonging to the owner of the goods, whereby the owner could do whatever was necessary short of making the product.

In the present case, when the mesh was removed and the product was stripped bare, the remainder was merely an important component which could be used as a skeleton to make the product anew. The judge was wrong in identifying the product in the way that he did, and the Court of Appeal was right to reverse the judgement at first instance.


This is the name given to a defence – more a manner of pleading, in fact56 – whereby the defendant argues that the accused product or process (which is not identical to that in the patent) is not actually novel, raising issues of non-infringement and validity at a single stroke. This impales the claimant on the horns of a dilemma: either way, they lose. If they argue for a wide interpretation of the claims and prove that the infringing act has been committed, they fatally injure their patent: if they restrict the claims to avoid the novelty challenge, they lose the infringement claim. The classic case is Gillette Safety Razor Co v Anglo-American Trading Co57 which gives the technique another name: the Gillette defence.

Restrictions on the exclusive rights

The exclusive rights of the patent owner will sometimes be reduced to a mere right to remuneration for the use of the invention.

Crown use

There are extensive provisions permitting the Crown to make use of patented technology in certain circumstances. These are too detailed for consideration here58.

Licences of right

Section 46 of the Patents Act allows the proprietor of a patent to apply to the Comptroller for an entry to be made in the register to the effect that licence are to be available under the patent as of right. The patent is referred to subsequently as being endorsed ‘licences of right’.

The effect of this is that the proprietor gives away the monopoly nature of the patent. It cannot thereafter (unless the endorsement is cancelled under section 47) withhold a licence from anyone who wants one. The benefit to the proprietor is that the renewal fees become significantly less.

The terms of the licence, particularly relating to royalties, are to be settled by the parties. If agreement cannot be reached, it falls to the Comptroller to settle the terms. Existing licence holders may benefit from a decision to have the patent endorsed: they may apply to have their licences changed to ones on the same terms.

If the patent has been endorsed, in most infringement actions the defendant can avoid having an injunction granted against it by undertaking to take a licence on the terms so settled. This does not apply where the infringement consists of importing goods from outside the EC.

Compulsory licences

The Comptroller may order compulsory licences to be made available in certain circumstances (section 48). The power also allows the Comptroller to endorse the patent.

The grounds are set out in section 48(3). They cover:

  • non-working of the patent in the UK:

  • (where the patent is for a product) unsatisfied demand in the UK (or demand met by imports):

  • import substitution preventing the patent being worked in the UK:

  • refusal to licence on reasonable terms, leading to loss of exports, problems with working a dependent patent, or prejudice to the establishment or development of commercial or industrial activities in the UK:

  • specified conditions imposed in licences.

Section 50 sets out the purposes for which the Comptroller should exercise the power to order compulsory licences. He is charged with securing that patents which can be worked in the UK are worked there and that the proprietor of the patent receives a reasonable remuneration. He also must ensure that the interests of any person working or developing a patent in the UK under the protection of a patent are not unfairly prejudiced.

Competition Commission reports

The Competition Commission may report that certain matters operate against the public interest. This may occur on a monopoly reference, on a merger reference or on a competition reference.

If the Comptroller concludes that the matters complained of include the terms of licences granted by the proprietor of a patent, or the refusal of the proprietor to grant licences, he may cancel or modify any offending condition in a licence and make licences available of right. (Section 51).


Validity may be put in issue (inter alia) in an infringement action (section 74). This is not a defence, but a form of counterattack which complements the defence. Any person can apply for a declaration of invalidity.59 The grounds are set out in section 72, which is a complete statement of the law, meaning that old cases are of no relevance to its interpretation.60 They are:

(a) the invention is not a patentable invention;

(b) that the patent was granted to a person who was not entitled to be granted that patent;

(c) the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art;

(d) the matter disclosed in the specification of the patent extends beyond that disclosed

in the application for the patent, as filed, or, if the patent was granted on a new

application filed under section 8(3), 12 or 37(4) above or as mentioned in section

15(9) above, in the earlier application, as filed;

(e) the protection conferred by the patent has been extended by an amendment which should not have been allowed.

Of these the most commonly encountered grounds are that (a) and (c). Patentability is discussed at length in section Error: Reference source not found and sufficiency is discussed in detail in paragraph Error: Reference source not found.



Damages and costs can only be recovered for infringement where the patent is only partially valid if the claimant can show that the specification for the patent was framed in good faith and with reasonable skill and knowledge (see Section 63(2) of the Patents Act 1977). In Nutrinova Nutrition Specialities and Food Ingredients GmbH v Scanchem UK Ltd61 (18th May 2000) the High Court applied this principle to hold that infringement had taken place.

A party who infringes a patent will be liable for damages even though the patent is later held invalid. The principle of res judicata in English law means that once a matter has been litigated between the parties it cannot be reopened between them. On the other hand, Article 33 of the Community Patent Convention stipulates that if a patent is found to be invalid it will be revoked from the date of grant. The Court of Appeal held in Coflexip SA v Stolt Offshore MS Ltd.62 that the doctrine of res judicata applied not only to matters the subject of judgment between the parties but also every relevant point that they might have raised at the time (Henderson v Henderson63). Cause of action estoppel, where the cause of action in later proceedings is the as that in earlier proceedings, applied to the act that was held at trial to be an infringement. Issue estoppel, where a matter has been litigated and decided between the parties, applied to the additional acts of infringement that had been raised in the enquiry about damages after the trial. Both forms of estoppel prevented the defendant from relying on the revocation of the patent, which had occurred after trial but before the damages enquiry.

Innocent infringement

An infringer who does not know or have reasonable grounds to believe that a patent exists will not be liable to damages or an account of profits.64

1 This is one of those areas where the US patent system differs from most of the rest of the world. There, protection runs for 20 years from the date on which the patent is granted.

2OJ 1992 L182/1, now consolidated in Regulation 469/2009.

5British United Shoe Machinery Co v Simon Collier (1910) 27 RPC 567, SKF v Douglas [1991] FSR 522.

6Menashe Business Mercantile Ltd v William Hill Organisation Ltd [2003] RPC 575.

7This could include an invitation to treat: there is no need fora contractually-binding offer. No infringement takes place, however, if the supply is to be after the patent has expired.: Gerber Garment Technology Inc v Lectra Systems Ltd [1995] RPC 383.

8Section 60(1)(a) and see Smith Kline & French Laboratories v Harbottle (Mercantile) [1980]. On import and export see SKF v DDSA [1978] FSR 109.

9Section 60(1)(b) and see Fürr v Truline [1985] FSR 553.

10Section 60(1)(c). As to the meaning of the word ‘directly’, see Pioneer Electronics v Warner Music Manufacturing [1997] RPC 757 (CA).

11United Telephone Co v Sharples (1885) 2 RPC 25.

12Dunlop Tyre Co v British Colonial Motor Co (1901) 18 RPC 313.

13Gerber Garment Technology v Lecra Systems [1995] RPC 383.

14British United Shoe Machinery Co v Simon Collier (1910) 27 RPC 567.

15SKF v Harbottle [1980] RPC 363, which referred to Article 25 of the CPC not Article 7 of the Regulation.

16Hoffmann-LaRoche v Harris [1977] FSR 200.

17Pioneer Electronics Capital Inc v Warner Brothers Manufacturing Europe GmbH [1997] RPC 757 (CA).

18 [2004] UKHL 45. MFI took no part in the proceedings having settled the action at an early stage.

19See paragraph Error: Reference source not found.

20Tanglass Oy v Luoyang North Glass Technology [2006] FSR 608.

21 [2002] EWCA Civ 1702.

22 [2001] ADR.L.R. 11 July.

23 American Home Products Corporation v Novartis Pharmaceuticals UK Ltd. (Patents Court, Laddie J, 6 December 1999, reversed by the Court of Appeal, 22 July 2000).

24Hewickson v Tallon [1965] RPC 434.

25EMI v Lissen (1939) RPC 23, 39 (per Lord Russell).

26Sweet & Maxwell, 3rd ed. 1998.

27[1968] FSR 100, [1969] RPC 367 (HL).

28 A story is told of a scientist in a well-known British company seeking a solution to a technical problem. He commissions a search of the patent literature. The searcher eventually comes back to him with some good news and some bad news. The first piece of good news is that the solution to the problem is in the patent literature. The bad news is that the patent is still subsisting. However, there is a further piece of good news: the patent belongs to the very company for which the scientist works.

29Another piece of specialist patent vocabulary. In mathematics, of course, an integer is a whole number, and by analogy in patents it is an element of an invention, one of the smallest whole units into which the invention can be broken down.

30Marconi v British Radio Telegraph & Telephone Co (1911) 28 RPC 181, Birmingham Sound Reproduction v Collaro [1956] RPC 232.

31 [1963] RPC 61.

32[1969] RPC 367.

33[1981] FSR 60, [1982] RPC 183 (HL).

34Codex v Racal-Milgo [1983] RPC 369.

35 You might think, as I did, that had the word ‘substantially’ been included so as to qualify ‘vertically’, much trouble could have been avoided. The claim does contain that word, but it qualifies the word ‘horizontally’ which describes another part of the lintel. The draftsman of the patent believed that the concept of ‘extending vertically’ simply meant that the top of the face had a more elevated position than the bottom, not that it had to be directly above, so the word ‘substantially’ would to his mind have been redundant.

36[1990] FSR 181.

37For example, Daily v Berchet [1992] FSR 533, Electrolux v Black & Decker [1996] FSR 595, American Home Products Corp v Novartis Pharmaceuticals UK Ltd [2001] RPC 1.

38Note 29, supra.

39Note 39, supra.

40[1995] RPC 585. See also Jacob J’s judgment in Beloit v Valmet, [1977] RPC 489.

41[2000] EWCA Civ 209.

42Wahrheit v Olympia Tools Ltd [2003] FSR 95.

43Pharmacia Corp v Merck & Co Inc [2002] RPC 775 (CA) and Merck & Co Inc v Generics (UK) Ltd [2004] RPC 607.

44[2004] UKHL 46.

45 See Crespi, [2004] CIPA 693, 695.

46[2009] EWCA Civ 1062.

47339 U.S. 605, 607 (1950).

48(CA2nd Conn) 168 F2nd 691, 692.

49Section 60(5)(a).

50See Monsanto v Stauffer [1985] RPC 515.

51Section 60(1)(c).

52Stena Rederi Aktiebolag v Irish Ferries [2003] RPC 681.

53Section 64, as amended by the Copyright Designs and Patents Act 1988.

54Note 89, supra.

55[2000] 4 All ER 353, [2000] UKHL 42.

56Per Lord Evershed in Page v Brent Toy Products (1950) 67 RPC 4.

57(1913) 30 RPC 465.

58Sections 55 to 59.

59Cairnstores Ltd v AB Hassle [2002] FSR 95.

60Genetech Inc’s Patent [1989] RPC 147, Mentor v Hollister [1991] FSR 557.

61[2000] EWHC Patents 98.

62 [2004] EWCA Civ 213, [2004] FSR 34 .

63 [1843] 3 Hare 100.

64Section 62.