Nampak Plastics Europe Ltd v Alpla UK Ltd

This is an appeal by the claimant patentee (“Nampak”) from a decision of Birss J dated 3 July 2014 by which he gave summary judgment to the defendant (“Alpla”) in its action for a declaration of non-infringement of United Kingdom Patent No 2 494 349 (“the patent”). Nampak complains that the judge wrongly granted summary judgment without the benefit of hearing expert evidence which would throw light on the issues of construction and infringement and in consequence approached those issues on a basis which was at least arguably incorrect. Birss J granted permission to appeal. Mr Iain Purvis QC and Ms Anna Edwards-Stuart argued the case for Nampak. Mr Adrian Speck QC and Mr Henry Ward responded on behalf of Alpla.

Nampak began the present proceedings claiming that a first design of plastic milk bottle made and sold by Alpla, the ECO 1, infringed the patent. Alpla denied infringement and subsequently made an application for a declaration of non-infringement in respect of a proposed second design, the ECO 2, which it had not yet launched, no doubt hoping that the design changes introduced in the ECO 2 were sufficient to place the product outside, or as they would say, further outside the claims of the patent. Nampak having declined the invitation to acknowledge non-infringement, Alpla applied for summary judgment. We are only concerned on this appeal with the summary judgment application in relation to the ECO 2. The trial of Nampak’s infringement action on the ECO 1 is fixed to come on in January 2015. In consequence we have heard this appeal with a degree of expedition (and in the vacation) in order to avoid potential disruption to trial preparation.

Nampak’s pleaded case in its action for a declaration of non-infringement was that it did not admit that the ECO 2 was not an infringement of the patent. Following a request for further information, Nampak stated that it did not have a positive case on infringement. As the judge explained, this turned out not to be the case: Nampak’s case before the judge was that the ECO 2 did infringe. Nevertheless the judge allowed Nampak to argue its positive case on infringement before him, as he considered that it would be unfair to refuse to do so. By its respondent’s notice, Alpla argues that the judge was wrong to permit Nampak to advance a positive case, and it would not have been unfair to refuse to permit it to do so.

The provisions of CPR Part 24 which allow for cases to be disposed of by summary judgment without a trial apply to patent cases as they do to other disputes. Applications for summary judgment in patent cases have not often met with success, however.

Thus, in Monsanto & Company v Merck & Co. Inc [2000] RPC 77 the issue was whether there was infringement by virtue of the presence of a compound at the final stage of a manufacturing process of the active ingredient, but not subsequently, when the claim required the use of the active ingredient “to prepare a medicament”. Pumfrey J held that there was no infringement. He considered that the issue involved no more than the construction of the words of the claim, and there had been no suggestion that there was any evidence which could have a bearing on that issue. The Court of Appeal allowed the appeal. Aldous LJ, with whom Auld LJ agreed, took a different view of the construction of the claim from Pumfrey J, and said this:

“I do not believe that the judge was right to conclude that the alternative case put forward by the patentees is unarguable upon the assumed facts. Despite the view as to the meaning of claim 20 which I have expressed above, it would not be right, at this stage of the action, to come to any concluded view as to the ambit of claim 20. The patent must be construed as a whole and the claims interpreted according to the Protocol on Interpretation. The subject of the specification is complicated. To come to a concluded view, the mantle of a man skilled in the art must be adopted. That will require the aid of expert evidence.”

Auld LJ agreed that it was necessary to construe the patent as a whole and to “don the mantle of the person skilled in the art”. He said that it was:

“impossible at this stage of the action and certainly without evidence on the matter, to say that the plaintiff’s case for the wider construction … is so plainly and obviously wrong that the claim should be struck out.”

None of this goes as far as to suggest that in a straightforward case it may not be a legitimate exercise of the power in CPR Part 24 to grant summary judgment in a dispute about the construction of a patent. In Virgin v Delta [2011] RPC 18 Jacob LJ said this:

“Whilst the general rules as to summary judgment apply equally to patent cases as to other types of case, there can be difficulties, particularly in cases where the technology is complex. If it is, the court may not be able, on a summary application, to form a confident view about the claim or its construction, particularly about the understanding of the skilled man. On the other hand in a case such as the present, where the technology is relatively simple to understand, there is really no good reason why summary procedure cannot be invoked. No one should assume that summary judgment is not for patent disputes. It all depends on the nature of the dispute.

“That can cut both ways, of course. If the court is able to grasp the case well enough to resolve the point, then it can and should do so – whether in favour of the patentee or the alleged infringer.”

However in Saleem Khatri v Cooperative Centrale Raiffeisen-Boerrenleenbank BA [2010] EWCA Civ 397, Jacob LJ also said (not this time in a patent case):

“Although a case may turn out at trial not to be really complicated, it does not follow that it should be decided without a fuller investigation into the facts at trial than is possible or permissible on an application for summary judgment. Thus the court should hesitate about making a final decision without a trial, even where there is no obvious conflict of fact, where reasonable grounds exist for believing that a fuller investigation into the facts of the case would add to or alter the evidence available to a trial judge and so affect the outcome of the case."

It is clear that the fact that a dispute involves the resolution of an issue of construction of a patent does not automatically render it unsuitable for summary judgment. However it is necessary to proceed with caution given that the court is not being called upon, when construing a patent, to decide what the words of the patent mean to it, but what they would have been understood to mean by the person skilled in the art: see per Lord Hoffmann in Kirin-Amgen Inc v Hoechst Marion Roussel ltd [2004] UKHL 46; [2005] RPC 9 at [32] – [33]. Such an exercise is dependent upon the identity of the person skilled in the art and the knowledge and assumptions which one attributes to him or her.

That said, it remains the law that expert evidence is not admitted as to the meaning of ordinary English words which have no special or technical meaning in the art. Once equipped with evidence as to the knowledge and assumptions of the person skilled in the art, the determination of the meaning which the words used in a patent claim would convey to one skilled in the art is for the court.

It follows from what I have said that, on a summary judgment application such as this, it is necessary for a party who claims that the court is inadequately equipped to decide an issue of construction to identify, perhaps in only quite general terms, the nature of the evidence of the common general knowledge which he proposes to adduce, and to be in a position to explain why that evidence might reasonably be expected to have an impact on the issue of construction. If that party is not able to do so, it is open to the court to conclude that he is simply hoping that “something may turn up” and that his defence does not have the necessary “reality” to avoid summary judgment under Part 24.

Given the very large number of plastic milk bottles manufactured each day, it is not surprising that manufacturers are keen to restrict so far as possible the amount of plastic used to make them whilst retaining sufficient structural integrity to make them fit for purpose. A blow-moulded milk bottle is manufactured from a “parison” which is a lump of plastic which is blown outwards from the centre towards the walls of a mould. As a consequence of this process, the parts of the mould which are furthest from the centre of the mould where the parison is initially placed are the most stretched and therefore thinned. In what the patent describes as a known container, the footprint is substantially rectangular in outline in plan view. Thus the corners are the furthest points from the centre and most subject to the thinning problem. The invention is said to provide a milk bottle that overcomes or mitigates this problem.

The patent then goes on to describe various “aspects of the invention”, the description of each of which is supplemented in general terms by “exemplary” or “preferred” embodiments. A common feature of everything described is that the containers have a “part line” which represents the two halves of the mould and which bisects the container in a vertical plane (through the handle, if it has one). It is explained, starting at page 3 line 20, that, in the invention and unlike the conventional substantially rectangular containers, the maximum radial extent from the centre is not “away from the part line” (e.g. at the corners), but at the point of intersection of the container with the part line (i.e. on the part line itself). This is said to reduce the tendency for localised thinning. However, despite the emphasis on this idea, the specification does not make use of it directly as a feature of the claims.

As the judge noted, the description of the first of these “aspects” appears to describe quite closely (without expressly saying so) the first of three specific embodiments of the invention tackled in more detail at a later stage. This embodiment is shown in figure 3, which I reproduce below:

The specification notes that, at the left hand end in the diagram, the footprint of the container consists of “two curved or radius sections separated by a straight section … or may consist of a continually curving section” and that the right hand end defines “divergent portions which extend at an acute angle to the part line of the container”.

It is convenient to refer at this point to later parts of the specification which describe the figure 3 embodiment in more detail. Thus at page 15 line 16 onwards it is explained that the opposed side surfaces 114 and 116 are parallel and that this has advantages for automated filling lines (which have guide rails which would be parallel to these side surfaces). However, it is explained that:

“… the side surfaces may define a slight curvature or other non-linear configuration whilst retaining at least a general alignment with the part line ... (and thereby advantageous alignment on filling lines, e.g. relative to opposing guides between which the containers travel on said filling lines).”

At page 16 line 6 the specification explains (still in relation to figure 3):

“In effect, the footprint defines a significantly truncated rectangle, wherein the maximum radial extent of the footprint from the centre point is greatest along the part line …rather than away from the part line … (as in the case of conventional rectangular or square containers).”

The specification then goes on to describe the opposing longitudinal ends 118 and 120. The left hand end 118 is described in similar terms to before. A rounded end of this kind is said to give improved resistance to bulging as compared to squares or rectangles. The right hand end 120 is said to include a straight section 130 extending generally perpendicular to the part line and which separates the divergent portions 126 and 128. It explains that this avoids the use of a sharp corner at the angled end, which might otherwise lead to the deformation of the opposite end of another container when the containers are being moved on the filling line. The straight section 130 is said to be of the same length as the straight section 124 on the opposite end of the container, but in other embodiments both sections may define a slight curvature, but are nevertheless aligned at least generally perpendicular to the part line, and define generally transverse surfaces for abutment between adjacent containers on a filling line.

Thus the purpose of the straight sections is both to prevent a sharp corner at the angled end and to provide an abutment surface for adjacent containers.

A further aspect of the invention, apparently tracking figures 8 and 9 of the patent, is described from page 6 line 20 onwards. It is enough to show figure 8, which is the plan view:

The figure 8 footprint is described as generally octagonal, with first and second pairs of opposing sides. These can be seen at top and bottom and left and right in the figure 8. The two pairs are described as “orthogonal” (i.e. at right angles to each other). The two sides which intersect the part line are said to be further away from the centre than the two which are parallel to it. The specification again stresses that the maximum radial extent from the centre is along the part line and not away from it as in conventional containers. The specification also draws attention to the fact (page 8, line 4 and again at page 8, line 15 and page 20 lines 14-15) that the length of the sides which intersect the part line is less than the diameter of the pouring aperture. It does not explain why this is significant. Both pairs of sides are allowed to follow a slight curvature, as with figure 3. This is said to be shown in the figure.

A further aspect is described by reference to figures 14 and 15. This is more of a truncated rectangle than “generally octagonal”. Figure 14 is reproduced below:

The specification describes this aspect as having “four truncated corner regions”. Again, this is said to reduce the tendency for localised thinning compared with rectangular footprints.

The specification later describes this embodiment as having four major sides arranged as two opposing pairs, the shorter pair being those which bisect the axis. There are also four minor sides extending between the four major sides at an angle of inclination to the part line. From the description and the figure it is clear that in this embodiment the length of the major sides which intersect the part line is greater than the width of the aperture.

The judge and the parties adopted the following breakdown of claims 1 and 2 of the patent:

A milk container according to claim 1 wherein the sides in said first opposing pair are straight and are bisected by the part line.

The feature of the claim on which (or rather, on the absence of which) Alpla based its application for summary judgment was feature K of claim 1 which requires the length of the pair of opposing sides which intersect the part line to be shorter than the width of the aperture.

The judge explained the design of the ECO 2 and the issues between the parties at [49] to [56] of his judgment in terms which I gratefully adopt:

“A drawing of the footprint of the ECO 2 (4 pint) is set out below.

This is a plan view from above. The pouring aperture can be seen and the handle is on the right. The part line runs horizontally. The first opposing sides are the sides which run perpendicular to the part line. Alpla submits the first sides are plainly larger than the diameter of the aperture. Therefore the product does not infringe. Alpla submits these first sides have a slight curve but that does not take them out of the claim.

Alpla produced a colour mark-up of the b/w original drawing. Two green horizontal lines have been added for the purposes of argument (they come out grey in b/w). The drawing is:

Alpla submits that the first sides extend between the lines shown in green running across horizontally in the picture. The relevant left hand first side is marked in red.

Nampak say Alpla's characterisation of the sides in the container is not correct. To explain its case, Nampak produced a drawing which divides up the shape of the footprint into regions numbered 1 to 12 as shown below:

The footprint has been turned round. The part line now runs vertically. Different regions of the perimeter have been numbered 1 to 12. There is also a pair of vertical lines drawn for the purposes of the case.

Nampak says the relevant first opposing sides are the regions numbered 7 and 1. These regions are shorter in length than the diameter of the pouring aperture. Each truncated corner consists of two regions: 2+3, 5+6, 8+9 and 11+12. On this basis the ECO 2 satisfies features J and K of the claim. Whether the bottle actually infringes the claim depends on all the other claim features being satisfied but at least at this stage, the only features focussed upon are features J and K. Thus submits Nampak, since they are satisfied by the ECO 2, the application for a declaration of non-infringement should be rejected.

So really the dispute is a simple one. Alpla contends that one relevant side consists of regions 6, 7 and 8 together and the other relevant side consists of regions 12, 1 and 2 together. If that is right then the bottle plainly does not satisfy the claim because the length of each of those sides is much greater than the aperture diameter. Nampak contends that the relevant sides are just regions 7 and 1. The other regions should be regarded as part of the truncated corners. If that is right then the bottle satisfies feature K because the side is shorter than the aperture diameter. Nampak submits that its approach to the application of the claim to the ECO 2 has a realistic prospect of success and so the matter should go to trial.”

The judge dealt first with the construction of feature K. He regarded it as clear for summary judgment purposes that a side, in the sense meant by the patent, could be straight or slightly curved. A side did not cease to be a side merely because it was not geometrically straight. The patent said so in terms, and he did not regard it as realistic to suppose that evidence could come to light which would lead to a different understanding on this point. As a consequence he felt able to reach the following conclusions about the construction of feature K:

“This conclusion has important consequences when a skilled person comes to consider how to characterise a container for the purposes of the claim in order to decide whether it is within the claim (either for the purposes of validity or infringement):

i)

First, since a side may be have a slight curve, a minute analysis of whether a part of the shape is geometrically straight or has a slight curve is unlikely to matter.

ii)

Second, one cannot avoid the claim by having a footprint with no straight elements. A footprint with no straight elements may well fall within the claim. The fact that a region may include straight and slightly curved elements does not, of itself, mean that a region is not a "side".

iii)

Third, in a region which has both straight elements and slightly curved elements, one cannot simply ignore the curved regions and take the view that the straight element is the "side" in question.

iv)

Fourth, this conclusion applies just as much to a side or face which is part of a truncated corner as it does to the opposing sides of the overall rectangular shape. In other words the fact a region has a slight curve does not prevent its being regarded as part of a truncated corner region.”

The judge rejected an argument by Nampak that the approach to determining infringement was to draw a notional rectangle (supposedly representing a conventional container) around the ECO2. Those parts which were the truncated corner were those which diverged from the rectangle. The judge considered that this ignored the patent’s teaching that sides could be slightly curved. He also rejected as too vague Nampak’s argument that one need to know how the bottle behaved on a filling line before one could determine infringement. His conclusion is contained at [74] to [76] of his judgment:

“A skilled person would consider the question arising in this case by looking at the shape in plan view, just as in the figures in the patent. If pairs of opposing sides can be identified by observation then no problem arises. If a skilled person looks at the ECO 2 they see a broadly rectangular shape in which all four of the corners have been truncated by removing a generally triangular piece. There are two pairs of opposing sides. The sides in the pair parallel to the part line are straight. The sides of the other pair have a slight curve but are still generally perpendicular to the part line. The precise shape of each perpendicular side consists of a straight part in the middle and two gently curving parts extending to the truncated corners. This is the natural way of characterising the shape of the ECO 2.

The only arguments advanced why a skilled person would not characterise the product in this way are (i) Nampak's argument based on the notional rectangle, which I have rejected as a matter of construction, and (ii) Nampak's argument based on the behaviour of the bottle on a filling line which I have rejected because it is too vague.

Considering the shape of the Alpla ECO 2 product characterised in this natural way, it does not infringe because it does not satisfy feature K. The length of the first pair of opposing sides is greater than the diameter of the pouring aperture.”

On this appeal Mr Purvis renews the arguments which failed to persuade the judge. He submits that it is possible, by applying the notional rectangle approach to identify a straight side, which, as claim 2 shows, is the paradigm case of a side. Figure 14 adopts this approach, he submits, showing curved sections next to a straight side as part of a truncated corner and not as part of the side. Figure 3 also shows straight opposing sides bounded by curved sections. He submits that, the judge purported to characterise the ECO2 through the eyes of the skilled person, when he could not do so without expert evidence as to the knowledge or attributes that the skilled person would possess. The judge was wrong to reject Nampak’s case as to why expert evidence would be likely to be relevant to whether the adjacent curved sections, or any part of the curved sections, should be included within the opposing sides.

Mr Speck supported the judgment and its detailed reasoning. He stressed those parts of the specification which gave latitude to the designer to depart from straight sides. The purpose of the opposing sides was also to prevent sharp ends damaging adjacent containers.

This is not a complicated case. It follows that the warnings in the authorities about the dangers of summary judgment in patent cases do not apply with such force. I should nevertheless have in mind that, even in a case which appears simple at this stage, there may be reasonable grounds for allowing a fuller investigation at a trial.

Before turning to the question of infringement by the ECO2, it is necessary to construe feature K of the claim. That cannot be done in isolation, as the various features of the claim are to a degree inter-dependent. Its relationship with feature J, which requires truncated corners, is particularly important.

A consideration of the specific embodiments of a patented invention can be a useful aid to construction of a claim. If there is a debate as to the scope of a particular word or phrase in a claim, it may be helpful to examine specific embodiments which the patentee has stated fall within the claim. Normally one would not expect the patentee to exclude subject matter which is expressly or impliedly stated to be an embodiment of his invention. If a propounded construction of a word or phrase would exclude an embodiment of the invention, then it is legitimate to infer that that construction is too narrow. But like most canons of construction this is not an absolute rule and must yield to the common sense and understanding of a person skilled in the art. Features drawn from some but not all of the specific embodiments may be added to the claims during prosecution of the patent, and the need to make clear which embodiments do and which do not fall within the claims can be overlooked. In the present case the third embodiment described plainly does not have feature K, as is made plain in the body of the specification, because the sides in question are longer than and not shorter than the width of the pouring aperture. The skilled person would understand that the patentee had described an embodiment which fell, at least on that ground, outside the claims, not that “shorter than” must be construed as including “longer than”. Mr Purvis, quite rightly, did not advance any argument of this kind.

The judge also noted that it was not clear whether the left hand configuration in figure 3 fell within the claim because it might not satisfy feature J. Feature J calls, amongst other things, for a side or face which is at an angle to the part line. He doubted whether an entirely rounded corner could satisfy this language – but he did not feel it necessary to decide that issue. Of course, a further problem one has with the left hand configuration in figure 3, on the assumption that it falls within the claim, is how one determines where the side ends and the truncated corners begin, particularly in the case of the continuously rounded corner. Unless that can be done, it may be impossible to tell whether the sides are shorter than the aperture.

The claim defines a milk container having orthogonal pairs of sides, one essentially parallel to the part line and one essentially perpendicular to it. Neither pair is required to be straight: they may be slightly curved. The purpose of having these opposed pairs of sides is to retain at least some of the generally rectangular shape of the conventional bottle, so that one pair will align with the guides and the other pair will abut with adjacent bottles on the filling line.

The sides which bisect the part line are required to be shorter than the sides which are parallel to it. The judge ventured the suggestion that the purpose of this feature was to produce a profile which allows for a larger truncation of the corners than if all the sides had the greater length. This might help to achieve the purpose of ensuring that the point of maximum radial distance from the centre is at or close to the part line. I think that it is an entirely fair inference.

Feature K introduces the requirement that the shorter sides are shorter than the aperture. The specification does not explain why this is significant. Mr Purvis speculated that, because the size of such apertures is fairly standard, by requiring the sides to be shorter than the aperture the feature is adding an element of scale. Mr Speck did not accept that this was the purpose of the feature, and pointed to the submissions before the judge where neither side had come up with a purpose. I am prepared to assume in favour of Nampak that the purpose is that advanced by Mr Purvis.

In order to answer the question which is posed by the issue of infringement, it is necessary to see what guidance the patent gives as to how one determines the length of the relevant sides. In many cases this may be a straightforward exercise, as it is with some of the figures in the patent. But in other configurations, like the left hand side of Figure 3, it may not be so clear.

To my mind it is clear that the relevant sides called for by the claim are straight or slightly curved portions which abut an adjoining container on the filling line. But it cannot seriously be suggested that the side needs to abut the adjoining container along its entire length. So the purpose of the side – abutment - is not a guide to where the side starts and finishes.

More relevant is the apparent common ground between the parties recorded by the judge at paragraph 46 of the judgment that the side was the length bounded by the truncated corner. Given that the purpose of the length requirement is, as the judge found, to allow for larger corners, it would seem to follow that where the side ends the truncated corner begins.

The fact that both the sides and the truncated corner of the container called for by the claims may include curved portions necessarily means that the transition from side to corner may not be visible as a sudden change in angle or radius. Given that a completely rounded figure 3 embodiment falls within the claim, the process of identifying where the side ends and the truncated corner begins may not be a question of simple inspection. The key question for present purposes, as it seems to me, is whether, at the point that the section of perimeter passes the extremity of the aperture, it is properly to be regarded as part of the side or part of the truncated corner.

I agree with the judge that it is inappropriate to answer this question by superimposing the notional rectangle on the footprint. That approach ignores the possibility of curved sides. The fact that it is used for illustrative purposes in some of the figures in the patent is unpersuasive. Much more persuasive is the fact that such an approach would result in there being no sides at all in the completely rounded Figure 3 embodiment.

The conclusions which I have reached thus far on the construction of the claim are based on a reading of the specification. I am unable to see how they could be shown to be erroneous by subsequent evidence.

I turn therefore to consider the issue of infringement. Like the judge, I am unpersuaded that there is anything to be learned from the behaviour of the bottles in a filling line, given that it cannot be expected that the sides are required to abut adjacent containers along all their length.

My only point of disagreement with the judge’s ultimate approach to determining infringement, effectively by deciding how a skilled person would view the container by observation, is that it may inappropriately focus on the geometric appearance of the footprint, rather than asking the question I have identified above. The critical question is how one should characterise the section of perimeter both before and after the point at which they cross the width of the aperture. The judge was prepared to hold that the whole of the sections 12,1,2 and 6,7,8 were part of the side, and that the curvature of 12,2,6 and 8 did not go beyond the “slight” curvature permitted by the patent. They were still essentially perpendicular to the part line. Given the point made above, that a transition from side to corner may be smooth, the critical question is whether the transitions arguably occurs within sections 12,2,6 and 8 at a point short of the width of the aperture, so that the sides are less than the width of the aperture.

This point, however, does not assist Nampak. At the point at which the perimeter passes the extremity of the aperture it has departed less, not more, from the strictly perpendicular. It seems to me that the suggestion that the side has departed so much from the perpendicular at that point that it has become part of a truncated corner completely lacks reality. It is also not one on which the evidence of an expert in plastic milk bottle production could conceivably assist. It was for Nampak to show that there was at least some basis for supposing to the contrary, and it failed to do so.

It follows that I can see no purpose in sending this issue for trial. There is nothing which it has been credibly suggested could emerge at trial which could have a bearing on the conclusion I have reached. I would therefore dismiss the appeal. It is not necessary to resolve the issue raised by the respondent’s notice.