Virgin Atlantic Airways Ltd v Jet Airways (India) Ltd & Ors

In these proceedings, Virgin Atlantic Airways Limited (“Virgin”) alleges infringement of three European patents. They are numbers 1 495 908, 2 272 711 and 2 289 734. For convenience I will use the last three numbers 908, 711 and 734.

There are three infringement actions, brought against the airlines Delta Air Lines Inc. (“Delta”), Air Canada, and Jet Airways (India) Limited (“Jet”). Contour Aerospace Limited (“Contour”) is joined as a defendant to each of those actions with the relevant airline. Contour is an aircraft seat manufacturer. All three patents are in suit in these actions. By an order I made on 19 October 2011, the issues of alleged joint tortfeasance between these parties have been stayed, to be decided if necessary in any inquiry as to damages or account of profits. Jet has not taken any recent active part in the proceedings, but Delta and Air Canada have. In each action there is a counterclaim for revocation of the patents.

There is a fourth action before me, between Contour and Virgin, in which Contour seek a declaration of non-infringement in respect of a modified design of seat.

Finally, an appeal by Premium Aircraft Interiors (“Premium”, a company in the Contour Group) from a decision of the United Kingdom Intellectual Property Office (UKIPO) is also before me. The appeal relates to the propriety of the United Kingdom being designated in respect of 908. This point is also raised by way of defence in the infringement actions.

The 908 patent was granted in May 2007. It was the subject of an opposition in the European Patent Office (“EPO”) by Premium, as well as Airbus Industrie, the aircraft manufacturer, and Cathay Pacific, the airline. The opposition proceedings ultimately resulted in the patent being amended, so as to restrict its scope from that which was the subject of the original grant. In what follows I will endeavour to refer to “unamended 908” and “908” to indicate the unamended and amended patents respectively.

908 (in either amended or unamended form) relates to a seating system for passenger aircraft in which the seats are arranged in an “inward facing herringbone” configuration. The seats convert into lie-flat beds. The description in the patent shows how the seats can be arranged in a space-saving configuration. In addition it shows how a flip-over mechanism can make use of the back of the seat as part of the lie-flat bed. In this introduction, it will be important to keep in mind these distinct aspects of what is disclosed in 908: space-saving and flip-over.

The parties have been engaged in litigation for some years over the sale and supply of Contour’s Solar Eclipse seat which is alleged to infringe 908 both before and after amendment. The procedural history is relevant, somewhat complicated, and as follows.

Virgin sued Contour before the opposition proceedings had come to an end for infringement of unamended 908. These earlier proceedings are referred to as “the Contour Action”.

Before the Contour Action reached trial, Virgin sued three customers of Contour, the airlines Delta, Air Canada and Jet, in the present three “Airline Actions”, alleging that the airlines were liable for infringement of unamended 908 as joint tortfeasors with Contour. Those actions were stayed by formal or informal agreements pending the outcome of the Contour Action.

In the Contour Action at first instance, before Lewison J as he then was, Virgin were unsuccessful. Lewison J held unamended 908 to be valid over the cited prior art (principally BA First, Airbus and common general knowledge alone). He went on to hold that unamended 908 was not infringed by the Solar Eclipse seat. He held that the claims were limited to the flip-over mechanism, a feature not taken by the Solar Eclipse. Both sides appealed the respective adverse findings. Lewison J’s judgment was given on 21^st January 2009.

On 22^nd October 2009 the Court of Appeal gave judgment on the appeal. The Court of Appeal reversed Lewison J’s decision on infringement of unamended 908, holding that the claims were not limited to flip-over seats. Lewison J’s judgment that the unamended patent was valid over the cited prior art and common general knowledge was upheld, notwithstanding the wider construction adopted for the purposes of infringement. In short, the invention with which 908 was concerned was the spacesaving invention: and that invention remained novel and inventive having regard to the cited prior art.

After judgment in the Court of Appeal in the Contour Action, Contour asked the Court of Appeal to allow certain run-off supplies of Solar Eclipse seats to Delta pursuant to existing contractual arrangements. The Court of Appeal permitted this. Contour was required to procure undertakings from Delta not to use the Solar Eclipse seats on transatlantic routes competitive with Virgin. There is an issue between Virgin and Delta as to whether that undertaking applies after the amendment of the 908 patent.

Virgin had by now succeeded on both infringement and validity. In consequence, the Court of Appeal ordered an inquiry as to damages. Contour applied for permission to appeal to the Supreme Court in relation to the findings of infringement and validity, as well as in relation to the inquiry as to damages. The latter application for permission to appeal related to the point which arises out of the judgment in Unilin Beheer v Berry Floor [2007] EWCA Civ 364 as to recovery of damages at a time when validity is still in issue, as it was here, in opposition proceedings before the European Patent Office (“the Unilin point”). Under the law as it stands, this fact alone does not provide a ground for a stay of an inquiry as to damages, because damages are payable even if the patent is subsequently revoked or amended. The Supreme Court refused permission to appeal on the patent points, as well as, at this stage, on the Unilin point.

On 9^th September 2010 the Technical Board of Appeal (“TBA”) of the EPO decided that the 908 patent could be maintained with amended claims. The earlier decision of the Opposition Division (“OD”) had introduced an amendment to require that the claims were limited to seats which did not swivel (something which the Court of Appeal considered was already implicit in the claims). The main issue in the present trial is whether the amendments introduced before the TBA have narrowed the claims so as to affect the Court of Appeal’s finding of infringement by the Solar Eclipse seat.

For present purposes the relevant amendments introduced before the TBA were (a) the introduction into the claims of a requirement for a generally triangular support element and (b) the deletion of claims directed to individual seats for aircraft seating systems: the only claims in 908 are now to seating systems for aircraft. The additional feature was necessary because the TBA took the view that the unamended claim lacked novelty over BA First. In coming to that conclusion, the TBA took a different view of the construction of the unamended claim from that taken by the Court of Appeal. Once the claim was amended, the TBA rejected the validity attacks based on the cited prior art (which included BA First and Airbus) and a number of added matter attacks.

Following the TBA decision, and now that the patent had been amended in a way in which Contour contended meant that the Solar Eclipse did not infringe, Contour sought permission to appeal to the Supreme Court on the Unilin point for a second time. In March 2011, the Supreme Court granted Contour permission to appeal on the Unilin point. However, if the result of this action is that the Solar Eclipse seats are held still to infringe, the appeal will have no impact on the damages payable, and will therefore be unnecessary. Accordingly the Supreme Court has fixed its hearing for April 2013, when the outcome of this action will be known.

In November 2010, Delta applied for summary judgment in its Airline Action based on a number of non-infringement arguments. Delta succeeded before Arnold J on 30^th November 2010. Delta contended that the deletion of the claims to aircraft seats as a result of the TBA decision meant that there could be no infringement. The remaining claims, on Delta’s argument, required the seating system to be installed on an aircraft (“the aircraft argument”). This only happened abroad. Virgin had a counter argument, which arose if Delta were right on the aircraft argument, that there was infringement by supplying in the United Kingdom an incomplete kit of parts for assembly abroad (“the incomplete kit of parts argument”). Arnold J held that there was no infringement on the aircraft or incomplete kit of parts arguments. He held that it was arguable that, depending on the correct construction of the claim, there might be infringement by supplying a complete kit of parts, but this point did not arise as the kit of parts, on his view of the claim, required an aircraft.

On 23^rd February 2011, the Court of Appeal reversed the judgment of Arnold J on the aircraft argument. The Court of Appeal held that the claims did not require a seating system to be on an aircraft, merely to be suitable for an aircraft. As, on this basis, there had to be a trial, and as the kit of parts involved was now complete, the Court of Appeal did not need to decide the complete kit of parts argument.

On 14^th March 2011, Premium applied in the UKIPO under Rule 50 of the Patents Rules 2007 for the designation of the United Kingdom to be removed from 908, because of alleged irregularities in procedure in the EPO. On 10^th August 2011 the application was dismissed by the Hearing Officer, Mr A.C. Howard. Premium have appealed that decision, and Virgin have intervened in the Appeal. That Rule 50 appeal is before me in the present proceedings as well.

The application for 711 was published on 12^th January 2011 and the patent granted on 29^th June 2011. The application for 734 was published on 2^nd March 2011 and the patent granted on 19 October 2011.

The claim based on the Divisionals is made against Contour alone, as Virgin do not believe that the airlines will buy more Solar Eclipse seats from Contour. The claim is made quia timet, and only for injunctive relief, because the Divisionals were not published until after the sales of the Solar Eclipse were made.

In all this, Mr Richard Meade QC, Mr Alan Maclean QC and Mr Henry Ward argued the case for Virgin. Mr Iain Purvis QC and Mr Brian Nicholson argued the case for Contour and Premium. Mr Adrian Speck QC and Mr Pushpinder Saini QC argued the case for Air Canada. Mr Benet Brandreth argued the case for Delta. Ms Fiona Clark appeared on behalf of the Comptroller of Patents.

Contour called Mr John Higgins to prove the Product Description, in particular on the Solar Eclipse seat. Mr Higgins is the former Business Development Director of Contour. He is now employed as Principal Engineer for Heath Tecna, Inc. in the

The parties called experts in the fields of aircraft seating system design, mechanical testing, and foreign law. In addition both sides called a patent agent in relation to procedure before the EPO.

Virgin called Mr Jaime Moreno as its aircraft seating design expert. Mr Moreno was Virgin’s expert in the Contour Action. Mr Moreno runs a design studio in Madrid and has worked on a number of aircraft design projects. Although he could not claim as much experience in aircraft design at the priority date as the defendants’ aircraft seating system design expert, I did not think that this mattered. His evidence in crossexamination was entirely straightforward and helpful. Mr Purvis pointed to one or two instances where Mr Moreno had retracted statements made in his reports, but did not suggest that I should discount his oral evidence as a result.

Contour and Delta called Mr Martin Darbyshire as their aircraft seating system design expert. Mr Darbyshire is managing director of a design company, tangerine. From 1998 to its launch in 2000 he led the team which created BA’s Club World business class lie-flat seat called Yin-Yang. From 2003 to 2005 he led the team which created the second BA Yin-Yang seat. He was at the priority date, and remains, a distinguished and successful aircraft seating system designer.

Mr Meade submitted that Mr Darbyshire had “limitations and shortcomings as an expert witness”. He submitted that Mr Darbyshire was of an inventive and imaginative disposition, that he had used hindsight in his analysis of obviousness, had oversimplified the issues, been selective in his evidence, had relied on instructions and materials supplied to him without discrimination, had allowed his report to be prepared in a sequential fashion calculated to lead to Virgin’s invention, and given evidence inconsistent with his previous report. Mr Meade was careful to point out that this was not an attack on Mr Darbyshire’s integrity or a suggestion that his evidence was designed to mislead.

For my part I found Mr Darbyshire to be a highly knowledgeable and helpful expert witness. It is true that, when I come to resolve the issues of inventive step, I must bear in mind that Mr Darbyshire’s level of skill is at the high end of the spectrum, well beyond those of average skill in the art. But that is not a limitation or shortcoming of Mr Darbyshire as an expert witness. It has never been the case that the court asks to hear only from those who personify the unimaginative skilled person, Jacob LJ’s “nerd”: see Technip France SA’s Patent [2004] EWCA Civ 381;[2004] RPC 46 at [7] - [12]. In that passage he explained that it was not so much the expert’s view on the ultimate issue of obviousness which mattered, but the reasons he or she gives for that view.

In analysing the issue of obviousness, the court must of course be wary of hindsightdriven arguments. Mr Meade said that Mr Darbyshire was so encumbered with his knowledge of Virgin’s UCS seat, said to be made in accordance with the invention, that he was ill-placed to give evidence about whether it was obvious without knowledge of the invention. If this were a criticism at all, it is one that could be levelled at Virgin’s expert as well. In a case where the commercial embodiment of an invention is well known, all the experts will have to attempt to exclude that knowledge from their analysis. Whether Mr Darbyshire in fact used hindsight in his reasoning is something which is best considered in dealing with obviousness, rather than at the stage of evaluating him as an expert witness.

Mr Meade made other criticisms but it is sufficient for me to say that I did not consider any of them to be valid criticisms of the manner in which Mr Darbyshire gave his oral evidence.

Virgin’s expert in relation to the experiments was Mr Mark Harris. Mr Harris is a consultant mechanical engineer specialising in electromechanical and computerdriven products. Contour’s expert in relation to experiments was Mr Peter Hansen, also a mechanical engineer. Mr Hansen has extensive experience in mechanical testing of materials and components. No serious criticism could be made of the way in which either gave his evidence.

Air Canada called Professor David Newland, who is Emeritus Professor of Engineering at Cambridge University. He is, as Virgin accept, a very eminent and distinguished engineer. He took the task with which he was entrusted extremely seriously, making real efforts to understand the detailed functioning of the reclining mechanism in the Solar Eclipse chair accused of infringement. Professor Newland did not claim any experience in relation to aircraft seat design. However, the issues of mechanical engineering on which he was giving evidence did not, in the main, require such expertise. Virgin complain that Professor Newland had more opportunity to inspect the seats than was afforded to their own expert, that Professor Newland had had a discussion with Contour’s engineers, and that he had conducted some preliminary experiments with feeler gauges which did not form part of the Notice of Experiments. I am satisfied that none of this is properly to be regarded as a criticism of Professor Newland.

Although the evidence of foreign law received no detailed attention at the trial, I record that both sides adduced evidence from foreign lawyers of the German, Dutch and French law relating to infringement by supplying a kit of parts.

Finally Virgin adduced evidence from Mr Christopher Davies of Frank B Dehn and the defendants from Ms Judith Caldwell of Keltie LLP. Both are patent attorneys with experience of EPO practice relevant to the non-designation point, which was largely uncontroversial. Only Mr Davies was cross-examined. Virgin wanted his evidence for the fact that some argument about construction had been put to the Technical Board of Appeal, whilst Contour sought to cross-examine him, until I called a halt, about that and some other irrelevant matters. Neither side emerged with credit from this manoeuvring, but this did not reflect on Mr Davies in any way.

Before about 1996 narrow and wide bodied long haul passenger aircraft typically adopted a seating system for first and business class consisting of a number of rows of side by side seats, facing forward, on an axis parallel to the centre line of the aircraft. The seats were referred to at trial as “big recliners”, and would typically have a seat pan, back rest and head rest. The backrest would recline into the space to the rear of the seat, but not form a flat bed.

In 1996 BA introduced a design of seat which gave first class passengers the option of lying flat. A BA First seat is shown below, taken from Figure 8 of Mr Darbyshire’s report:

These seats were arranged, in the commercial embodiment, in an outward- facing herringbone arrangement. In this context “inward-” or “outward-” facing is with reference to the aisle, so these seats faced away from the aisle and towards the window.

The BA First seat had a rectangular headrest and was surrounded by a back-shell and privacy screen to shield the occupant from the view of other passengers. Each passenger also had a small ottoman (footstool) opposite their seat, on which a guest could sit to talk to them, and which became part of the bed once the seat had fully reclined. The reclining mechanism operated by means of a ‘trolley’ base which moved forwards along a track as the seat was reclined from a sitting position to a flat bed position.

A number of airlines followed BA First with the introduction of lie-flat seats. In 1997 United Airlines introduced a first class seat in an outward facing herringbone. In 2000 American Airlines introduced a seat which could swivel to an outward facing herringbone, but which faced forward for take off and landing.

Others did not adopt the herringbone. So Swiss Air and Singapore Airlines adopted a seat with a sufficiently long pitch to allow for reclining into a lie-flat position. Virgin Atlantic adopted an angled lie-flat seat, the J2000, for its Upper Class. The seats were arranged in rows facing forwards and although they reclined significantly to provide a bed, they did not go completely flat. A picture of the J2000 is shown below:

When reclining the seat into bed mode, the passenger’s feet were moved into the space beneath the seat in front of them (which reduced the pitch required to allow the seat to recline to that extent).

In 2000 BA launched its Yin-Yang seat unit, so called because it arranged pairs of seats in a head to toe arrangement:

Yin-Yang was the first lie-flat bed for business class. It won significant acclaim but was also the subject of criticism. For example, some passengers did not like facing backwards.

In the Contour Action, Lewison J identified the skilled addressee of the 908 patent in the following terms. There was no dispute about it in this action, so I set out what he said here:

“191.

… The team is a team of aircraft and transportation designers comprising designers and engineers with a knowledge or aircraft seats in general, and in particular reclining aircraft seats. The team members would include:

i)

Members skilled in designing aircraft layouts and fitting seats and units into such layouts;

ii)

Designers and engineers with at least first degree level and actual experience;

iii)

Some members with experience in the regulatory side of aircraft seating layouts and ancillary items such as crash testing and stresses, the transfer of loads from seats to the aircraft structure, and decompression requirements.” The specification of 908

The disclosure of the unamended 908 specification has been summarised both in the judgment of Lewison J [2009] EWHC 26 (Pat) at [201] to [217] and in that of the Court of Appeal [2009] EWCA Civ 1062 at [31] to [43]. Some of this is directed to the flip-over point which is no longer in issue. I summarise the description below, so far as necessary for this action.

At paragraph [0004] the patent describes what was referred to at trial as the “big recliner” in the conventional arrangement of aircraft seats. The relevant disadvantage is increased pitch, making it uneconomic for business class. At [0006] the patent refers to the BA First application. It points out that the seating units in BA First are oriented at an acute angle to the longitudinal axis of the aircraft fuselage:

The arrangement in BA First is said to have the advantage that:

“by incorporating an additional, secondary seat in the flat sleeping surface together with back-rest, seating portion and leg-rest of the primary seat, it is possible to form a long sleeping surface which is able to accommodate comfortably passengers having a height of greater than 6 ft (1.83m).”

BA First is said to suffer from disadvantages, the relevant ones being that it requires more cabin space than a conventional layout of seats and the seat itself occupies a very large floor area and is therefore unsuitable for use in business class.

The specification then goes on to consider Yin-Yang. This seating arrangement is also said to be extravagant in the space available in business class and the bed is too short to accommodate tall passengers comfortably.

The patent next sets out the objects of the invention, which can be summarised, so far as relevant as follows:

At [0017] the patent contains a paragraph which has its origins in the consistory clause of unamended 908. Now the function of the consistory clause is performed by simply stating in the first sentence that the invention is as set forth in the independent claims. The paragraph goes on as follows:

“There is provided a passenger seating system for an aircraft, comprising a plurality of seat units, each seat unit defining a notional longitudinal seat axis and comprising a supporting structure adapted for attaching the seat unit to a floor of an aircraft and means forming or being configurable for forming a seat comprising a seat-pan and a back-rest, said seat units being arranged to form a column defining a notional longitudinal column axis, in which column said seat-units are arranged sideby-side in longitudinally offset relation at an acute angle to the notional column axis, thereby defining a space to the rear of each seat, each seat unit further comprising means forming or being configurable for forming a substantially flat bed, so that when the seat unit is formed into a bed a major proportion of the bed is disposed forwardly of the position that was occupied by the seat (Seating system of the type disclosed e.g. in [BA First]). The flat-bed extends rearwardly into said space behind the seat. The invention also provides a seat unit for such a passenger seating system." (emphasis supplied)

Here, as elsewhere in the specification, the patentee makes clear distinctions between the system (i.e. the overall arrangement), the seat unit (i.e. the totality of parts associated with one passenger’s seat) and the seat itself (i.e. the seat pan and the backrest). It is clear that this paragraph does not describe the totality of the invention, not least because at least two limitations of claim 1 (the generally triangular support element and the limitation of the acute angle) are not mentioned. The paragraph conveys the broader notion of a space behind the seat into which the flat bed extends. As yet there is nothing about how the space is defined beyond the fact that it is behind the acutely angled seat. As to the support element, the next paragraph, [0018] states:

“Each seat unit includes a first passenger supporting element in said space to the rear of the seat, which first passenger supporting element forms part of said flat bed when the seat unit is formed as a bed.”

I think the skilled person would understand this paragraph, in context, as teaching him that in the space behind the seat, and not forming part of the seat itself, is a further element. When the seat unit is formed as a bed, this extra element, not part of the seat, becomes part of the bed.

The specification goes on to say in [0019] that the space to the rear of the seat is “generally triangular or trapezoidal”, and that the acute angle is in the range of 30-60^o. It further explains that the seats may be in a “herringbone arrangement”.

At [0020] there is the first mention of an “extension surface”:

“Said seat units may be disposed adjacent a side wall of the vehicle and face inwardly. Preferably, said accommodation cabin comprises two opposing side walls, and a column of seat units may be positioned contiguously or closely adjacent to each wall such that each seat faces into the cabin, with an extension surface behind the back-rest of the seat disposed adjacent the wall. The seats may thus have their backs to the vehicle wall giving the cabin as a whole an uncrowded appearance.” (emphasis supplied)

It is this arrangement, of seats facing into the cabin with their backs to the aircraft wall, with an extension surface behind the back rest (in the form of the passenger supporting element), which the specification describes in more detail thereafter. This again would convey to the reader that the extension surface/passenger supporting element is something separate from the seat.

The paragraph on which attention in the present case focused most intensely is [0025] which states:

“Each seat unit comprises a first, preferably fixed, passengersupporting element in said space to the rear of the seat, which first passenger-supporting element is disposed substantially coplanarly with said one or more movable elements when said movable elements are configured in the bed mode and is adapted to form part of said flat bed. Said first passengersupporting element is generally triangular. It will be appreciated that the first passenger-supporting element is only used by a passenger when the seat unit is arranged in the bed configuration, and accordingly the seat unit may be arranged such that the first passenger-supporting element extends into a lateral recess defined by the concave cabin sidewall to maximise the use of space in the cabin. The first passengersupporting element may be fixed in said space to the rear of the seat.” (emphasis supplied)

I think the emphasised words confirm the impression that the skilled reader would have obtained from the passages I have referred to earlier in the specification, that the passenger support (or supporting) element, although part of the seat unit, does not form part of the seat.

Virgin advance a different interpretation of this paragraph. They start by drawing attention to the fact that the supporting element is only “preferably fixed”. Thus, it is to be inferred that it may be moveable, and does not have to be present in the space behind the seat all the time. The sentence beginning “It will be appreciated…” simply refers back to the fixed alternative.

I cannot accept Virgin’s case on paragraph [0025]. I think that [0025] teaches very clearly that the passenger supporting element is not used when the seat is in the seat mode. The skilled person would understand that this is because, as he or she has learned from the earlier passages I have quoted, the passenger supporting element is not part of the seat in seat mode. It is something to be found “in the space to the rear of the seat” [0018]; it is an “extension surface behind the backrest of the seat” [0020]. The skilled person would understand the “preferably fixed” language as allowing for the supporting element to be something other than fixed (although quite what is not disclosed), but not as disclosing or allowing its use by the passenger when the seat is in seat mode.

Paragraph [0029] reinforces this meaning:

“Advantageously, the seat unit may be oriented at an angle of between 35 and 55^o, preferably 40 to 50^o,relative to the longitudinal axis of an aircraft cabin such that an extension surface behind the back-rest element extends into a recess defined by a typical concave aircraft cabin interior wall. Whilst the area of the cabin juxtaposed the concave cabin wall is not suitable, and has insufficient headroom, to accommodate the back-rest element in the upright position, it can be used in accordance with the present invention to accommodate the rear extension surface which forms part of the bed surface in the bed configuration.”

Again, the only candidate for the “rear extension surface” is the passenger support element. This is “behind the back-rest element” and must therefore be separate from it.

The remainder of the specification is taken up with a description of the specific embodiment. The layout of the seating units is shown most clearly in figure 1A:

The figure shows three adjacent seat units forming part of a column. Each seat unit has a front end (51), and rear end (52) and two opposing sides (53) and (54). At the front end there is an ottoman (65). Along one side of each seat unit is an upstanding

Figure 2 shows a perspective view of the system, with one seat in bed mode and two seats in seat mode:

(76) denotes a removable infill element which is inserted between the rear surface

The specification explains that:

“As best seen in FIG. 1A, the space 36 to the rear of the seat 71, 72 of each seat unit 40 is thus occupied by the first surface 47 of one seat and the second surface 48 of the other adjacent seat, said first and second surfaces 47, 48 of the one and other seat units 40 respectively being divided from one another by the privacy screen 60 of the one seat unit 40. The space 36 behind each seat 71,72 is thus used to extend the length of the bed surface 47, 48, 67, 74, 76 provided by the seat unit 40 in the bed configuration rearwardly of the seat 71, 72 into said space 36.”

When the seat is in the upright position the space behind the seat is concealed by the back rest and can, as the specification points out at [0050], be used for storage, supported by the surface (47).

Overall the specification points out at [0049] that the bed surface provided can be at least 80 inches (2.032 metres) with a pitch between adjacent seats of 50 to 60 inches (1.27 to 1.52 metres).

Claim 1 is directed to the column of seats adjacent the sidewall of the aircraft, whereas claim 2 is directed to back to back columns which may go into the centre of the cabin. It is enough to set out claim 1, with added reference numbers, but without the feature reference numerals in the claims:

In Kirin Amgen v TKT [2005] RPC 9 the House of Lords explained that the determination of the extent of protection only involves asking what a person skilled in the art would have understood the patentee to have used the language of the claim to mean. Guidelines to assist the court in construing the patent are summarised by the Court of Appeal in Virgin Atlantic v Premium Aircraft [2009] EWCA Civ 1062; [2010] FSR 10 at paragraph 5. The approach is termed “purposive construction” because it has regard to the inventor’s purpose. The guidelines in Virgin point out that “purpose is not the be all and end all. One is still at the end of the day concerned with the meaning of the language used.”

The first issue of construction concerns feature 21: “a generally triangular passenger support element is disposed in said rearward space substantially coplanarly with said one or more movable elements when said movable elements are configured in the bed mode”. A surprising number of issues need to be considered in relation to this phrase. These are:

On the first of these issues, the defendants submit that:

Virgin submit that:

On this issue I prefer the defendants’ construction. Both the specification and claims make it clear that the passenger support element is something different from the moveable passenger bearing elements.

As to Virgin’s argument in (a) above, I accept that the moveable passenger bearing elements do not have to form part of both the seat and the bed: the seat pan (71) is a moveable passenger bearing element, but is lowered beneath the bed when the seat unit is in bed mode. But it does not follow that the passenger support element and the moveable passenger bearing element can be the same physical thing. As to (b), and as I have indicated, I think the skilled person would understand from [0025], and the passages which precede it, that the passenger support element was something not used in seat mode, and that that applied whether the passenger support element was fixed or moveable. I appreciate that it is the claim which I have to interpret and that I should not elevate [0025] to the status of claim language: but [0025] provides very strong support for the proposition that the ordinary meaning of the claim, which specifies different elements using different language, is the correct one. As to (c), I bear in mind again that the claim does not include the words “extension surface”, but those words reinforce the view that the passenger support element is something separate

I turn therefore to the second aspect of the meaning of this feature, namely whether the passenger support element needs to be in the rearward space all the time, and in particular when the seat is in seat mode. Virgin’s submissions are:

The defendants’ case is:

I prefer the defendants’ construction. Taking first the teaching of the specification, Virgin are seeking to build far too much on the fact that the specification says at one point that the passenger support element is “preferably fixed”. These words do not give the reader any idea of what the less preferred or “unfixed” embodiment would be like. They may simply be emphasising that the element is fixed to the aircraft floor. It is certainly not the case that the skilled reader is presented with a fixed and a moveable option. There is no specific teaching that the element is moveable, far less any teaching that it could be moved outside the space behind the seat.

The final sentence of [0025] might in my judgment suggest that the passenger support element might be fixed outside the rearward space: but Virgin do not rely on that sentence as conveying any additional teaching, (no doubt because it was added in the course of prosecution). Accordingly the skilled person approaches the claims with the very clear idea that the separate passenger support element is a part of the seat unit which is to be found in the space behind the seat, and is preferably fixed. He or she would understand the temporal requirement in the claim as relating to the requirement for coplanarity, in accordance with the teaching at [0025].

The arguments about purpose are, to my mind, not very compelling in either direction. It is, however, incorrect to say that the defendants’ construction would strike the skilled person as devoid of purpose. Of course, had the other pointers been more favourable to Virgin, one might have concluded that the storage advantage was only applicable to the fixed, and not the moveable, option. The defendants’ argument about the absence of teaching of how to make the headrest work as the passenger support element does not form part of my reasoning.

I consider separately the consequences for added matter if I am wrong about construction.

Next I consider the “substantially coplanarly” requirement. The defendants submit:

Virgin submit as follows:

In my judgment, a passenger support element which is underneath the moveable passenger bearing elements cannot be described as substantially coplanar with them. The purpose of the requirement for coplanarity is to allow for a substantially flat bed.

Finally I turn to the nature of the support element. The defendants submit that the element must provide direct support, in that the passenger must rest on the element itself. In my judgment, given my conclusion that the passenger support element is separate from the moveable passenger bearing elements, and coplanar in the sense I have described, it is obvious that it provides direct support to the passenger. It may, of course, include cushioning, in which case the cushioning will form part of the passenger support element.

The passenger support element must be adapted to form part of the bed. Virgin submit that this feature is satisfied by anything which forms part of the bed. The patent uses the term “bed surface” when it means the surface. The defendants submit, to the contrary, that the passenger support element is adapted to form part of the bed surface.

Both sides accept that this point is related to the coplanarity requirement. In my judgment the passenger support element is required to form part of the bed surface, not merely some part of its supporting structure or mechanism.

I deal with a number of points under this heading:

The first of these points was decided by the Court of Appeal in its judgment on the Delta summary judgment application in relation to 908.

As I have indicated, Arnold J had held that, on its true construction, claim 1 of 908 was limited to a “plurality of seat units assembled and arranged on an aircraft”. There was accordingly no infringement unless there was infringement under section 60(1)(a) or by virtue of a kit of parts doctrine: see [2010] EWHC 3094 (Pat) at [84].

In the Court of Appeal, Jacob LJ[2011] EWCA Civ 162 described the first (and in the event the only) issue which they had to decide in the following terms at [8]:

“Do Contour's acts within the UK fall within the scope of the amended Patent, and more specifically does the Patent claim cover a ship-set before it is installed on the aircraft?”

The Court of Appeal went on to answer that question in the affirmative. Contour accept that, in the absence of a relevant amendment to the claim from that which was considered by the Court of Appeal, I should not adopt a different construction for the purposes of this case. They submit, however, that Virgin must amend the claim from that considered by the Court of Appeal, because part of the language which was relied upon by the Court of Appeal constitutes added matter. It is therefore necessary, as a preliminary matter, to decide whether there is any substance in the added matter attack.

The attack focuses on feature 12: “wherein at least some of the seat units are arranged to be disposed adjacent a sidewall of the aircraft”. These italicised words were not present in the corresponding claim in the application, claim 47, which simply says that the seat units are disposed adjacent a side wall.

In the Court of Appeal these italicised words were the focus of one of Virgin’s arguments. This argument is recorded in the judgment of Jacob LJ in the following terms at [29]:

“As for [this feature], Mr Meade submitted it supported Virgin's construction, not Delta's. This was by reason of the words arranged to be disposed adjacent a sidewall. If the claim required that the units were in fact so disposed, you would merely say that. Arranged to would be completely redundant words on Delta's construction. But they must have been intended to add something: that in context could only be a capability of being so disposed. ”

Accordingly, it is argued before me that the words did in fact add the notion that the units were not actually disposed adjacent the sidewall.

Virgin submit that there is in fact no added matter. The skilled person reading claims 44 and 47 in the application would see that it was “for an aircraft”. Accordingly this already brings to mind the notion that the system may not be on an aircraft. The use of the word “disposed” as opposed to “arranged to be disposed” in claim 47 would not detract from that disclosure.

I prefer Virgin’s submissions on this question. I do not think that the addition of the words “arranged to be” has altered the technical disclosure. Virgin offered to amend the claim in various ways if I held there was added matter. There is no need for me to consider those amendments.

Accordingly, there is also no need for me to accept Contour’s invitation to revisit the judgment of the Court of Appeal on this “for/on an aircraft” point.

I turn therefore to the second point: what is required for an off-aircraft system?

Contour submit that the word “system” conveys no more than the word “apparatus”. That is what Arnold J held at first instance in the Delta summary judgment application at [58]. It must convey more than the expression “seat unit” which is also used in the claims. They submit that the critical features of a seating system according to claim 1 are that it comprises:

Contour also rely on what the Court of Appeal said at [34] about what amounted to a system:

“…The claim is for a system for an aircraft. The skilled reader would expect the language to be confined to a realistic complete system for an aircraft.”

Accordingly Contour submit that there is no infringement unless the seat units are actually arranged so that they form a realistic complete column, defining a notional longitudinal column axis and that the seat units are arranged side-by- side in longitudinally offset relation at an acute angle. This must be done in the UK, otherwise there is no relevant act of infringement.

I reject Contour’s submissions. The skilled reader would understand that a system of the kind described in the 908 patent will only need to be fully assembled on board an aircraft. Given that I have agreed with the Court of Appeal that the patent envisages a system for, rather than on, an aircraft, I do not see why the patent would be read as requiring any particular degree of assembly or arrangement of a set of seat units off the aircraft. The columns of seats are not going to be placed on board the aircraft in their assembled state: it would in most cases be physically impossible to do so.

I do not think that the observations in the Court of Appeal’s judgment at [34] assist Contour. They do not deal with the question of whether the realistic complete system needs to be assembled off the aircraft. There is no issue in the present case that Contour made purpose-built complete systems, accompanied by detailed instructions, albeit not completely assembled.

I have come to this conclusion by a process of purposive construction of claim 1 of 908, against the background of the particular product to which it extends. It will not be in many cases where a claim will be construed to cover anything less than the entire assembled product.

It follows that I do not have to consider the further submissions based on a freestanding kit of parts doctrine. I would prefer to leave the issue of whether there is any general doctrine of kit of parts for a case where it arises directly. Just as the English courts have eschewed a free-standing doctrine of equivalents, expanding the claims outside the scope afforded by their true purposive construction, I would be reluctant to take the first step towards any other free standing doctrine of infringement. It is, after all, perfectly possible to formulate a claim expressly as a kit of parts.

There is a further issue about the construction of this feature, which it is more logical to consider in the context of obviousness over BA First.

The Solar Eclipse seat is described in a Product Description served by Contour and supported by the other defendants. It includes a side view of the Solar Eclipse seat in the lie-flat position.

In plan view, but not to scale, also in lie-flat position, it looks like this:

The headrest is to the right in both drawings. There is no dispute that it is generally triangular:

The seat is at a slight angle (3^o) to the floor of the aircraft to compensate for the inclination of the aircraft when at cruising altitude.

Beneath the headrest is the rear console. Unlike the bed this is parallel to the floor of the aircraft, and therefore at a small angle to the bed surface. This has a complex three dimensional shape, illustrated thus:

The rear console houses cabling and power for the actuators and lighting equipment in some seat units and also houses the power for the in-flight entertainment. The part marked Part 1 is forward of the passenger when seated and houses the life vest or is used for stowage. Part 2 is aft of the passenger when seated. Part 3 comprises elements A and B. Element B is a recessed portion. Part 4 is a decompression grille.

Virgin submit that the requirement for a triangular passenger support element is met by the headrest. In my judgment the headrest does not meet the requirement of the claim. Firstly, it is one of the moveable passenger bearing elements. Secondly, it is not disposed in the rearward space when the seat is in seat mode. These conclusions follow from the construction of the claims which I have adopted. Virgin’s secondary case - rear console

In the alternative Virgin submit that the requirement for a triangular passenger support element is met by the rear console. They identify the part of the rear console relied on as that marked in yellow in the figure which I have reproduced above.

In my judgment the argument fails at the first hurdle, as the rear console surface is not substantially coplanar with the passenger support elements. The surface of the rear console is beneath the moveable passenger bearing elements.

Both sides adduced experimental evidence in relation to the extent of contact between the headrest and backrest and the rear console. I will deal with these very shortly having regard to my earlier finding. Virgin’s experiments were withdrawn at the beginning of the trial when it emerged that the pressure transducers used for their measurements had not been calibrated properly or at all, and that they were not prepared to call the experimenter. That leaves the Product Description and the defendants’ experiments.

The defendants’ experiments were conducted on two Air Berlin seats which I had the opportunity of inspecting. Virgin had many criticisms over the way in which information about the construction of the seats and their testing had been supplied to them. I have to do the best with the evidence which has been adduced, whilst making no assumptions in Contour’s favour. In the end a fairly clear picture of the relationship between the underside of the headrest/backrest and the top of the rear console emerged. My findings are these:

Although the evidence on the point is weak, I conclude that the wall does satisfy the requirement of the claim for support. However, there is in my judgment no triangular support element. Nothing other than the wall can provide any meaningful support. The fact that the wall adjoins a generally triangular area which plays no support role does not turn the wall into a triangular support element. The console argument therefore fails.

The abuse of process argument arises in the following way. Claim 3 of unamended 908 (the B1 specification as it was called) calls for:

“A seating system as claimed in claim 1 or claim 2 wherein each seat unit includes a first passenger supporting element in said space to the rear of the seat, which first passenger supporting element forms part of said flat bed when the seat unit is formed as a bed”

In the Contour Action this claim was originally alleged to be infringed. However, pursuant to permission granted by Kitchin J, as he then was, in July 2008 Virgin reamended its particulars of claim to abandon its allegation of infringement of this claim. A similar argument is based on the fact that Virgin acknowledged noninfringement of claim 1 of GB 2 296 103 (“GB 103”) which includes claim 1 which concludes with the words

“including a first passenger supporting element in said space to the rear of the seat, which first passenger supporting element forms part of said flat bed.”

The abuse of process argument is only run by Contour and Delta, not by Air Canada. It follows that even if I decided the point in favour of Contour and Delta I would have to come to an independent view in the case of Air Canada. Accordingly I have had to decide the point on the merits without reference to Virgin’s previous conduct. As I have come to the conclusion that the Solar Eclipse does not infringe the amended claims, there is no purpose in extending this already long judgment by a consideration of the principles of Henderson v Henderson abuse of process and their application to the facts of this case.

Figure 12 of the Product Description illustrates a variant Solar Eclipse headrest which it says has been offered for sale since April 2011:

Virgin no longer assert this variant headrest infringes claim 1 on the “headrest” basis because it is not generally triangular, but did maintain that it would infringe on the “rear console” basis. It follows from my findings in relation to the original headrest, that an arrangement with this variant headrest does not infringe.

Infringement of 711 is alleged on a quia timet basis against Contour only. Claim 1 of 711 differs from claim 1 of 908 in that it is a claim to a seat unit, not a seating system for an aircraft. There are other differences as well but they do not matter.

The issues of infringement surrounding the passenger support system are the same as for 908. There is thus no infringement of 711.

Infringement of 734 is again alleged on a quia timet basis against Contour only. The significance of the claims of 734 is that they do not have the passenger support element. Claim 1 is to a seating system for an aircraft. Claim 10 is to a seat unit.

It follows from my earlier findings that claims 1 and 10 of 734 would be infringed by the sale of the Solar Eclipse seat.

I mentioned that there is a fourth action concerning a modified Solar Eclipse (not to be confused with the variant headrest). It includes a back shell, similar to BA First.

Contour attack the validity of 908 on the grounds of added matter and obviousness. Air Canada and Delta also run added matter attacks, but largely as squeezes on infringement.

In Vector Corp v Glatt Air Techniques Ltd [2007] EWCA Civ 805, [2008] RPC 10, Jacob LJ approved his own earlier statement (as Jacob J) in Richardson-Vicks' Patent [1995] RPC 568 at 576 where he summarised the rule against added matter in a single sentence:

I would add that it is clear that added subject matter may exist by express addition of words, or by implication from a change of wording: see per Kitchin J (as he then was) in European Central Bank v Document Security Systems [2007] EWHC 600 at [100]. “at least some”

Contour say that the addition of the words “at least some” in feature 12 of claim 1 (and the corresponding feature of claim 2) teaches that some only of the seats of a column need to be adjacent the side wall. Virgin suggest that “at least some” takes account of the seats in the centre of the aircraft which are back to back and not adjacent the sidewall. The point is a pretty trivial one, but I think in context Contour are right. Virgin do not suggest that there is any corresponding teaching in the application. It follows that there is added matter. Virgin have offered to delete the words if I came to that conclusion. They should do so.

It is convenient to mention that a similar point arises on 734, which falls to be dealt with in the same way.

The defendants submit that, on Virgin’s construction of this feature, there would be added matter. That is because the temporal requirement would now for the first time qualify the disposition of the passenger support element and not merely the coplanarity. On this construction the claim would require that “a … passenger support element is disposed in said rearward space … when said moveable elements are configured in the bed mode.” There was never any temporal element attached to the disposition of the passenger support element in the application.

Virgin submit that this objection is only about what the claim covers, not what subject matter is disclosed. They submit that there is no teaching that the passenger support element need not be in the rearward space at other times, such as when the seat is in seat mode. The claim, on this construction, covers such an arrangement, but does not teach it.

I think there is a subtle addition of subject matter if Virgin are right on construction. Virgin are correct in principle that there is no addition of subject matter merely because something is now covered by the claim which was not covered before. However, by attaching the temporal requirement to the disposition of the passenger support element, the specification is now teaching by implication that that disposition is not required at other times. On Virgin’s construction the disposition requirement is to be regarded as packaged together with the coplanarity requirement. The skilled person would understand that the coplanarity requirement is only achievable when the seat is in bed mode, and will definitely not be there when the seat is in seat mode. Accordingly there is an inference by association that the disposition need only be as specified at those times as well.

Virgin do not suggest there is any teaching in the application that the passenger support element need not be in the space when the seat is in bed mode. Accordingly if I had acceded to Virgin’s construction, the patent would be bad for added matter.

A similar but unpersuasive point was run on the meaning of coplanar. It was said that if I adopted (which I have not) Virgin’s meaning of coplanar so as to include the possibility that the passenger support element could be coplanar with the moveable elements above it, this would be to add matter. However there was no material change in wording here between the application and the granted patent. There is therefore nothing in this point. Other less significant points were taken, for example on linguistic changes from “supporting” to “support” and “flat bed” to “bed”, but I was not persuaded by them.

Contour do not attack the inventiveness of 908 if they succeed, as they have, on the issue of infringement. They recognise that an additional passenger support element, positioned in the rearward space, was an inventive improvement over the common general knowledge or any of the cited prior art. I therefore have to consider the issue of validity only on the basis that I am wrong on the issues of construction which led me to hold that the Solar Eclipse did not infringe. On this basis the patent extends to seats where there is no separate passenger support element, and the requirement for a passenger support element may be met, as Virgin contended, by a headrest.

Contour advanced two distinct arguments of obviousness. The first is based on the decision of the Technical Board of Appeal in the EPO in Agrevo. The second is a more conventional obviousness attack. Both are largely founded upon the disclosure of BA First which I now describe.

BA First is UK Patent Application No. 2 326 824A, published January 6^th 1999. It discloses both inward and outward facing herringbone arrangements. The layout of seats in the inward facing arrangement can be seen in Figure 4:

The generally triangular space behind the seat back is designated by the numeral (21). Figures 14 and 16 in BA First show the seat from a side view, in seat and bed mode respectively. I reproduce these figures below, in the annotated form used by Lewison J in his judgment in the Contour Action.

As Lewison J explained:

“The vertical line [in Figure 14] is an addition to show where the rear of the seat back is when the seat is in seat mode. Rear of that line is space within the shell that surrounds the seat. Figure 16 in BA First illustrates the seat when reclined into bed mode. It will be seen that approximately half the head rest is now disposed rearward of the vertical line which marked the rear of the seat back when in seat mode.”

It is convenient to deal here with an issue of construction of the 908 patent relevant to the case based on BA First. The relevant claim features are 12 to 14:

“wherein at least some of the seat units are arranged to be disposed adjacent a sidewall of the aircraft and face inwardly thereby to define between the rear of each seat and the sidewall a space when the seat unit is configured as a seat”

Lewison J dealt with this issue of construction at [241] of his judgment by accepting Virgin’s construction as follows:

“Virgin Atlantic argue that the space to which this integer refers is not a space inside the seat; but is limited to a space that is formed as a result of the arrangement of seats in an inward facing herringbone. In effect, therefore the space in question is the equivalent space to that designated by reference numeral (21) in the BA First patent. Virgin Atlantic's interpretation is, in my judgment, supported by the following considerations:

i)

The use of the word "thereby" indicates that the space in question is defined as a result of the arrangement of seats in an inward facing herringbone. The space between the rear of the seat back and the shell in BA First exists however the seat is placed within the aircraft;

ii)

The space in question is identified by the reference numeral (36) which identifies the triangular or trapezoidal space between the back of the seat and the side wall of the aircraft;

iii)

The description of the BA First patent in paragraph [0006] of the specification identifies the same space in BA First as being used for a counter top or a cupboard; and it is that space that reappears in the consistory paragraph [0017] in the description of the BA First patent.”

In the Court of Appeal (Jacob and Patten LJJ, Kitchin J) Contour challenged this construction. They dealt with the issue in this way:

Mr Purvis boldly invites me to take a different view from Lewison J, a patent judge of very considerable experience, and all three of the distinguished patent judges in the Court of Appeal. He says that the claim has been amended since their judgments, and the matter needs to be approached afresh. Virgin acknowledge that the patent has been amended, but point out that it has been narrowed by the amendment, leaving the relevant feature in place. So the amendments can hardly have removed a distinction between BA and the patent.

It is trite law that a patent should not be construed by reference to the prior art, unless it is prior art which the patent acknowledges. In Beloit Technologies Inc. v Valmet Paper Machinery Inc. [1995] RPC 705 at 720 line 25, Jacob J, as he then was, said this:

“I believe Article 69 of the EPC does not legitimately allow our courts to construe claims using the prior art either to widen them or to narrow them. There is normally no reason to suppose the patentee when he set the limits of his monopoly knew of a particular piece of prior art which is therefore irrelevant in deciding what those limits are. Of course the position is different if the prior art is specifically acknowledged in the patent. The purposive construction would lead to a construction of a claim which did not cover that acknowledged prior art: it can hardly have been the inventor’s purpose to cover that which he expressly recognises was old.”

One can see that this principle is firmly in play in the judgment of the Court of Appeal. The feature in question was the only thing which stood in the way of a finding of anticipation. The fact that the skilled person is faced with a conundrum, that the patentee cannot have intended to cover BA First, means that “only if no other possible construction is possible would the skilled man be forced to conclude that the patentee had claimed that which he knew was old.”

Mr Purvis submits that it is no longer necessary to go to any particular lengths to construe this feature of the claim, as the skilled person would now see, even on Virgin’s construction, that the claim includes a requirement for a triangular headrest not present in BA First. The task for me is therefore to re-evaluate this issue of construction without reliance on the Beloit exception to the general rule that the prior art is irrelevant.

Whilst paying tribute to the attractive and sustained way in which the argument was presented by Mr Purvis both in oral argument and in writing, I am not prepared to depart from the construction arrived at by Lewison J and the Court of Appeal. Whilst the Beloit principle was part of their reasoning, it was not the totality of it. I consider that what remains of the reasoning in the judgments of Lewison J and the Court of Appeal adequately supports the conclusion they came to. This feature remains a distinction over BA First.

I appreciate that in coming to this conclusion I must, by implication be differing from the view of the Technical Board of Appeal, who acceded to the argument that unamended 908 was anticipated by BA First. However, it is not the practice of the Boards of Appeal to give reasons for arriving at their interpretation of claims. I am not therefore able to analyse their reasoning in order to see whether I am persuaded by it. I also appreciate that the consequence of respecting the Court of Appeal’s construction is that Virgin did not need to amend to avoid the effect of BA First. This has caused me to ponder long and hard before reaching the conclusion that the amendments which Virgin made have the effect of excluding the Solar Eclipse. In the end I have not felt able to follow their construction, applying conventional principles. It would be wrong to adopt a different approach to construction simply because of the history of the way in which the feature entered the claim.

Contour argue as follows. First, they remind me that in Dr Reddy’s Laboratories v Eli Lilly [2009] EWCA Civ 1362, Jacob LJ, having reviewed the EPO jurisprudence on obviousness including Agrevo T 0939/92 said at [50]:

“The EPO jurisprudence is founded firmly around a fundamental question: has a patentee made a novel non-obvious technical advance and provided sufficient justification for it to be credible?”

Secondly, they point out two other propositions which can be extracted from the Agrevo case, and were not in dispute:

Next Contour say that the only distinction between BA First and the patent in suit is the triangular shape of the headrest. This, they say, is the result of accepting Virgin’s construction which would enable it to win on infringement, and Contour’s construction of the rearward space. The triangular shape, say Contour, is entirely arbitrary.

I cannot accept these submissions. Firstly, it does not get past first base, as I have rejected Contour’s construction of the rearward space. Secondly I cannot accept that a triangular headrest is an arbitrary feature. In the context of the rest of the claim, the generally triangular headrest allows advantage to be taken of the triangular rearward space. This remains true, despite the fact that it is possible to devise arrangements where full advantage is not taken of the triangular shape, such as where the seat is spaced from the aircraft wall.

This aspect of Contour’s case accepts the construction adopted by the Court of Appeal of the rearward space. It involves abandoning the rearward shell of the BA seat, moving the position of the privacy screen to give more space to recline into and reshaping the headrest.

In Conor v Angiotech [2008] UKHL 49; [2008] RPC 28 at [42] Lord Hoffmann approved the following statement by Kitchin J in Generics (UK) Ltd v H Lundbeck A/S [2007] RPC 32 at [72]:

“The question of obviousness must be considered on the facts of each case. The court must consider the weight to be attached to any particular factor in the light of all the relevant circumstances. These may include such matters as the motive to find a solution to the problem the patent addresses, the number and extent of the possible avenues of research, the effort involved in pursuing them and the expectation of success.”

It is convenient to address the question of obviousness by using the structured approach explained by the Court of Appeal in Pozzoli v BDMO [2007] EWCA Civ 588; [2007] FSR 37. This involves the following steps:

I have identified the person skilled in the art above. The common general knowledge would include the prior art commercial seating arrangements which I have described. The skilled person would have known of the possibility of inward facing herringbones, although would not be particularly motivated to adopt them. He or she would also have known that it was possible to arrange seats in an aircraft at an angle of greater than 18^o, but that was not something which had yet been done on a commercial passenger aircraft, even in the commercial embodiment of BA First. It would be known that this would require compliance with requirements for airbags and harnesses, and a more extensive testing programme. In the Contour Action the Court of Appeal identifies the inventive concept as the “space packing idea”. In the present case, on the hypothesis on which I have to adopt for this part of the case, it involves space packing with a triangular headrest.

In order to arrive at the inventive concept from the disclosure of BA First, the skilled person would have to adopt an angle of greater than 18^o and an inward facing herringbone. These are not differences from the disclosure of BA First, although, so far as the angle is concerned, it involves taking literally the angle shown in the figures. There is no doubt that BA First does not show a triangular headrest, and on the construction adopted by the Court of Appeal (and by me) the seat does not move into the relevant space. The Court of Appeal said this about obviousness in the Contour Action:

I have to resolve the issue of obviousness in the present action based on the evidence which I heard, not that which Lewison J heard.

Mr Moreno frankly accepted that the precise choice of shape of headrest was a matter of design. Although in BA he said he would have preferred a rectangular headrest because it was bigger and could be adjusted, he said you could have circular headrests as well. In some situations, such as in Concorde, an upright seat might have to be curved into a rough semicircle or triangular shape.

Mr Moreno also gave evidence to the effect that, if an inward facing herringbone was adopted, one would only need that part of the shell or privacy screen which separated the seat in bed mode from the adjacent seat. The back part of the shell was not strictly required. However this was not Mr Higgins’ view, from his knowledge of the BA seat, as he believed that the back shell formed a part of the reclining mechanism, by forming an interference fit, and could not therefore be so easily dispensed with.

Mr Moreno also accepted that one could “play around” with the privacy screen between the two seats. He did not accept that if one’s herringbone was at less than 18^o you would have enough space for a dining table or accessories. It would not really be useful to do that unless one went to an angle of greater than 18^o, although one would gain a little space. He described what was involved as experimenting.

This stepwise analysis does not really undermine Mr Moreno’s overall view that the invention was not obvious. Moreover Virgin attacked the reasoning at every stage:

In the end, like Lewison J and the Court of Appeal in the Contour Action, I was unpersuaded by the evidence in this action that the invention was an obvious step to take in the light of BA First. It is a case where it is legitimate to ask why such a simple step, if obvious, was not taken by anyone else before the priority date. The obviousness attack therefore fails on this wider construction as well.

I did not detect any suggestion that Contour could succeed on 711 if they did not succeed on 908.

A European patent shall, subject to the provisions of paragraph 2, confer on its proprietor from the date of publication of the mention of its grant, in each Contracting State in respect of which it is granted, the same rights as would be conferred by a national patent granted in that State….

Any infringement of a European patent shall be dealt with by national law.

The request for the grant of a European patent shall contain the designation of the Contracting State or States in which protection for the invention is desired.

The designation of a contracting state shall be subject to the payment of the designation fee. The designation fees shall be paid within six months of the date on which the European Patent Bulletin mentions the publication of the European search report.

The designation of a Contracting State may be withdrawn at any time up to the grant of the European patent. Withdrawal of the designation of all the Contracting States shall be deemed to be a withdrawal of the European patent application.

Designation fees shall not be refunded.