England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE CHANCERY DIVISION
(PATENTS COURT)
MR JUSTICE LADDIE
HC 02 00440

		Royal Courts of Justice Strand, London, WC2A 2LL
		1 April 2004

Lord Justice Jacob :

1. The appellants (whom I will collectively call Coflexip, though there have been name changes as recorded in the full title page to this judgment) are the patentees and exclusive licensee of European Patent UK No. 0478,742 whose priority date is 30^th March 1990. The patent relates to a method and device for the laying of flexible pipes offshore. The key claims for the purposes of this appeal are claim 1 (the widest method claim) and claim 3 (the widest device claim).
2. Unusually, the patent has been litigated before. Its claims have already been construed in some respects (see para. 40 below). And its inventive concept has likewise been determined before (see para.117 below). Coflexip sued a company called Stolt. At first instance Laddie J (by "the first judgment" of 22^nd January 1999) held that claims 1 and 2 (the method claims) were valid and had been infringed. In a supplementary judgment (also 22^nd January 1999) he construed claim 3 narrowly, holding that it would be invalid if it had a wide construction.
3. The Court of Appeal upheld Laddie J's first judgment as to the validity of the method claims. But his supplementary judgment was reversed, the Court of Appeal favouring the wide construction and holding that claim 3 was valid nonetheless.
4. In this case the proceedings before Laddie J below were effectively an action for infringement and a counterclaim for revocation, though in fact hostilities began with the claim for revocation. Laddie J held that the respondents' (whom I will call Rockwater) device did not fall within claim 3 and that the patent was invalid for want of novelty ("anticipation") and obviousness. There was no assertion of infringement of the process claims. Both sides were content to accept such points of construction as were determined by this Court in those earlier proceedings.
5. Before us Mr Andrew Waugh QC and Dr Justin Turner represented Coflexip. Mr Simon Thorley QC and Mr Thomas Mitcheson represented Rockwater.

The skilled man

6. The "man skilled in the art" is invoked at many critical points of patent law. The claims of a patent must be understood as if read by that notional man – in the hackneyed but convenient phrase the "court must don the mantle of the skilled man." Likewise many questions of validity (obviousness, and sufficiency for instance) depend upon trying to view matters as he would see them. He indeed has statutory recognition – Arts. 56, 83 and 100 of the EPC expressly refer to "the person skilled in the art."
7. It is settled that this man, if real, would be very boring – a nerd. Lord Reid put it this way in Technograph v Mills & Rockley [1972] RPC 346 at p.355

"… the hypothetical addressee is a skilled technician who is well acquainted with workshop technique and who has carefully read the relevant literature. He is supposed to have an unlimited capacity to assimilate the contents of, it may be, scores of specifications but to be incapable of scintilla of invention. When dealing with obviousness, unlike novelty, it is permissible to make a "mosaic" out of the relevant documents, but it must be a mosaic which can be put together by an unimaginative man with no inventive capacity."

8. The no-mosaic rule makes him also very forgetful. He reads all the prior art, but unless it forms part of his background technical knowledge, having read (or learnt about) one piece of prior art, he forgets it before reading the next unless it can form an uninventive mosaic or there is a sufficient cross-reference that it is justified to read the documents as one.
9. He does, on the other hand, have a very good background technical knowledge – the so-called common general knowledge. Our courts have long set a standard for this which is set out in the oft-quoted passage from General Tire v Firestone Tire & Rubber [1972] RPC 457 at 482 which in turn approves what was said by Luxmoore J in British Acoustic Films 53 RPC 221 at 250. For brevity I do not quote this in full – Luxmoore J's happy phrase "common stock of knowledge" conveys the flavour of what this notional man knows. Other countries within the European Patent Convention apply, so far as I understand matters, essentially the same standard.
10. The man can, in appropriate cases, be a team – an assembly of nerds of different basic skills, all unimaginative. But the skilled man is not a complete android, for it is also settled that he will share the common prejudices or conservatism which prevail in the art concerned.
11. None of the above is controversial. However, sometimes the requirement that the skilled man be uninventive is used by counsel for a patentee in an attempt to downgrade or dismiss the evidence of an expert called to say that a patent is obvious – "my witness is more nerdlike than his" is the general theme. I do not find this a helpful approach. It is frequently invoked and Mr Waugh QC invoked it in this case in an effort to downgrade Rockwater's expert evidence on obviousness given by Professor Witz. Mr Waugh said his witness, Mr Nash was more appropriately qualified than Professor Witz, that the latter, because he had patents in his name "was of an inventive turn of mind."
12. I must explain why I think the attempt to approximate real people to the notional man is not helpful. It is to do with the function of expert witnesses in patent actions. Their primary function is to educate the court in the technology – they come as teachers, as makers of the mantle for the court to don. For that purpose it does not matter whether they do or do not approximate to the skilled man. What matters is how good they are at explaining things.
13. But it also is permissible for an expert witness to opine on an "ultimate question" which is not one of law. I so held in Routestone v Minories Finance [1997] BCC 180 and see s.3 of the Civil Evidence Act 1972. As regards obviousness of a patent Sir Donald Nicholls V-C giving the judgment of the Court of Appeal in Mölnlycke v Proctor & Gamble [1994] RPC 49 at p. 113 was explicit on the point:

"In applying the statutory criterion [i.e. as to whether an alleged inventive step was obvious] and making these findings [i.e. as to obviousness] the court will almost invariably require the assistance of expert evidence. The primary evidence will be that of properly qualified expert witnesses who will say whether or not in their opinions the relevant step would have been obvious to a skilled man having regard to the state of the art."

14. But just because the opinion is admissible:

"it by no means follows that the court must follow it. On its own (unless uncontested) it would be "a mere bit of empty rhetoric" Wigmore, Evidence (Chadbourn rev) para. 1920. What really matters in most cases is the reasons given for the opinion. As a practical matter a well-constructed expert's report containing opinion evidence sets out the opinion and the reasons for it. If the reasons stand up the opinion does, if not, not. A rule of evidence which excludes this opinion evidence serves no practical purpose. What happens if the evidence is regarded as inadmissible is that experts' reports simply try to creep up to the opinion without openly giving it. They insinuate rather than explicate" (Minories at p. 188)

15. Because the expert's conclusion (e.g. obvious or not), as such, although admissible, is of little value it does not really matter what the actual attributes of the real expert witness are. What matters are the reasons for his or her opinion. And those reasons do not depend on how closely the expert approximates to the skilled man.

General background – common general knowledge in this case

16. The current Patents Court Guide requires that:

"The parties should endeavour to produce a composite document setting forth the matters alleged to form part of the common general knowledge and, where they disagree, what that disagreement is."

17. That was not done in this case, though I think the requirement had not appeared in the Guide by the time of the trial before Laddie J. Fortunately it does not matter because, in the end, I do not think there was any dispute as to what was and what was not common general knowledge. Fortunately also, there was no dispute between the parties that Laddie J's helpful exposition of the general background is not contested. I lift it with gratitude:

"5. The patent is concerned with a process and equipment for laying pipes and other conduits in deep water. It has particular application in off shore oil and gas fields. A variety of pipes and cables, some of very great length, may have to be used, for example, to connect the well to a site on dry land. The patent in suit is concerned particularly with the laying of flexible pipes and cables from a pipelaying ship and overcoming the problems associated with passing the pipe or cable from the ship into the water. It is convenient to start by considering the properties and behaviour of submarine pipes.

6. There are different types of submarine pipes. Although they have many features in common and the same laws of physics applies to them all, in the art they are divided into two broad classes, namely rigid pipes and flexible pipes. The former are normally made of steel. Sometimes they are coated in concrete or plastics materials. They are capable of being laid in very deep water. The latter are normally made up of a number of layers of composites and reinforcing materials such as steel braids. Because their walls are made up of a number of interacting layers, those walls tend to be very thick. Most members of the general public would regard them as stiff or rigid. However, in the technical world of submarine pipes, they are considered to be flexible. As a practical matter those in the art have little difficulty in distinguishing one type of pipe from the other. There are difficulties associated with laying both, particularly in deep water. The two broad classes of pipes cannot be distinguished on the basis of size alone. Each can be made in a variety of sizes and their size ranges overlap. However, as Mr Nash, Coflexip's expert, explains, because of their thick walls, flexible pipes are much heavier than the same sized rigid pipes.

7. The differences between typical rigid and flexible pipes can be explained as follows. A copper pipe used in a domestic water system can be regarded as rigid. If one end is clamped and the pipe is not too long, it will maintain an almost horizontal position without support at its free end. Its reluctance to bend can be referred to as its bending stiffness. If the end of the pipe is pulled down a bit it will flex. It is a bit "springy". When the force is removed, it will return to its original position. However, if more force is applied, a point will arrive at which parts of the pipe's surface will begin to stretch. It can be made to take up a permanently bent configuration. A plumber will do this by applying force to a copper pipe in a pipe bender. It is said to have been subject to plastic deformation. These characteristics were explained by Mr. Nash in his first report as follows:

"15. Rigid pipe has a finite bending stiffness. That is to say, if a sufficient load is applied to a length of rigid pipe it will deflect (i.e. bend). A rigid pipe can be bent elastically up to its elastic limit. This means that provided the pipe is not bent beyond its elastic limit (or yield stress), it will return to its original shape after the bending force has been removed. If a rigid pipe is bent beyond its elastic limit, this will result in the pipe being plastically deformed. This means that if the bending force is removed, the pipe will not return to its original profile, i.e. a permanent curvature will be induced into the pipe.

16. A plastically deformed pipe can, within limits, be returned to its original profile. This is achieved by applying a sufficient bending force to the pipe in the opposite direction to that which originally caused the plastic deformation. In other words, if a straight section of pipe has been plastically bent one has to overstress the pipe (i.e. apply a stress greater than the yield stress) in the opposite direction in order to straighten it. This process, if carried out correctly, will not affect the pressure containing properties of the pipe, nor its resistance to hydrostatic forces. However, if a rigid pipe is plastically bent beyond a certain minimum radius of curvature, called its ultimate bending radius, the pipe will suffer permanent localised buckling or crimping. This will irreparably damage the integrity of the pipe."

8. These characteristics are exhibited by rigid steel submarine pipes. As long as they are not subjected to too much force they will be springy. If more force is applied they can be plastically deformed so as to take on a permanently bent shape. If they are bent beyond their ultimate bending radius, they buckle. It may take quite a lot of force to plastically deform a rigid pipe. If such a pipe is plastically deformed so that it adopts a curved shape, it will need to be plastically deformed again to return it to its straight configuration. Depending on the size and composition of the pipe, this also may involve the use of a lot of force.

9. The behaviour of a typical flexible pipe is somewhat different. The bending stiffness of flexible pipe is far less than that of rigid pipe. The concepts involved can be explained sufficiently accurately for the purpose of this action by reference to the behaviour of a garden hose. If it is held horizontally at one end it will tend to droop. It has a much smaller ultimate bending radius than a rigid pipe, so it will have to be bent much more acutely before it buckles. On the other hand it is less readily plastically deformed. If bending force is removed from a flexible pipe or hose which has not buckled it will tend to return to its original shape. It is less prone than rigid pipes to acquiring a permanent bent shape. However if a garden hose is fed over a hose reel and a weight is put on the free end, the latter will tend to pull the hose down onto the reel. The greater the weight, the more pronounced this effect will be. Because the walls of the hose are comparatively flexible this has the effect of squashing the hose onto the reel. The cavity in the centre of the hose will become oval and, at some point, it will close completely. When this happens the pipe is said to be crushed. Again these characteristics were explained by Mr. Nash in his first report:

"18. The bending stiffness of flexible pipe is several orders of magnitude less than that of rigid pipe. This is due to the lower bending stiffness of the materials which make up the flexible pipe and also because each layer is, to a degree, able to move relative to its neighbouring layer (in non-bonded flexible pipe). This means that flexible pipe can be bent to a much smaller radius of curvature than rigid pipe without exceeding its elastic limit. If one were to bend a flexible pipe beyond its elastic limit (also referred to as damaging bend radius or minimum bend radius), the pipe would be irreparably damaged so that it could not be straightened back to its original shape and its pressure containing properties would probably be impaired. However, flexible pipe can be bent to a very much smaller radius of curvature than rigid pipe without reaching its damaging bend radius. There is no plastic deformation with flexible pipe."

10. Rockwater's expert, Professor Witz, explains that the distinction between rigid and flexible pipes is not sharply defined. He says that when bending is applied to rigid and flexible pipe, there is much in common ground in their resistance and behaviour. Both flexible and rigid pipe exhibit finite bending stiffness. He disagrees with Mr Nash's statement that there is no plastic deformation with flexible pipe. If a flexible pipe is subject to severe bending (i.e. it is subject to a "sharp" as opposed to a "gentle" bend), the innermost steel carcass may seize and plastically deform. But he accepts that the limiting bending criteria for rigid pipe are reached at a bending radius higher than the corresponding bend radius for an equivalent flexible pipe and that flexible pipe will have different parameters to rigid pipe. He points out that buckling and crushing can occur in both rigid and flexible pipes. However, because the walls of a rigid pipe are, by definition, comparatively rigid, they withstand crushing better. Thus rigid pipes are less sensitive to crushing and more sensitive to plastic deformation and buckling than their flexible equivalents. As was not in dispute, before laying a new pipe, load calculations should always be carried out to ensure that the pipe can be laid successfully with the apparatus available.

Laying rigid submarine pipes

11. When rigid submarine pipes were first laid in water from ships, they were fed off the back of the ship in a more or less horizontal direction. The ship is loaded with numerous lengths of straight pipe which are welded one by one on board to the end of the pipe being fed into the water. As more and more pipe is fed off the back of the ship, the weight of the unsupported pipe grows. This pulls down on the end of the pipe causing it to curve. This curvature of the pipe at the end near the laying vessel is called the overbend. To avoid permanent deformation here, the pipe is supported by a long curved guide, called a stinger. The stinger maintains a set minimum radius of curvature so that there is no plastic deformation of the pipe. With very rigid pipes (e.g. pipes of large diameter) being laid in very deep water, the stinger might have to be very long indeed. For example, a stinger of about 130 metres might be needed to lay rigid pipe in water 1000 metres deep. In such a case, the tension created by the 1000 metres of suspended pipe could be in the region of 60 metric tonnes. For that reason, where rigid pipe is to be laid in deep water, the stinger could extend well beyond the stern of the pipe-laying vessel. This could compromise the vessel's stability, particularly in rough weather.

12. Furthermore in all cases steps have to be taken to ensure that the pipe does not contact the end of the stinger (or is carefully controlled on contact). The reason for this is that if the vessel suddenly backed up or moves forward too slowly relative to the pipe pay-out speed, there is a risk that the pipe will be bent more sharply (i.e. be subjected to a smaller bend radius) at the end of the stinger, thereby risking damage to the pipe. It may buckle at that point. Similarly, in adverse weather, the vessel may pitch and roll. If the pipe is in contact with the very end of the stinger and the stern of the ship rises as the prow falls, there will be a tendency to bend the pipe at an acute angle (i.e. a small bend radius) at the end of the stinger where it is no longer supported.

13. The greater the depth of the water and the bigger the pipe, the greater the weight of the suspended pipe. Furthermore the vessel will be pulling in the direction of lay. The weight of suspended pipe and the tension imparted by relative movement of the vessel away from the pipe lying on the seabed will tend to pull the pipe towards the sea bed and off the ship. In some cases, the weight of the suspended pipe can be reduced by adding floatation devices to it as it is paid out. Indeed, pipe is normally laid in a sealed condition in which it is full of air since this will give it some buoyancy. Notwithstanding the adoption of these procedures, there will continue to be a substantial tension in the pipe tending to pull it off the vessel. This has to be prevented by means of some device on the vessel which pulls in the opposite direction. This can be done by a mechanical "hand" which pulls on the end of the pipe as it leaves the ship and stops it going overboard. The mechanical hand is called a tensioner. It must have the capacity to match the tension trying to pull the pipe off the vessel. Using more technical terminology, the tensioner "reacts" the tension in the pipe. In other words it acts as a brake on the pipe. Whichever terminology is used, the concept is the same; the suspended pipe under the influence of gravity pulls towards the seabed. To prevent the pipe moving in that direction in an uncontrolled manner, an equal and opposite pull has to be exerted, for example by the tensioner. The pipe can be likened to the rope being pulled by two teams in a tug-of-war. A great deal of energy may be expended by each team, i.e. each is applying a lot of tension in opposite directions, even though the rope does not move at all or only moves slowly. It can be said that the tension created by one team is being reacted by the other and vice versa. If either team lets go, the other will fall backwards. Similarly, if the pipe under the vessel is severed near the surface, the seaboard side will fall to the seabed and the vessel-side pipe will be brought back sharply on board by the effect of the tensioners.

14. Thus the tensioner must have the capacity to exert a tension on the pipe which matches (i.e. reacts) the tension generated by the suspended pipe as it is being laid. In fact, it may need to have a higher capacity than that. As mentioned above, when the pipe is being laid it will normally be full of air. This gives it buoyancy. However the tensioner may be used to recover the pipe from the seabed in which event the latter may be full of water. The buoyancy will have disappeared. The pipe may now have an effective weight considerably greater than it had when being laid. The tensioner may be made up of a number of smaller tensioners in sequence capable of exerting a total pull which exceeds that of the suspended weight of pipe. Pipelaying in this fashion can be illustrated as follows:

Figure 1:



15. In this Figure, the pipe-laying ship is being driven towards the right. If it were not, the end of the pipe where it leaves the stinger would be pulled hard down by the weight of the pipe below it. It would therefore be pulled down and might hit the end of the stinger. This could cause high localised forces which tend to buckle the pipe. This horizontal thrust of the ship results in the pipe lifting off the stinger before it reaches the end of it. Because the horizontal thrust lifts the pipe off the stinger (in other words reduces its tendency to fall down vertically), this can be used to reduce the overall length of the stinger to some extent. However increased horizontal thrust means greater use of the vessel's engines and this costs money. The location at which the pipe lifts off the surface of the stinger to enter the water is called the pipe suspension or lift off point. It will be appreciated that the location of this point will vary from time to time throughout the laying operation. The lighter the pipe being suspended and the greater the forward thrust of the vessel's engines, the "flatter" the take off trajectory of the pipe. This means that the pipe suspension point moves up the stinger towards the vessel. On the other hand heavier pipe (for example pipe being laid in deeper water) and lower engine thrust will result in a steeper take off trajectory of the pipe. In such circumstances the pipe suspension point moves down the stinger towards its seaboard end. Changes of the pipe suspension point will also be caused by pitching of the vessel due to adverse weather conditions. Professor Witz explained during his oral testimony that the stinger will be sized so the lift off point is broadly in the middle and the installer has some degree of latitude in adjusting the horizontal tension and, therefore, the effective lift off point to keep the pipe within its lay configuration.

16. Figure 1 above also illustrates how the rigid pipe flexes. It describes an "S" shape in the water. As a result, this type of rigid pipelaying is called "S-lay". The deeper the water, the nearer the vertical the central section of the pipe will be as illustrated below.

Figure 2



17. Because in deep water the pipe will tend to take up the shape shown in Figure 2, an alternative type of rigid pipelaying can be used in such situations. The angle at which the pipe is fed off the end of the ship is matched, as nearly as possible, to the natural angle which the pipe will take up in the water (its so-called catenary shape). The result is that the pipe is fed off the end of the ship at an angle to the surface of the sea. This is illustrated in Figure 3 below. Because the pipe takes up a generally "J" configuration, this form of pipelaying is called "J-Lay".

Figure 3



In this illustration, the pipe is loaded onto the ship in short straight lengths. Individual lengths of pipe are lifted onto the angled ramp at the stern of the vessel, held by clamps and welded onto the upstanding end of the pipe which leads down into the sea.

18. Needless to say, the deeper the water, the more near the vertical the top part of the pipe will be. There is a minor dispute between the parties as to how extensively J-Lay was practised before the priority date of the patent, but there is no dispute that it was well recognised as a possible way of laying rigid pipe and had been used.

19. In both S-Lay and J-Lay one of the objectives is to prevent the pipe being bent at too acute an angle (i.e. with a small bend radius). If this is not avoided, there is a risk of the pipe buckling. A similar problem exists at the seabed. It will be seen in all the Figures above that the pipe curves up from its horizontal position on the seabed. This is called the sag bend. Once again, the pipelaying must be conducted in a way which ensures that the bend radius at this point is not too small for the pipe being laid. This is a point which will be considered more fully in a moment.

20. Up to this point, pipelaying of rigid pipe has been described which involves welding together straight lengths of rigid pipe. Rigid pipe can be bent and spooled onto a large reel. This involves plastic deformation of the pipe. If the pipe from such a reel is laid from a pipe-laying vessel, it will need to be plastically deformed back into its substantially straight configuration. The reel is supplied with drive motors. These may be used to spool pipe onto the reel. They can also be used to apply axial pull to the pipe so as to counter or react the axial tension created by the pipe suspended under the vessel.

Laying flexible submarine pipes

21. With flexible pipe the standard form of laying used in modest depths of water is to pass it more or less horizontally from a reel (sometimes called a winch) over a curved chute (sometimes called an "overboarding gutter") or wheel (called an "overboarding wheel", "laying wheel" or "sheave") and down into the water. The catenary of such pipe from the point where it leaves the gutter or wheel to the sea bed is J-shaped although, because of the shape of the path taken by the pipe from the reel to the seabed, this is sometimes referred to as S-Lay. Once again, to prevent the weight of pipe under the ship from stripping the rest of the pipe from the reel, some form of tensioning device is used to react the tension created by the suspended pipe. This can take the form of tensioners or the wheel on which the pipe is loaded can be powered. The "Recommended Practice for Flexible Pipe" API 17B published by the American Petroleum Institute in June 1988 illustrates both of these methods and describes them as the most common. These illustrations are set out as Figures 4 and 5 below.

Figure 4



22. In 1990, powered reels were not suitable to lay flexible pipe in deep water where a tension capacity of more than typically 30 tonnes was required. Mr Coutarel, Coflexip's Product Research and Development Manager, said that at that time the tension capacity provided by the reel was around 10-30 metric tonnes. This was to be compared with linear tensioners. Mr Nash explained that in the Norske Shell Draugen field in about 1992, tensioners of 60 tonne capacity were used and Mr Coutarel explained that Coflexip used a tensioner of 125 tonne in 1991. Furthermore tensioners can be used in series. Accordingly, linear tensioners were used to provide the tension when greater than 30 tonne tension capacity was required. An arrangement using tensioners is illustrated in Figure 5 below.

Figure 5



23. The use of the laying wheel illustrated in this Figure and the chute in Figure 4 is to prevent the pipe from being crushed (i.e. flattened) by the effect of the weight of the line below the ship. The increased crushing forces produced by laying heavier pipe in deeper water can be offset by increasing the radius of the chute or wheel over which the flexible pipe is overboarded into the water. This was described by Mr Nash in his first report in a passage which he said was common general knowledge:

"Crushing of the pipe at the overboarding point

36. As the water depth increased, the tensile load in the pipe also increased due to the greater weight of suspended pipe. The increased tensile load became a particular concern where the pipe passed across the overboarding gutter (or wheel). The gutter provided the reaction to a combination of the tensile load in the pipe between the gutter and the installation reel or tensioner, and the free hanging pipe suspended from the gutter. This reaction was spread approximately uniformly over the length of the contact between the pipe and the gutter. If the radius of the gutter and, therefore, the pipe contact length, were to remain the same, the crushing force per unit length of contact increased as the water depth and pipe suspended weight increased. To avoid or minimise the increase in crushing load as the water depth increased, one had to increase the radius of the gutter, and hence the contact length. In this way the increased crushing load was compensated by an increase in the support length."

24. In the case of flexible pipes, just as with rigid pipes, there will be a sag bend at the junction between the pipe lying horizontally on the seabed and the near vertical portion leading down from the pipe-laying vessel. Again, if this sag bend is too acute, there is a risk that the pipe will be deformed at the bend. To avoid this, the vessel is driven in the pipe-laying direction so as to impart tension to the line in a horizontal direction. It will be appreciated that the same effect is achieved with rigid pipe. This is, in substance, a mirror of what happens on the stinger – as the vessel is driven forward the pipe takes a more shallow trajectory. Whether considering rigid or flexible pipe, additional tension applied in the direction of lay reduces the angle of the sag bend and thereby reduces the risk of damage to the pipe at that location. This effect is illustrated in the following figure.



Figure 6

25. The drawing on the left illustrates a case where there is no forward motion of the pipe-laying vessel. The one on the right is an exaggerated depiction of the effect of driving the vessel to the right. In the left hand drawing, the only tension in the pipe is the vertical one trying to pull the pipe towards the sea. The vessel has to counter or react this tension. In this case it is the reel to the right of the vessel which reacts that tension. In the right hand drawing, the tension in the pipe can be considered to be made up of two components. One is the vertical component generated by the effect of gravity on the suspended pipe. It is pulling the pipe towards the seabed. The second is a horizontal component which is trying to pull the pipe to the left. Both of these components have to be reacted on the pipe-laying vessel. Mr Nash explained the use of horizontal forces to avoid problems at the sagbend as follows:

"In order to prevent the pipe buckling at the sagbend a horizontal tension was applied to the pipe by tensioners situated on the deck of the vessel. In this way a minimum radius of curvature was maintained at the sagbend. This avoided the risk of the weight of the pipe tending to straighten itself vertically and creating overstress in the sagbend (caused by a small radius of curvature)." (First Report paragraph 85).

Although in this passage he was discussing the sagbend problem in relation to rigid pipes, the same principles apply to flexible conduits as well.

Overboarding accessories

26. Flexible pipes are normally made in the factory in very great lengths. Mr Coutarel gave evidence that a flexible pipe with an 8 inch internal diameter may be made in individual lengths up to 10 km and a 12 inch pipe may be made in individual lengths of up to 5 km. These lengths have to be joined together. Each length of pipe has an end fitting at each end. To extend a flexible pipe, the end fitting on one length must be connected to the end fitting on another to make what is known as an intermediate connection. Mr Coutarel explained that end fittings vary between 0.5 and 4 metres in length depending on the size of the pipe and its application. So an intermediate fitting may be from 1 to 8 metres in length. End fittings are made of steel and usually have a diameter twice the diameter of the flexible pipe to which they are fitted. They are rigid. Other types of rigid accessories which are installed on flexible pipe during production are bending stiffeners which are used on pipes called dynamic risers. Bending stiffeners are generally up to 7 metres long and are made of steel and polymer material. The diameter of a bending stiffener is normally 3-4 times the size of the diameter of the flexible pipe. There are other rigid accessories such as buoyancy modules and anodes.

27. These rigid accessories have to be overboarded. This means that they have to be fed over the gutter or wheel and fed into the water. If such an accessory is fed round a wheel or gutter, the tension in the pipe will act on the pipe/accessory junction to give very high buckling forces over a very short distance. The problem is illustrated in Figure 7 below.

Figure 7



28. This propensity for local buckling has been a serious problem where the laying of flexible pipe is concerned. A significant number of possible solutions have been proposed and quite a few of them have been put into practice. Coflexip gave evidence relating to, and produced drawings of, some of these. Mr. Coutarel explained that one way of overboarding the rigid accessories without damaging the pipe is to use a crane or the so-called A-frame which is sometimes located at the stern of the vessel. As the rigid accessory is unrolled from the reel, the laying operation is halted before the rigid accessory reaches the overboarding gutter. The crane hook is attached to the rigid accessory. In this way the tension load from the weight of the suspended pipe is transferred from the reel to the crane which then operates to lift the rigid accessory up and over the overboarding gutter. The crane then lowers the pipe and the rigid accessory attached to it below the level of the overboarding gutter. If the rigid accessory is an intermediate connection between two lengths of pipe, the tension is then taken up once again by the reel and the crane is disconnected from the rigid accessory. The laying process then continues. In this case, during the operation of the crane, the powered reel no longer pulls the pipe. The load is taken by the crane.

29. Another arrangement used and published before the priority date of the patent in suit operates as follows.

Figure 8(a)



30. This shows the flexible pipe being fed from the tensioners at the right towards the laying wheel. The drawing illustrates a case in which some 300 tonnes of pipe is suspended off the end of the vessel. The rigid accessory is approaching the upper surface of the laying wheel. At this point a lift off device which is located between the tensioner and the laying wheel and is normally not in contact with the pipe, is lifted up so as to support the underside of the pipe and push it in an upward direction. This has the effect of lifting the right hand end of the accessory. This is illustrated in Figure 8(b) below.

Figure 8(b)



31. The lift off device continues to rotate until it is in a position to allow the accessory to travel vertically past the laying wheel as shown in figure 8(c) below.

Figure 8(c)



32. Finally the lift off device is pulled back to its original position.

Figure 8(d)



33. It will be seen that although the pipe and the accessory are kept under tension at all times, the arrangement avoids there ever being an occasion on which there is any bending at the accessory/pipe junction.

34. Further methods for overboarding involve moving the accessory round the laying wheel or overboarding gutter under very little tension. One method of doing this, which was the subject of a Coflexip patent, is illustrated below.

Figure 9



35. A collar is fitted around one of the end fittings and cables (9) are attached to it in order to connect it to the winch used when a pipe has to be abandoned, e.g. because of bad weather, and recovered (hence the name "A&R" winch). The tension load is transferred from the tensioners to the A&R winch. The tensioners are then opened and the A&R winch cable is unreeled to allow the intermediate connector to pass through the tensioners. Unreeling is continued until the collar reaches a tilting frame (5). When the collar reaches the tilting frame it automatically engages with the frame. By continuing to unreel the A&R winch cable, the tilting frame is able to rotate and thereby lift the intermediate connector off the wheel. The tension applied inboard of the tilting frame, e.g. by the tensioners, is turned off or reduced. This means that there is bending but no significant tension on the vessel side and tension but no bending on the sea side of the intermediate connector. It will be appreciated that this is similar in principle to the use of the A-frame discussed in paragraph 28 above. This is illustrated in the following Figure.

Figure 10



36. Once the tilting frame has completed its rotation, the collar automatically disengages from the frame. The tension is transferred back to the tensioners and the A&R winch cable is disconnected. The tilting frame is brought back on board and normal laying is resumed.

Summary of the relevant problems in laying flexible pipe.

37. From what has been said above, it can be seen that among the problems facing those wishing to lay flexible pipe were the following. First there is a need to avoid bending the pipe at the interface between the pipe and the rigid accessory at the same time as a high axial tension is being applied at that point. If a high axial tension is applied to the pipe at the interface at the same time as it is subjected to bending stress, there is a risk that the pipe will become damaged. Second there could be a problem which arises out of the radial contact loads induced by the axial tensile load (pipelay tension) when the flexible pipe is bent around the overboarding gutter or wheel. These contact loads are generally acceptable in shallow or medium water depth, but they tend to crush the pipe in deep water. The same problem exists if high axial tension is applied to the pipe while it is spooled on its reel."

18. The last passage is of particular importance. Both sides accepted it. In my judgment it fairly represents part of the common general knowledge though it was not shown that all of the particular prior art proposals formed part of the common general knowledge. Moreover it seems clear that those problems had been long-standing at the date of the patent. Mr Thorley endeavoured to suggest they were new or fairly new as real problems in that use of flexible as opposed to rigid pipe was only becoming significant around the date of the patent and that it was only around then, when deeper water pipe laying was coming in, that they mattered. I do not think he made that out – and the prior proposals for dealing with the problems of flexible pipe show that those in the art had been tackling the problem for some time. The Judge did not make a finding in his summary that the relevant problems were new.
19. Mr Waugh pointed out that the Judge had omitted a couple of other prior proposals, proposals that he set out in his Stolt judgment at paras. 17-19 (the "Chinese finger") and 23-24. I do not set them out here. But it is to be noted that the latter proposal did contemplate using a suspended movable clamp to take vertical tension – close to but not the same in concept to the use of vertical tensioning means in the patent in suit. Mr Waugh prays it in aid as another "miss" when considering obviousness, see below.
20. Gusto (EPA 0 054 332) 1982 is a further proposal, from 1982, showing the problem of pipes exposed to tensile and bending loads simultaneously and a possible solution – a rather Heath Robinson affair fixed under the vessel. It reinforces the Judge's conclusion that there were problems to be solved.
21. There is one further matter of common general knowledge which I should note, since it is relevant to Mr Thorley's non-infringement argument. It is simply this: it would be apparent to the skilled man that, at least in millpond conditions, when a vessel was moving forward and actually laying a flexible pipe along the sea bed by a vertical lay (as in the patent), the pipe would necessarily be at a slight angle to the vertical from the point where it last made contact with any part of the vessel. The pipe could not go down completely vertically – if it did so it would coil up on the bottom. So the vertical part of the J of a J-lay must be slightly off vertical. I accept that. It is self evident. Professor Witz gave as a typical angle 2^o off vertical.

The Patent

22. This is a relatively short document (shorter than any of the judgments). It is accompanied by drawings which, by the terms of Art 69 itself (never mind the Protocol), are to be used to interpret the claims.
23. The patent opens with the following sentence:

"The invention relates to a method and a device for effecting the laying of flexible conduits, in particular of tubular flexible conduits, comprising tensioning means disposed in the vertical path of the said tubular flexible conduit."

So two things are made clear at that outset – there is a vertical tensioner and the invention is for flexible conduits. Before carrying on with the text I go straight to the drawing – which in the real world is what skilled men do. For convenience I use Laddie J's version, which omits some irrelevant detail:



24. There some important points at once apparent:

i) The flexible conduit sags as it comes off the drum 2 – showing that it is not under tension.

ii) It follows that it can go over the guide 4 without danger of deformation or crushing;

iii) After the guide it goes into the tensioning means, a pair of "linear winches", 6. It is the tensioning means which finally brings the conduit into the vertical. The means grips the conduit strongly enough to take its entire weight down to the sea bottom. The conduit "dangles" from the tensioning means in a J lay;

iv) There is a table 7 through which the dangling conduit passes;

v) The conduit passes into a shaft, 8 in the middle of the vessel, which provides an opening into the sea. Such shafts are well-known and go by the name "moonpool" shaft, the patch of sea at the bottom being called a "moonpool". The conduit passes through the moonpool shaft but if the ship rocks enough it will brush against its bottom edges.

vi) There is an unnumbered derrick to the right of the guide 4.

25. A second drawing illustrates how the apparatus is used to overboard accessories.



26. a shows the doors open. The conduit (not under tension) runs over the guide 4 into the vertical tensioning means which holds it vertical. By stage b the end of the conduit, to which is fitted an endpiece 21, has arrived at the end of the guide 4. The unnumbered derrick has a winch 19. On this is a cable 20. That cable is attached to the endpiece. It takes the weight of the conduit. So the main vertical tensioning means can be opened, c, the conduit lowered until the endpiece is at table level, d, and the doors closed so as to enable workers to get to the endpiece, e. At this point the table provides support for the suspended conduit and the cable 20 is released. A new section of conduit, with a frontpiece, can now brought over 4, turned through the opened vertical tensioners to be joined to the endpiece at the table. When that is done the vertical tensioners are closed to grip the conduit, g. Finally the doors are opened, h and the endpiece lowered into the sea.
27. I go back to the text of the patent. The next sentence says:

"It is known that the laying of tubular flexible conduits is effected from a vessel. Such a conduit may be stored in a basket with a vertical axis or on a drum with a horizontal axis. The conduit is unwound substantially horizontally from the storage means."

This acknowledges the known fact that the conduit, when it first comes off the storage means, is horizontal.

28. Then the well-known prior art is acknowledged:

"The conduit is retained by tensioning means ensuring its suspension. The tensioning means support the mechanical tension produced by the weight of the unwound conduit and thus spare the storage means the task of having to sustain this weight. After the storage means, the tubular flexible conduit passes over a deflecting element, such as a wheel or a curved chute placed for example at the rear of the vessel and ensuring the guidance of the flexible conduit and the passing over the side of the vessel, the axis of the conduit passing from a substantially horizontal direction to a steeply slanting direction, it being possible for the angle of inclination to range from a few degrees to approximately 15 to 20 degrees relative to the vertical. The conduit is submerged in the water, to be deposited on the seabed. The tensioning means, as well as the wheels ensuring the passing of the horizontal path to the vertical path therefore have to support the weight of the tubular flexible conduit comprised between the vessel and the bottom as the laying vessel advances. The tensioning means, as well as the deflecting element, must therefore withstand the pull exerted by the portion of the flexible conduit suspended between the vessel and the bottom. For the purpose of being able to lay flexible conduits with relatively large diameters in great depths of water, one has to use tensioning means and deflecting elements whose dimensions and cost pose problems and which create an excessive space requirement on the bridge of the laying vessel."

29. So, in the prior art:

i) The tensioning means is horizontal;

ii) There is a deflecting element (wheel or chute) for "ensuring guidance" of the conduit overboard;

iii) The guidance takes the conduit from horizontal to a "steeply slanting direction."

iv) This ranges from "a few degrees to approximately 15^o to 20^o from the vertical".

30. The next paragraph says in effect that in the prior art the storage means and tensioner can be combined in a winch – a sort of powered and braked big reel from which the conduit is unwound. The paragraph after says the trouble with that is that for bigger pipes and depths, you need a bigger wheel and there are practical limits to this. I here put into my own words the rather more formal language of the two paragraphs which read:

"It is also known that for effecting the laying, equipment such as a winch is used combining the functions of storage and of tensioners of the flexible conduits, which also makes it necessary to pass the flexible conduits over a deflecting element to place it in the water. Such a winch, as well as the deflecting element, assume dimensions and space requirements that are excessive when the diameter of the conduit and the depth of water increase. As the diameter of the tubular flexible conduit and the depth increase, the size of the wheel becomes increasingly larger. Such a wheel may have a diameter of the order of 10 metres for a depth substantially equal to 500 metres."

"With the devices of the known type, it is not possible to exceed this order of depth. Wheels with a larger diameter are difficult to make."

31. The patent then passes to another known problem:

"Moreover, there are problems in passing the end fittings for connecting sections of the tubular flexible conduits, or of auxiliary accessories mounted on the flexible conduits, such as collars, stiffeners, buoys, anodes, etc.; because these relatively rigid accessories, whose external transverse dimension is greater than the external diameter of the flexible conduits must pass over the deflecting element."

This is the problem illustrated in Laddie J's paragraph 30 quoted above.

32. The patent then acknowledges two pieces of specific prior art which were aimed at the two problems, one of these being the Gusto patent already mentioned.
33. Then comes a statement of the object of the invention. This is of course important if one is to construe the claims purposively. It reads:

"The object of the present invention is the laying of flexible conduits at depths that are substantially greater than those which are feasible by using the known means, as, for example, a depth of the order of 1000 to 2000 metres. The device in accordance with the present invention must be capable of withstanding considerable tractive forces which may reach and even exceed 250 tons in the case of a conduit with a diameter substantially equal to 30 cm for a depth of 1000 metres. With the exception of sections having a rigid accessory, or having an increased rigidity relative to the flexure of the flexible conduits, and whose external bulk exceeds the external diameter of the flexible conduits, the lowering of a continuous length of a tubular flexible conduit is ensured by the main tensioning means which take up the mechanical tension exerted by the portion of the flexible conduit suspended on the tensioning means between the laying vessel and the bottom. In these conditions, the tension ahead of the tensioning means being exerted on the portion of the flexible conduit up the line and coming from the storage means is very low and preferably virtually negligible. It has been found that in these conditions, it becomes possible to subject a portion of the conduit situated in the portion ahead, between the storage means and the tensioning means, to relatively extensive bending, this being explained by the fact that in this case, there is no combination of the bending with an axial tractive load. The device in accordance with the invention thus makes it possible, without resorting to bulky and sophisticated equipment, to effect changes imposed on the flexible conduits between the point where it is extracted from the storage means and the point where it plunges into the water, the axis of the flexible conduits having, for example, to pass from a horizontal direction to a direction that is close to the vertical. It is thus possible to cause the flexible conduits to pass over a guide means, such as a curved chute or a wheel with a small diameter, including the passing of rigid and bulky accessories mounted on the conduit. The tensioning means used are of the linear winch type. It has been found that winches of this type make it possible to increase the load of the axial tension exerted by the flexible conduit and which has to be taken up by the tensioning means, which makes it possible to increase for a flexible conduit of a given structure, the depth of water in which it can be laid."

34. A number of points emerge:

i) The invention is intended for deep water use. It follows that it is likely to be unfair to the patentee to read his claims as covering an apparatus which cannot be used in deep water.

ii) Save where there is an accessory passing, it is the main tensioning means which carries the weight, sea-side.

iii) On the shipside of the main tensioning means there is "very low and preferably virtually negligible" tension. It is this which enables the conduit to pass over a guide or wheel of relatively small dimension so as to "pass from a horizontal direction to a direction that is close to the vertical." There is no simultaneous tension and bending.

iv) A linear winch can do the job of taking all the tension caused by the weight of the conduit and that it is this possibility which makes it possible (for a given conduit structure) to lay deeper than would otherwise be possible.

35. The patent goes on to discuss various types of linear winches – those with caterpillars, or tyres, a stepwise gripper or hybrid. What will not do are "winches with a motorised drums, capstans with single or twin drums, or multiwheel capstans are not considered to be linear winches."
36. Next the patentee discusses the "auxiliary retaining means." Nothing turns on these. Then he comes to the "table support", describing how this has open and closed positions. Of the latter position he says it "nevertheless allows the flexible conduits to pass freely without the accessories." This is important when one comes to construction. It means the patentee has contemplated the conduit passing through a small hole in the closed doors for at least some of the laying operation.
37. There then follows detail which it is not necessary to repeat. The description at the end of that second embodiment in which two auxiliary tensioners are used contains the following passage:

"The tubular flexible conduit 3 leaves the linear winch at a considerable slant, it being possible for its angle relative to the vertical to vary according to the depth of the water and the circumstances of the laying operation. This angle can generally amount to 5 to 10 degrees, but it can also be extremely small, the conduit being virtually vertical, in particular because of a great depth of water, or yet again attain approximately 15 degrees. One of the original features of the invention lies in the absence of any means for guiding the tubular flexible conduit after it has left the main tensioning means. However, it should be noted that the main tensioning means 6 can themselves ensure guidance for the tubular flexible conduit and a deviation relative to the vertical. For example, the multicaterpillar tensioning means generally permit a deviation of more or less 10 to 15^o relative to the vertical."

38. Next are the claims. Only claims 1 and 3 really matter, though claim 9 also plays a minor part in relation to novelty:

Claim 1

"A process for laying from a floating support (1), a flexible conduit (3) comprising a rigid accessory (21, 21a, 21b) mounted on the said flexible conduit (3) and having an outer dimension larger than the outer diameter of the latter,

wherein one unrolls a flexible conduit (3) gripped at its outer surface by linear winch-type tensioning means (6) with a substantially vertical axis, and wherein

(a) to cause the section of the flexible conduit (3) whereon the accessory is mounted to pass through the said tensioning means (6), the said rigid accessory (21) is connected to auxiliary tensioning means (19, 20), so as to take up the pull exerted by the flexible conduit (3) by the said auxiliary tensioning means (19, 20), while the flexible conduit (3) is lowered through the free space between the said laterally moved-apart tensioning means (6);

(b) the outer surface of the flexible conduit (3) is gripped ahead of the rigid accessory by the said tensioning means (6) after they have been brought together, so that the pull exerted by the flexible conduit (3) is again taken over by the said tensioning means (6);

(c) after the auxiliary tensioning means (19) have been released, the unrolling of the flexible conduit (3) is resumed by the said tensioning means (6), the latter comprising the last means for guiding the conduit at the level of the floating support.

Claim 3

"A device for operating the process according to any one of claims 1 and 2, comprising:

- linear winch-type tensioning means (6) with a substantially vertical axis, capable of ensuring the normal lowering of the flexible conduit (3) by gripping the outer surface of the flexible conduit (3) and capable of being laterally moved apart, the said tensioning means (6) comprising the last means for guiding the flexible conduit (3) on board the floating support(1),

auxiliary tensioning means (19, 20) comprising at least one elongate movable traction element (20) capable of being connected to the rigid accessory (21, 21a, 21b) mounted on the flexible conduit (3)."

Claim 9

"A device according to claim 7 or 8, characterised in that it includes a supporting structure (5), such as a derrick mounted on the floating laying support (1), and allowing the tensioning means (6), the auxiliary tensioning means (19, 20), and the guidance means (4) to be supported."

39. It is an unusual claim structure, a process claim followed by a product-for-carrying-out-the-process claim. Moreover the process claim requires various items of hardware and is thus not "pure process." Nonetheless I do not think that the skilled man, to whom it is addressed, would have much difficulty in following it, guided as he will be by the drawings.
40. I begin by noting that the claims have has already been construed by this Court in Stolt. It was held that:

i) Flexible conduit "should be interpreted as including more than what was known as flexible pipe, but not so widely as to cover what was known in the industry as rigid pipe" (per Aldous LJ at para. 18).

ii) The pull referred to in the phrase so as to take up the pull exerted by the flexible conduit" means "that exerted by the flexible conduit seaside of the main tensioners" (para. 21). Aldous LJ went on to accept that "there will inevitably be some tension [shipside] due for example to weight of the pipe and friction, but that does not detract from the teaching of the specification that the tension from the pipe seaside of the tensioners is to be taken by the main tensioners, or when an accessory is passed, by the auxiliary tensioner." "Pull" has to be construed in context, just as "vertical" had to be so construed in Catnic v Hill & Smith [1982] RPC 183.

iii) A device for operating in claim 3 means "a device suitable for operating" (para. 26).

iv) Last means for guiding the conduit at the level of the floating support did not exclude bumpers or deflectors located below the tensioners to prevent the pipe which is being laid from hitting the moonpool [at its base] or the moonpool doors (paras. 29-30, endorsing Laddie J in Stolt at paras. 65-67). As I read Aldous LJ he was not concerned to give a complete meaning of this phrase, which has assumed much more importance in this case.

The first issue – does the Toisa Perseus fall within claim 3?

41. I come at last to the first point in this case. It turns on a question of construction. So I begin by reminding myself of the principles:

(a) The first, overarching principle, is that contained in Art 69 itself. Sometimes I wonder whether people spend more time on the gloss to Art 69, the Protocol, than to the Article itself, even though it is the Article which is the main governing provision.

(b) Art 69 says that the extent of protection is determined by the terms of the claims. It goes on to say that the description and drawings shall be used to interpret the claims. In short the claims are to be construed in context.

(c) It follows that the claims are to be construed purposively – the inventor's purpose being ascertained from the description and drawings.

(d) It further follows that the claims must not be construed as if they stood alone – the drawings and description only being used to resolve any ambiguity. The Protocol expressly eschews such a method of construction but to my mind that would be so without the Protocol. Purpose is vital to the construction of claims.

(e) When ascertaining the inventor's purpose, it must be remembered that he may have several purposes depending on the level of generality of his invention. Typically, for instance, an inventor may have one, generally more than one, specific embodiment as well as a generalised concept. It is the latter which matters when construing the claim, particularly the widest claim. Otherwise one is in danger of being unfair to the inventor. I put it this way in Tickner v Honda [2002] EWHC 8 (Patents) at paragraph 28:

"The whole approach goes by the sobriquet "purposive construction". You learn the inventor's purpose by understanding his technical contribution from the specification and drawings. You keep that purpose in mind when considering what the terms of the claim mean. You choose a meaning consistent with that purpose – even if that involves a meaning which, acontextually, you would not ascribe to the word or phrase. Of course in this exercise you must also be fair to the patentee – and in particular must not take too narrow a view of his purpose – it is the widest purpose consistent with his teaching which should be used for purposive construction."

(f) Nonetheless 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. Hence the other extreme of the Protocol – a mere guideline – is also ruled out by Art 69 itself. It is the terms of the claims which delineate the patentee's territory.

(g) It follows that if the patentee has included what is obviously a deliberate limitation in his claims, it must have a meaning. One cannot disregard obviously intentional elements. Hoffmann LJ put it this way in STEP v Empson [1993] RPC at 522:

"The well known principle that patent claims are given a purposive construction does not mean that an integer can be treated as struck out if it does not appear to make any difference to the inventive concept. It may have some other purpose buried in the prior art and even if this is not discernible, the patentee may have had some reason of his own for introducing it."

(h) It also follows that where a patentee has used a word or phrase which, acontextually, might have a particular meaning (narrow or wide) it does not necessarily have that meaning in context. A good example of this is the Catnic case itself – "vertical" in context did not mean "geometrically vertical", it meant "vertical enough to do the job" (of supporting the upper horizontal plate). The so-called "Protocol questions" (those formulated by Hoffmann J in Improver v Remington [1990] FSR 181 at p.189) are of particular value when considering the difference of meaning between a word or phrase out of context and that word or phrase in context. At that point the first two Protocol questions come into play. But once one focuses on the word in context, the Protocol question approach does not resolve the ultimate question – what does the word or phrase actually mean, when construed purposively? That can only be done on the language used, read in context.

(i) It further follows that there is no general "doctrine of equivalents." Any student of patent law knows that various legal systems allow for such a concept, but that none of them can agree what it is or should be. Here is not the place to set forth the myriad versions of such a doctrine. For my part I do not think that Art. 69 itself allows for such a concept – it says the extent of protection shall be determined by the terms of the claims. And so far as I can understand, the French and German versions mean the same thing. Nor can I see how the Protocol can create any such doctrine.

(j) On the other hand purposive construction can lead to the conclusion that a technically trivial or minor difference between an element of a claim and the corresponding element of the alleged infringement nonetheless falls within the meaning of the element when read purposively. This is not because there is a doctrine of equivalents: it is because that is the fair way to read the claim in context.

(k) Finally purposive construction leads one to eschew what Lord Diplock in Catnic called (at p.243):

"the kind of meticulous verbal analysis which lawyers are too often tempted by their training to indulge."

Pedantry and patents are incompatible. In Catnic the rejected "meticulous verbal analysis" was the argument that because the word "horizontal" was qualified by "substantially" whereas "vertical" was not, the latter must mean "geometrically vertical."

42. I return to apply these principles to this case. The dispute is over the meaning of last means for guiding the flexible conduit. I must first, to explain how dispute arises, describe the relevant part of the Rockwater vessel, the Toisa Perseus. Although it has often been said that the question of construction does not depend on the alleged infringement ("as if we had to construe it before the Defendant was born" per Lord Esher MR in Nobel v Anderson (1894) 11 RPC 519 at 523), questions of construction seldom arise in the abstract. That is why in most sensible discussions of the meaning of language run on the general lines "does it mean this, or that, or the other?" rather than the open-ended "what does it mean"?
43. There are some difficulties with describing the details of Toisa Perseus and particularly in what function, if any, they have. Rockwater provided a product description in accordance with CPR Part 63 Practice Direction 5.1 or its equivalent predecessor. But they never formally verified it or "put on" (as the Australians say) a witness who could be cross-examined about it. In any future case where the product or process description machinery is used, it is desirable that, unless it is accepted, it be formally proved – if the claimant requires - by a witness at trial.
44. The general structure of the Toisa Perseus is shown in the following drawings taken from the product description. The first shows the general arrangement of the ship, as in fig. 1 of the Patent:

The second drawing shows the detail around the table and moonpool:



45. The drawings shows a close similarity with those in the Patent. But there are three differences:

i) First, the tensioner is set back towards the stern rather than being over the centre of the moonpool shaft.

ii) Next the aperture in the moonpool doors is offset with respect to the hanging conduit. If it hung perfectly vertically with the doors open in millpond conditions then it would just touch the corner of the aperture instead of passing through the middle. In plan view from the top, the aperture and conduit would look like this:





It follows that when the vessel was actually pipelaying – moving to the right – the conduit would press somewhat against the stern-side of the aperture. In other words instead of coming down from the tensioning means at a slight angle (which, as I said, Professor Witz estimated as typically 2^o from vertical) when the moonpool doors were closed it would come down vertically to the aperture in those doors. The J-lay angle could only start there.

iii) The last modification is that attached to the underside of the moonpool doors is an upside-down square funnel. This is in two halves; one half being on each door so that the funnel is complete when the doors are closed. It is shown as 12 on the above drawing. With the doors shut, again in millpond conditions, the conduit may be prevented from dropping straight into the sea from the aperture – it may brush the stern side of the funnel.

46. Rockwater gave no evidence as to the purpose of these modifications. Nor did they give any direct evidence as what effect they had as a practical matter. Their case is that the offset aperture in combination with the flared horn are the last means for guiding the flexible conduit on board and not the main vertical tensioner.
47. Prof. Witz illustrated Rockwater's case with the following diagram of what happens when the conduit drops straight into the sea from the tensioner.



48. This is of course grossly exaggerated – it shows an angle of nearly 45% whereas in real life it is more likely to be of the order of 2^o. The notion here is that the "guidance" is being provided by a slight wrap-around the bottom end ("fore") of the tensioner caterpillar.
49. Rockwater develop their position by pointing out that in the Toisa Perseus all the horizontal component of force is taken by the flared horn when the ship is moving forward pipelaying to the sea bed. So much was accepted by Mr Nash in cross-examination.
50. They further seek to support their position by reference to the passage in the specification dealing with the second embodiment which says the conduit "leaves the linear winch at a considerable slant" and that the angle can "generally amount to 5 to 10^o" but can also be smaller in great depth or "yet again attain approximately 15^o." They pray in aid the remainder of the passage too – particularly the sentence saying the tensioners "can themselves ensure [perhaps "permit"] guidance for the … conduit and a deviation relative to the vertical."
51. The patentees on the other hand say that the offsetting and provision of the horn is irrelevant. They say that the heart of the invention is to use a vertical tensioning means, one strong enough to take the pull of the conduit as it falls to the seabed. Realising that can be done, coupled with the provision of an auxiliary tensioning means to overcome the overboarding problem, is what the invention is all about. The inventor's purpose is to be found in the explicit "object of the invention" set out in the patent. The "deflection" being spoken of in the claim is not some minor change of direction after the tensioner: it is apparatus which effects "changes imposed on the flexible conduit between the point where it is extracted from the storage means and the point where it plunges into the water". In other words the patent is talking about a substantial deflection – and the last substantial means of deflection is the tensioning means.
52. I accept that contention. The general idea is to effect the change of direction from horizontal to approximately vertical before the flexible conduit is put under tension. If one does that, then the conduit is not bent under significant tension. That is the inventor's purpose. The "guidance" he is referring to is the totality of the means whereby the conduit is brought to the substantially vertical. So the tensioning means is indeed the last for guiding – it provides the last significant guidance for the flexible conduit.
53. This is borne out by a consideration of the drawing. The conduit is unreeled in a tensionless condition. It goes over the chute 4. The chute is the first means for guiding. If the conduit were left to its own devices as lay over the chute it would come off at an angle – "flexible conduits" are not truly floppy as we saw from a sample in court. But the tensioning means brings it into the vertical. Thus it is the last means of guidance. I do not think the inventor would have regarded a subsequent brushing with the corner of an aperture of a moonpool door as "guidance," any more than the occasional brushing of the bottom of moonpool opening counted as guidance in the Stolt action. Both may take up horizontal components of force but these are trivial compared with the weight of the conduit.
54. It is true that in the Toisa Perseus the whole of the horizontal component of force is taken – but it is not the horizontal component of force which really matters – that is slight compared with the much greater vertical component. There is indeed some analogy with Catnic itself here – the slightly angled backplate of the infringing lintel also took the vertical component of force – the component that mattered.
55. What then of the passage in the description of the second embodiment and Professor Witz's diagram showing the angle at the end of the tensioner? Mr Thorley strongly relies on this. He says it shows that the patentee regarded the lower end of the tensioner as being a means for guidance and indeed the last means for guidance. I think there is some substance in this, though it does mean, as Mr Thorley submitted in answer to a question from Pill LJ, that one is to regard the lower end of the tension, as a "directional device". Most people would not so describe it. However, it does not follow that a particular passage describing a particular embodiment should necessarily be taken as limiting the generality of words used in the claim. What matters is the overall disclosure or purpose.
56. Here, until one comes to that passage, in my judgment a wider concept of "guidance" is being considered and it that wider concept which is to be found in the claim. I do not read the passage, as did the Judge, as indicating that "any guide means downstream of the tension[er] was an essential requirement of the invention." I read it, in the context of the disclosure as a whole as indicating that it is essential that there be no significant guide means downstream. What amounts to "significant" depends on the purpose. The purpose is to avoid the sort of guidance under tension in the prior art,
57. Consideration of the patent diagram and its discussion of the moonpool doors confirms this. The patentee clearly expects his vessel to be able to lay conduit with the moonpool doors closed so that the conduit runs through a small aperture. The skilled man would appreciate that, depending on the depth of lay and sea conditions, the conduit would brush more or less frequently against the edges of the aperture. It could indeed run against it and be "guided" (using the word out of context) somewhat by it.
58. Likewise there is the realistic probability that, depending on conditions, the conduit will also come into contact with and thus be "deflected" by the bottom edges of the moonpool shaft – the point already covered in the Stolt judgment. Incidentally one wonders here, looking at the first of the figures of the Toisa Perseus whether the conduit could touch the horn – as shown it would seem that as it was angled back it would touch the bottom of the moonpool shaft first. This was not a point that was or could be explored properly in evidence because, as I say, no relevant witness was called.
59. In short, I construe "last means for guiding" as meaning "last means for providing substantial guidance". And I do not think the offset aperture and upside down horn of the Toisa Perseus do that – the job has already been done by the vertical tensioner.
60. The Judge thought otherwise. He construed the words this way:

"It seems to me that a "means for guiding" is a part which is capable of applying that lateral force during a significant part of the operation of the pipelaying vessel within its normal operational envelope. For example a guide which would be in almost constant contact with the conduit during pipelaying in shallow water may not be in contact with it for much of the time when pipelaying in particularly deep water where the top of the catenary is practically vertical. Thus for the majority of the time in deep water the guide does not react the horizontal tension. It would nevertheless still be a guide means or means for guiding" (para. 79).

61. But this has all sorts of uncertainties – temporal and climatic considerations, size of conduit, depth and so on. None of these (save for depth) are referred to in the patent. I do not think it is correct. At its heart is a concentration on lateral forces, but no-where in the patent is there an indication that it is concerned with these as such. What matters is the overall tension in the conduit – a tension which is nearly entirely due to its weight and whose really significant component is vertical.
62. The Judge considered that:

"the invention includes the alleged benefit of having a tensioners which not only applies vertical tension to the line but is capable of reacting horizontal tension for the purpose of guiding the conduit. The benefit of all the forces being able to be reacted by the tensioner is missing" (para. 91).

63. Professor Witz offered two reasons for saying that there were problems with the arrangement. Firstly that there would be bending over a relatively short radius under high tension as shown in his exaggerated diagram causing a risk of conduit collapse. Secondly that such bending would cause wear and tear on the tensioner.
64. But there is no evidence that in reality either problem would really exist or be significant. It is Rockwater who chose to use an offset aperture and horn - so a direct explanation as to why they are there and what happens in practice, is telling in its absence. Nor do I read the patent as confining itself to a case where the relatively much smaller horizontal component of tension must be taken by the tensioner. I cannot think of any reason why the inventor would want so to confine the meaning of his claim (cf. Lord Diplock's "nutshell" question in Catnic).
65. The Judge also likened the horn to a "stinger" as used in the prior art. But stingers were not used for flexible conduits – their use is described by the Judge in his paragraph 15. Without the stinger the weight of a rigid pipe fed out horizontally would cause it to bend too sharply and thus be damaged. The horn here does not have the same function at all.
66. Accordingly I conclude that the Toisa Perseus falls within claim 3, considering it as I have when arranged with its moonpool doors closed.
67. There is an altogether other reason for holding that the claim covers the Toisa Perseus. Consider the vessel laying conduit with its doors open. Then there is no question of the conduit coming into contact with the doors or the horn. The only possible obstruction to the conduit going directly into the sea from the tensioner would be the bottom of the moonpool shaft – functioning just like the bumper in Stolt.
68. Mr Thorley had no real answer to this. He said the ship was not intended to be operated that way, that there were safety reasons why both his clients and the patentee's device had to be run normally with the doors closed. That is no answer. Given the construction of claim 3 adopted in Stolt, the vessel is capable of carrying out the process of claim 1 with its moonpool doors open. Looking at it another way, if the Toisa Perseus had been prior art, claim 3 would have been anticipated. Novelty could not have been saved by saying "but it is not meant to lay conduits with the doors open." So far as I can see the Judge did not deal with this point at all.
69. So for two reasons the Toisa Perseus falls within claim 3.

Issue 2 Anticipation (want of novelty)

The Appellate approach to novelty

70. I come to the next major issue, was a vessel falling within claim 3 disclosed in either of two prior art citations called Recalde US (US Patent No. 4,721,411) and Recalde GB (UK Patent Appn. No. 2,178,129A). There is little difference between the two and, as did the parties, I shall concentrate on Recalde US. I shall just call it Recalde. It was published in January 1988, so just less than 3 years before the priority date of the Patent in suit.
71. Before coming to the detail I begin with proper approach for the Court Appeal in relation to the issue of novelty. Mr Thorley contended that it should be the same as that to be taken in relation to a finding on the question of obviousness. That has recently been restated in several cases. In Biogen v Medeva [1997] RPC 1 at p. 45 Lord Hoffmann said when discussing the issue of obviousness:

"The need for appellate caution in reversing the judge's evaluation of the facts is based upon much more sold grounds than professional courtesy. It is because specific findings of fact, even by the most meticulous judge, are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance (as Renan said, la vérité est dans la nuance), of which time and language do not permit exact expression, but which may play an important part in the judge's overall evaluation. It would in my view be wrong to treat Benmax as authorising or requiring an appellate court to undertake a de novo evaluation of the facts in all cases in which no question of the credibility of witnesses is involved. When the application of a legal standard such negligence or obviousness involves no question of principle but is simply a matter of degree, an appellate court should be very cautious in differing from the judge's evaluation."

72. Similar expressions have been used in relation to similar issues. The principle has been applied in Pro Sieben Media v Carlton [1999] 1 WLR 605 at pp. 613-614 (per Robert Walker LJ) in the context of a decision about "fair dealing" with a copyright work; by Hoffmann LJ in Re Grayan Building Services [1995] Ch 241 at p.254 in the context of unfitness to be a company director: in Designers Guild v Russell Williams [2000] 1 WLR 2416 in the context of a substantial reproduction of a copyright work and, most recently in Buchanan v Alba Diagnostics [2004] UKHL 5 in the context of whether a particular invention was an "improvement" over an earlier one. Doubtless there are other examples of the approach.
73. It is important here to appreciate the kind of issue to which the principle applies. It was expressed this way by Lord Hoffmann in Designers Guild

"Secondly, because the decision involves the application of a not altogether precise legal standard to a combination of features of varying importance, I think that this falls within the class of case in which an appellate court should not reverse a judge's decision unless he has erred in principle."

74. I do not think the question of novelty involves the application of a "not altogether precise legal standard." It involves a precise standard. Nor is it applied to a "combination of features of varying importance." On the contrary one must look for every claim element to see whether it is fully disclosed in the prior art. It may be that in some cases, Biogen-type principles may come into the question at an earlier stage. I have in mind, for instance, where the court has to evaluate evidence as to what a particular prior use actually was and whether it was enabling or where there is an evaluation of what is exactly disclosed by something like a photograph (see van der Lely v Bamfords [1963 RPC 61), or even, perhaps, the meaning of a technical term or phrase where experts have disagreed. But in a case such as this, where the issue is simply what does the prior art describe and does it fall within the claim?, the Biogen principle does not apply.

The legal test for destroying novelty

75. There was no dispute as to this. It well-known and that set out in the judgment of this Court in General Tire v Firestone Tire [1972] RPC 457 at pp. 485-456.

""If the prior inventor's publication contains a clear description of, or clear instructions to do or make, something that would infringe the patentee's claim if carried out after the grant of the patentee's patent, the patentee's claim will have been shown to lack the necessary novelty, that is to say, it will have been anticipated. The prior inventor, however, and the patentee will have approached the same device from different starting points and may for this reason, or it may be for other reasons, have so described their devices that it cannot be immediately discerned from a reading of the language which they have respectively used that they have discovered the same device; but if carrying out the directions contained in the prior inventor's publication will inevitably result in something being made or done which, if the patentee's patent were valid, would constitute an infringement of the patentee's claim, this circumstance demonstrates that the patentee's claim has in fact been anticipated.

If, on the other hand, the prior publication contains a direction which is capable of being carried out in a manner which would infringe the patentee's claim, but would be at least as likely to be carried out in a way which would not do so, the patentee's claim will not have been anticipated, although it may fail on the ground of obviousness. To anticipate the patentee's claim the prior publication must contain clear and unmistakable directions to do what the patentee claims to have invented: Flour Oxidizing Co Ltd v Carr & Co Ltd (1908) 25 RPC 428 at 457, line 34, approved in BTH Co Ltd v Metropolitan Vickers Electrical Co Ltd (1928) 45 RPC 1 at 24, line 1. A signpost, however clear, upon the road to the patentee's invention will not suffice. The prior inventor must be clearly shown to have planted his flag at the precise destination before the patentee."

76. It is also well to remember Lord Westbury's well known test in Hill v Evans (1862) 31 LJ Ch 457: does the earlier document give "for purposes of practical utility" the same information as the later patent so far as it relates to the claim concerned.
77. And yet another way of looking at the problem is to ask whether what is disclosed falls within the claim – if it had been later would it infringe?
78. All these tests (which are different aspects of the same thing) apply here. There was no dispute as to that. The dispute was what Recalde discloses. If it discloses all the hardware capable of being used for the process of claim 1 then it is a "device for operating" the process and will anticipate.

The Recalde disclosure

79. Recalde was a proposed, but never-built, development of the actual pipelaying vessel called The Apache which was the unsuccessful piece of prior art cited in Stolt. Recalde was not cited in that case. It is an immensely detailed document, almost descending to every nut and bolt – over 1400 parts are numbered on the 47 drawings. The text, apart from the claims, is 39 columns long. Professor Witz said he took two days to read it. I think there would be few who would disagree with the view that it is also immensely boring. That, of course, would not bother the skilled man who must be taken to read it through assiduously, both in relation to the question of novelty and in relation to obviousness. He is never bored. What I will do here is simply to describe the bits necessary for this judgment, mainly in my own words rather than the near-impossible language of the original.
80. I begin with figure 1 of Recalde:



81. It would have been recognised by the skilled man as a development of the Apache, used from 1979 onwards. Although he would not have been familiar with the detail of the vessel, he would have been aware that the overboarding mechanism was for getting rigid pipe overboard without the use of a stinger. Recalde develops that concept to enable an array of lines, including a rigid pipe to be laid simultaneously.
82. The general idea is that set forth in the abstract:

"A reel pipelaying vessel having multiple reels for laying operational lines in a wide range of water depths. An operational lines laying device is mounted on the vessel in order to provide supporting means for the simultaneous layout of a plurality of operational lines at a common velocity and in a predetermined relationship to one another. Motive means are interconnected to the multiple reels in order to control the rate of layout. Separate or combined straightening and tensioning devices can be employed for handling the rigid walled pipeline contained within the operational lines. The straightening and tensioning devices can be mounted for pivotal movement with respect to the vessel deck or can be operated in fixed planes with respect to the deck. Six or more operational lines can be simultaneously laid out with vessel described."

83. One gets from this:

i) Recalde is about laying a plurality of lines all at the same time and with a common velocity;

ii) For this purpose he has multiple reels, each holding one of the lines to be laid;

iii) Those reels have motive means to control the rate of lay. This means (and is confirmed in the description) that they have both motors and brakes;

iv) There may be straightening and tensioning devices for handling rigid pipeline. (Flexible lines of course do not need straighteners.)

84. The three drums on the right of fig. 1 contain the pipes to be laid. The "main reel" 20 carries rigid pipe, drums 40 and 36 lines (which may be pipes) which may be flexible. Three drums are shown but the patent contemplates that two pipes or lines may come off a particular drum – hence 6 in all. The various lines go from their drums over the "take off" drum 26 into a pipe take off structure (30). There is a straightener 32 (needed to bend back the pipe from main reel 20 because it is a rigid pipe) and a vertical tensioning device 34.
85. Also to be noted on fig. 1 is item 62 just to the stern side of the main reel at deck level. This is an "A&R" winch. "A&R" stands for "abandonment and recovery". This is a well-known device in pipelaying. If things get too stormy you may need to abandon the operation and let the pipe down to the sea bed. You do this by lowering it with a cable from an A&R winch. The upper end of the cable can, if necessary, be fixed to a buoy or otherwise it may remain attached to the winch. Recovery consists of using the winch to bring the end of the pipe up from the sea bed. Winch 62 has an associated storage reel for the cable. This is on the other side of the main reel at 50. The cable passes under the main reel over cable rollers 58,60 to winch 62.
86. Fig. 1 does not show much detail of what there is at the stern – the pipe take off structure. All one can see is that, very generally, it consists of a straightener 32 followed by a tensioner 34. The detail is shown in fig.8:



87. Here one has a side view of the "pipe array" 35 coming off the take off drum. The array goes into the straighter marked generally 32 then into the tensioner marked generally 34. The structure has another A&R winch marked 362. The patent allocates no special function to this A&R winch any more than it does to A&R winch 62. The reader would assume they are each provided for the normal purpose of abandonment and recovery.
88. Recalde is not on its face intending to deal with the same problems as those with which the patent in suit is concerned, namely getting a flexible line vertical whilst not under tension and providing a means for overboarding accessories. It does not in terms even mention overboarding or flexible pipes. Nor does it mention that shipside of the tensioner a flexible pipe can be virtually tensionless. Of course none of that matters if in fact the Recalde vessel could be used for the process of the patent.
89. The Judge found that it could. He held that a flexible conduit could be on one of the auxiliary drums, fed over the top of the take off drum, into the tensioner from which it would drop directly into the sea. The tensioner would thus be the main tensioner of the claims. Thus, he said (para. 115), "the major items of hardware required by claims and 3 of the patent in suit are present." As for the auxiliary tensioning means, the judge found that A&R winch 62 would serve the purpose.
90. These conclusions were attacked on three grounds. It is only necessary for one of them to succeed to defeat that novelty attack, but in my judgment all three grounds are correct.

The A&R Winch point

91. As I have said the Judge found that A&R winch 62 could act as the auxiliary tensioning means of the claim. But he fell into two errors here. First he overlooked the fact that Rockwater did not advance an attack based on that winch. Secondly it was not proved that there were clear and unambiguous directions in Recalde to so position the winch that it could be so used.
92. Rockwater's opening skeleton argument before Laddie J said this:

"49. The tensioner can be laterally moved apart … so as to allow the accessory to pass. The pull exerted by the flexible conduit would have to be accommodated whilst the tensioner was apart and the cable winch 362 was suitable for doing this"

No mention was made of winch 62.

93. Moreover it was never put that winch 62, as shown in the Recalde drawings, could, without appropriate sheaving (i.e. provision of grooved wheels – what most of us would call pulley wheels - to facilitate change of direction) of the cable. Mr Nash's cross-examination went like this:

Q. There are shown in fig. 8 and A&R winch 362; correct?

A. Yes.

Q. And that is a cable winch, as I understand it.

A. I presume so.

Q. And there is also another A&R winch. You can see it in fig. 5 or fig. 4 marked at 62.

A. Yes.

Q. Both of these would by definition be capable of taking the weight of whatever pipe it was that you were laying?

A. I am not sure that 62 was in a position to do that.

Q. Right. I am not worried about position. [my emphasis] I am just talking about ability at the moment.

A. There is no indication that it can be sheaved to do so, so it is just a picture with no indication as to how you might use it.

Q. But sheaving to carry out particular tasks was a conventional task that those operating these sort of vessels had to approach?

A. Yes, but I think it would be unlikely it sheaved for both 62 and 362. If they have identified 362 here in the pipe take-off structure and this pipe take-off structure is obviously movable, I would it would only have it sheaved up for one and was the one I would expect.

94. So no case was put that 62 was shown to be in the right position to be used, without sheaving, as an auxiliary tensioning means.
95. Moreover just looking at the drawings, one certainly cannot say that there is a direct line of sight from winch 62 to the take off drum. The side view in fig. 1 shows that drum 36 may be in the way. Mr Thorley pointed out that the drum was movable on runners from port to starboard and suggested it could be moved out of the way. But that is far from certain. Other drawings show the mounting could very well get in the way. Prof. Witz did not give evidence that there was a direct route from 62 to the take off drum. Part of his evidence in response to questions from the Judge ran thus:

"If you come to the stern of 62, you can see a black line, which I believe is the cable, and that is generally routed to the stern. You see it aligns with the pipe take off drum. How it goes over is not clear on the drawing, but that would be my preferred A&R winch in general to use."

And in further cross-examination he confirmed that:

"On fig 2. we can see the line coming off the A&R winch, but then it sort of vanishes. I am afraid I cannot make out where it goes after that. Unfortunately on fig. 1 it does not show the routing there."

96. Accordingly I have formed the clear view that the Judge was wrong to hold that A&R winch 62 could be used, without some unspecified modification, as the auxiliary tensioning means of the claims of the patent in suit.
97. There is this further point about 62. Even if it did satisfy claim 1, it is not mounted on a "supporting structure such as a derrick." So it could not anticipate claim 9. The Judge overlooked this when he held this claim anticipated (para. 167).
98. As for A&R winch 362, obviously it could not be used as the auxiliary tensioner for the patented process. It is below the linear tensioner and so could not hold an accessory above it such that the tensioner could be opened. Of course if you provided a sheave to take the cable up the take-off structure, it might be possible to use it, but Recalde contains no instructions to do this. The Judge rightly held (para. 166) that 362 was not novelty-destroying of the relevant claim element.

The all-the-pull point/strong tensioner points

99. These two points are related. The main tensioning means of the patent must be capable of "taking up the pull" of the flexible conduit. In Stolt that was construed as meaning the pull "seaside of the main tensioner." The Patentees say that there are no clear and unambiguous instructions in Recalde that the tensioning means there should be capable of doing this. They say:

(a) If one looks at the drawings the various pipes coming off the three reels are all shown as taut – that they are under tension. The contrast is with the picture of fig. 1 of the patent in suit, showing the flexible conduit as floppy.

(b) Recalde is concerned to feed out all his pipes at the same rate. For that purpose his drums are all powered and braked. They can and indeed must take up tension. Indeed Professor Witz accepted that was so:

"I have always said that my reading of this was suggesting that you kept some back-tension"

(c) It follows that the tensioner at the stern of Recalde need not be strong enough to take the entire tension seaside and there is no clear instruction to have one capable of doing that job. Its function, from Recalde's point of view, is to pull the rigid pipe through the straightener. That is confirmed by its size as shown in the picture. Moreover Recalde says it is "rather short" (Col 27₂₈); yet for the patentee's purpose (which remember includes deep water) you need a long one.

100. I think those points are right. Perhaps the most obvious point of all, however, is even simpler. Recalde shows a tensioner, but he does not say that it must be strong enough to take all of the sea-side tension. Actually the tensioner in Recalde is no different from that in Apache. It has the same purpose in Recalde. And it was rejected as anticipatory in paragraphs 49 – 50 of Aldous LJ's judgment in Stolt.
101. It is suggested that even a small tensioner might be able to do the job of carrying out the process of claim 1 when the vessel is not in deep water. Then the pull would far slighter and there is no limitation in claim 1 as to depth. This is actually a further, trivial, point of construction. I do not think it consistent with the purpose of the patent to read "pull exerted by the flexible conduit" as being limited to a slight pull. The apparatus of the invention is mainly for "laying .. in deep seas (a depth exceeding 900m)." To fall within the claim the vessel must be capable of deep sea laying. This means a strong tensioner. Of course this involves a question of degree – but that is no obstacle to a fair reading of the patent.
102. So I do not think there are clear and unambiguous instructions in Recalde that his tensioner be strong enough for the invention. He does not need it to be because his drums can and are expected to take up tension.
103. The Judge held otherwise. He, and Mr Thorley on the appeal, paid particular attention to the following passage at col.24_30-45.

"The auxiliary reels 36 and 40 are fitted with hydraulic motors which are used for spooling of operational lines on to the reels. It is preferable to provide either two or four such hydraulic motors for each of the reels 36 and 40. As shown in FIGS. 22 and 23, hydraulic motors 806 and 808 are mounted on reel support frame base 724 and are provided with sprocket gears 810 as shown in FIG. 23 which are intermeshed with sprocket chain 684. Upon operation of the hydraulic motors 806 and 808 in order to rotate the reel 40 in a clockwise direction as shown in FIG. 22, operational line(s) can be reeled on the storage drum 680 of auxiliary reel 40. During the unspooling operation the hydraulic systems providing power to the two hydraulic motors 806 and 808 can be operated in order to provide breaking force for the reel 40 in order to provide additional tension for the operational lines which are being paid out over the drum 26 for layout."

104. The suggestion is that by this passage Recalde is contemplating that the lines may be slack and the consequence is that when that is so the tensioner will be taking all the seaside force. I am quite unable to accept that. The passage itself speaks of "additional tension" suggesting there is some other source of tension. That could be from friction over the reel, the inherent braking of the motors (if they remain connected) or possibly something else. If Recalde was saying here that the lines may be slack ship-side of the tensioner, then he certainly was not so saying clearly and unambiguously. This passage is far from giving "for the purposes of practical utility" the information in claims 1 or 3 of the patent.
105. And I think the same of the slightly different passage in Recalde GB. This says (p.7_29-37):

""Each of these lines can be passed over the pipe take-off drum 26 and then passed through the straightening device 32 and the tensioning device 34, even though the plastic, electrical, and support lines may not require straightening and hence are passed through without the straightening device being in operative contact in order to use the layout drum 26"

106. The Judge said:

"What this passage appears to be saying is that all lines pass through the straightening and tensioning stations but the flexible ones miss out operative contact with the straightener because they do not need straightening. The inference is that the flexible conduits are in operative contact with the tensioner. That means that the tensioner applies tension to them."

107. Again I do not think this is clear and unambiguous. The lines which do not need straightening may well pass into the tensioner but I do not see that as necessarily so. More significantly I do not see any teaching that it is the tensioner which must take the entire seaside pull.
108. In this connection Mr Thorley, rightly, points out that there can be some pull shipside provided it is not enough to crush the flexible conduit as it goes over the guide. But I do not see that being taught by Recalde. He just shows tension, has drums which can cause and take tension and is wholly indifferent to shipside tension. There is no clear teaching that the shipside tension on a flexible conduit should be low enough to prevent crushing and no clear teaching that the tension sea-side of the tensioner should be strong enough to achieve that.
109. I should perhaps mention the other aspect in which Recalde GB differs from Recalde US. Recalde US only contemplates multiple pipelaying. This requires all the lines to be fed out at the same rate and under control which will involve tension. The Judge accepted that Recalde GB did contemplate laying a single, flexible, line. He went on to say, "in doing so the tensioner reacts to some or most of the axial tension." But that is not good enough. What is needed is substantially all the pull sea side. There is negligible pull shipside in the patent – and Recalde GB does not teach that explicitly. Hunting around for implication will not do.

Ability to pass accessories

110. As I have said, Recalde does not mention passing accessories at all. And I have concluded that there is no available "auxiliary tensioning means". But there are or may be other problems too. When a multi-array is being passed, Recalde expects one of the pipes to be rigid. If you open the main tensioner to allow an accessory to pass, you will not, as the accessory passes, be able to use the tensioner for its principal purpose – to pull the rigid pipe through the straightener. There was a suggestion that only one track of the tensioner could be opened and that this was possible with the device as disclosed. Professor Witz came up with this ingenious idea. It was not mentioned for this purpose by Recalde: the mechanism by which it is said this could be done was for a wholly different purpose, namely adjusting the curvature of the straightener/tensioner. In view of my conclusions on the other points concerning anticipation it is not necessary to pursue this point further here. It has significance, however, in relation to obviousness, to which I now turn.

Issue 3 Obviousness

111. There was no dispute as to the legal approach – that which is often helpful and is helpful in this case is the structured approach of Windsurfing v Tabur Marine [1985] RPC 59 at 73-4. I summarise the steps:

(1) Identify the inventive concept of the claim;

(2) Identify the common general knowledge;

(3) Identify the difference(s) between the prior art and the alleged invention;

(4) Decide whether those differences would have been obvious steps.

112. Nor were the well-known warnings against ex post facto analysis in dispute. Fletcher-Moulton LJ's judgment in British Westinghouse v Braulik (1910) 27 RPC 209 is as true today as when it was first said:

"'I confess that I view with suspicion arguments to the effect that a new combination, bringing with it new and important consequences in the shape of practical machines, is not an invention, because, when it has once been established, it is easy to show how it might be arrived at by starting from something known, and taking a series of apparently easy steps. This ex post facto analysis of invention is unfair to the inventors, and in my opinion it is not countenanced by English Patent Law.'

113. Nor was there any dispute that normally when a Judge has made a finding of obviousness, an appellate court should be slow to interfere, only doing so when the Judge has made an error of principle – see above.
114. Here the Judge was really in a difficult position to decide obviousness. For he had already concluded that the claims were anticipated. So, looking at Windsurfing step (3), he had decided there were no differences. His views on obviousness were therefore inherently conjectural. Indeed in theory he had to make several different conjectures – depending on how many points of difference from the prior art he should assume – just the A&R winch point, just the strong-enough tensioner point, just the all-the-pull point, or any combination of them. On my conclusions as to the differences from Recalde, obviousness actually has a concrete starting point.
115. So the Biogen principle as to the appellate approach to obviousness has no real application in this case. Or, looking at it another way, the Judge made an error of principle in his assessment of step 3 Windsurfing. Nonetheless I shall consider the Judge's views after setting out my own.
116. I start with the inventive concept of claim 1. It is for a process for laying a flexible conduit involving a combination of pieces of apparatus and steps. They are:

(1) provide a vertical linear tensioner;

(2) use this for unrolling the conduit;

(3) use this as the last substantial means of guidance on the vessel;

(4) provide an auxiliary tensioner;

(5) use this for passing accessories with the main tensioner open

117. Chadwick LJ put it this way in the Stolt case at para. 78:

"I agree with Lord Justice Aldous, for the reasons which he gives, that the inventive concept in the present case comprises the following elements: (i) it relates to a process for laying flexible conduits to which rigid accessories (of a bulk greater than the diameter of the pipe itself) have been or are to be attached, (ii) it provides for the pull to be taken by a tensioner ("the main tensioner") mounted in a vertical axis on the seaward side of the chute or guide over which the pipe or conduit must pass between the basket in which it is stored on deck and its entry into the sea, (iii) it requires that the operative parts of the main tensioner can be moved apart or opened, so as to release the pipe and allow the accessories to pass through it; and (iv) it provides for the use of an auxiliary winch to take the pull, again on the seaward side of the chute or guide over which the pipe must pass, when the accessory is passed through the main tensioner. The combination of those elements solves two of the problems associated with the laying of flexible pipe. First, by taking the pull on the seaward side of the chute or guide over which the pipe must pass between storage on deck and entry into the sea, it avoids the danger of crushing associated with the passing of flexible pipe over a curved guide under tension. Second, again by taking the pull on the seaward side of the chute or guide, it avoids the danger of local buckling, the interface between the flexible pipe and the rigid accessory, associated with overboarding accessories in a flexible pipe under tension."

118. Windsurfing step 2 requires the common general knowledge to be identified. That I have done by reference to Laddie J's uncontroversial description. His paragraph 37 is of particular importance here. There was a known problem with flexible conduit – how to get it into the sea from a horizontal feed and how to overboard accessories is, at the risk of a little oversimplification, the problem in a nutshell.
119. I am satisfied that this problem was long-standing. Mr Thorley suggested that the need was only very recent – because the proportion of flexible pipe laying as compared with rigid pipe laying only increased in the 1980s and because deep water pipe laying only became a significant problem at about the same time – shallow-water wells having been exhausted. I do not agree – nor did the Judge make any such finding. The various proposals for overboarding accessories or dealing with pipe crushing (including the "Chinese finger") are well before the date of the patent – showing (hardly surprisingly) that engineers could readily foresee the need to find a way of laying in deep water and were trying to find solutions.
120. Windsurfing 3 requires the differences to be identified. On my reading of Recalde they are:

i) No disclosure of a suitable auxiliary tensioning means;

ii) No disclosure of a suitable main tensioner;

iii) No disclosure of how to overboard accessories;

iv) No disclosure of how to deal with the problems of crushing under tension of flexible pipes.

121. These differences in my view mean that the inventive concept of the patent is miles away from Recalde. I think there was an invention here.
122. I am confirmed in this view by another consideration. All the "bits and pieces" of the invention were known separately for many years. The question "why was it not done before" is always a powerful consideration when considering obviousness, particularly when all the components of a combination have been long and widely known. Sometimes there is a good answer (e.g., no demand, not worth the expense, prior art only recent). Apart from the no-demand-until-recently point, Mr Thorley only offered two other explanations. He suggested that the expense of building a vessel capable of carrying out the patented process was not warranted until shortly before the date of the patent. This fails for two reasons: firstly it is the no-demand point in another form and secondly the expense is not all that great. Implicit in the argument is that a whole vessel would have to be specially constructed. But that is not so – existing vessels could be modified at the (for the oil industry) trivial expense of £3m.
123. Mr Thorley's other explanation is that Recalde was only published shortly before the date of the patent. But I do not think the skilled man would have begun to see this overly detailed complex description of a vessel for laying an array of lines (including particularly a main rigid pipe) as disclosing something which, if modified in a self-evident manner could be used for the process of the invention. You have to have the idea of the process first.
124. Start with the A&R winch. Of course once you have the idea of an auxiliary tensioning means shipside of the tensioner you could sheave one of Recalde winches to do the job. But you have to have the idea first. And that means you must see that the tensioner could, if you made it strong enough, take all of the pull. And you must also see that then a flexible pipe could be tensionless shipside and so got from horizontal to near vertical without danger of crushing. And that you could, if somehow the Recalde straightener were not in the way, or could be made to get out of the way (Prof. Witz's idea) overboard accessories by opening the tensioner and using a sheaved A&R winch cable. This is far from the thinking of an unimaginative skilled man.

The Judge's view on obviousness

125. I turn to the Judge's view. I think he took a too analytical view of the inventive concept – overlooking that it was to a combination not a series of unrelated elements. He said, repeating what he said in Stolt:

"The inventive concept embodied in the patent is the provision of linear winch-type tensioners with substantially vertical axes so arranged that, in use, they take all or substantially all of the pull exerted by the dependent conduit. As explained above the method of passing accessories through the tensioners by opening and closing them to allow the accessories through was known. In fact it is difficult to see how accessories could be made to pass through linear tensioners without opening them and no alternative mechanism was advanced as a possibility. There is no peculiar or unexpected interaction between the method of passing the accessories and the use of the vertical tensioners. The former can be ignored for the purpose of assessing obviousness since it adds nothing to the inventive concept in the patent".

126. That is not the same concept as that adopted by the Court of Appeal in Stolt. Nor was the Judge right to say that Mr Miller accepted that formulation – he did not, as appears from his opening. The Judge's formulation of the concept overlooks the fact that there is a combination here – vertical tensioner and a means of passing accessorises.
127. That the Judge focused on the vertical tensioner alone and overlooked the importance of an auxiliary tensioner also appears from paragraphs 56 of his judgment. He said this:

"Thus the purely structural features of the claim would be met by a device of the following general arrangement:



128. That is incorrect. Unless you have the idea of an auxiliary tensioner and provide one, you cannot get the accessory overboard. Mr Thorley accepted the Judge was in error here, saying that other passages in the judgment showed he had not overlooked the auxiliary tensioner. I cannot really see that is so.
129. There are, I think, other errors in the Judge's approach to obviousness. For instance he assumes (correctly) that Recalde teaches taking some of the tension seaside. But then he says it would be obvious to operate "so as to take most of the tension". But "most of the tension" is not enough – it needs to be virtually all. And unless the skilled man sees the virtue of that (and the rest of the process of the patent) it is not obvious to go to all of the tension. Moreover I think the Judge too readily thought that the skilled man would pay significant attention to the use of Recalde's multi-reel vessel just for the purpose of laying a single flexible pipe. Recalde US never mentions that idea, Recalde GB does but very obliquely.
130. Perhaps most fundamental is that the Judge fell into error in taking the stepwise approach contained in paragraphs 173-176. It is just the sort of approach which Fletcher-Moulton LJ warned against. So I think the Judge erred in principle here.

Conclusion

131. In the result I would allow the appeal on all points. The patent is valid and infringed.

Lord Justice Mummery :

132. I agree with both judgments.

Lord Justice Pill :

133. I also agree. I wish expressly to state my agreement with two findings of Jacob LJ upon the technical evidence. I agree with his finding at paragraph 121 that there was an invention. The inventive concept was clearly described by Chadwick LJ in Stolt when the patent was last before this court, and is cited by Jacob LJ at paragraph 117.
134. Secondly, though earlier in point of legal analysis, the judge's reliance on "A&R" winch 62 in the Recalde disclosure, under the heading "anticipation", was, in my view, misplaced. A&R is shorthand for "abandonment and recovery" and the purpose of such a device is described by Jacob LJ at paragraph 85. I agree with Jacob LJ's conclusion at paragraph 96. It is not surprising that winch 62 could not be used, without substantial modifications to the apparatus, for the purpose now proposed when it was introduced for a different purpose. If anything, Jacob LJ understates the position in paragraph 95. Analysis, with the help of Counsel, of the evidence demonstrated the obstacles to using winch 62 for the purpose now proposed. Indeed, no serious attempt had been made on behalf of the respondents to establish its suitability.
135. As to the "man skilled in the art", he is described by Jacob LJ as a 'nerd' (paragraphs 7 and 11) and as "not a complete android" (paragraph 10), which suggests that he is part of the way to being an android. A 'nerd' is defined in the Concise Oxford Dictionary (10^th Edition 1999) as "a person who lacks social skills or is boringly studious" and an 'android', in the same work, as "(in science fiction) a robot with a human appearance". I hope that those working in this field will not regard "men skilled in the art" as figures from science fiction who lack social skills. Jacob LJ, will think me less than supportive of the development of the language of the law but I do respectfully prefer, for its clarity, Lord Reid's terminology cited at paragraph 7 of the judgment.