Procter & Gamble Company v Reckitt Benckiser (UK) Ltd

This is an appeal from a judgment of Lewison J, [2006] EWHC 3154 (Ch). He held P&G’s registered community design (“RCD”) No, 000097969-0001 valid and infringed. The indication of the products to which the design is intended to be applied is “sprayers”. P&G use it in a number of countries (but not yet here) for an air freshener product called Febreze. The Reckitt spray canister held to infringe is for their Air-Wick room air conditioner.

Mr Henry Carr QC argued for Reckitt and Mr Roger Wyand QC for P&G.

The most important things in a case about registered designs are:

The registered design;

The accused object;

The prior art.

It follows that a place for evidence is very limited indeed. By and large it should be possible to decide a registered design case in a few hours. The evidence of the designer, e.g. as to whether he/she was trying to make, or thought he/she had made, a breakthrough, is irrelevant. The evidence of experts, particularly about consumer products, is unlikely to be of much assistance: anyone can point out similarities and differences, though an educated eye can sometimes help a bit. Sometimes there may be a piece of technical evidence which is relevant – e.g. that design freedom is limited by certain constraints. But even so, that is usually more or less self-evident and certainly unlikely to be controversial to the point of a need for cross-examination still less substantial cross-examination.

In Thermos v Aladdin [2000] FSR 402 at 404 I said:

Thermos was decided in relation to the then UK domestic law of registered designs. This case is the first to reach this Court concerning a
Community Design Registration (“CDR”) granted pursuant to Council Regulation EC 6/2002 (“the Regulation”). But everything I said then applies also to actions about CDRs too.

It follows that the design history of the P&G design, and whether Reckitt copied (which was hinted at darkly by Mr Wyand) was irrelevant. Of more relevance (though even this is secondary) is the fact that the P&G design (in the form of its physical embodiment) received some independent accolade. The Judge records this at:

The only other evidence of relevance in this case was some technical evidence about design freedom and purpose. I will come to it in due course but at this stage merely record that it was of minor significance, was largely self-evident and certainly not worth actual witnesses and cross-examination.

The Judge helpfully sets all that is material out, and I gratefully borrow it as Appendix 1.

There are six representations shown:

The category of product is described as “sprayers.” I have reproduced all 6 pictures. The Judge left out the top and bottom views. Rightly everyone ignored these as having any real significance. I reproduce them only so that readers can see that for themselves.

We also had physical embodiments of the design, both in plain white and in their actual Febreze get-up. It has long been the practice, where there is a physical embodiment of a registered design, for the court to look at that. Of course one has to be very careful that the physical object is a true reproduction of the design. In the present case it was accepted that that was so, subject to some minor variation so trivial as to be hardly discoverable even on a close comparison between the registered design and the Febreze sprayer.

The Judge reproduced pictures of this at [19]:

Unlike the registered design (which has a slightly tapered top to the can which blends with the top) this has a top which fits on a standard cylindrical canister. The Judge also reproduced as an Appendix pictures of the accused product viewed from the same position as each of the design representations. I reproduce his appendix as Appendix 2. There was some criticism that this did not give a fair comparison because of the manner of lighting. I do not go into this. The reader will be assisted by better photographs of the actual product which were supplied to us and which I reproduce as Appendix 3.

By Art. 10.1 “the scope of the protection conferred by a Community design shall include any design which does not produce on the informed user a different overall impression.”

The Regulation does not tell us much about the notional “informed user.” He/she is clearly not quite the same sort of person as the “person skilled in the art” of patent law. The equivalent to that person in the field of design would be some sort of average designer, not a user. Recital 14 assists a bit. It says:

The recital is actually framed around the requirement for registrability – whether the design has “individual character” – rather than the test for infringement. Curiously the reference to the “existing design corpus, taking into consideration the nature of the product to which the design is applied” is not expressly carried over into the text of any of the actual Articles of the Regulation. But self-evidently it is relevant to their interpretation. What it tells us is that for the purposes of registrability the notional informed user is to be taken as aware of other similar designs which form part of the “design corpus”. Further, the “overall impression” to the “informed user” is also a key ingredient of the infringement test. So for that test too the notional informed user must be taken to be aware of the “existing design corpus.”

The recital uses another phrase not carried over into the Articles – “clearly differs”. Only if the “overall impression” “clearly differs” from that of the “existing design corpus” will the design have an “individual character.” Plainly that is relevant for registrability, even if Art. 6.1 does not expressly use “clearly differs”. Does the phrase also apply to the infringement test? Does an accused design escape infringement only if its overall impression “clearly differs” from the registered design? All Art. 10.1 says is “different overall impression”. Does that really mean “clearly different” – the word “clearly” requiring much blue water between the accused and registered design for non-infringement?

The Judge thought it did (see [57]). But I do not. Different policies are involved. It is one thing to restrict the grant of a monopoly right to designs which are shown “clearly” to differ from the existing design corpus. That makes sense – you need clear blue water between the registered design and the “prior art”, otherwise there is a real risk that design monopolies will or may interfere with routine, ordinary, minor, every-day design modifications – what patent lawyers call “mere workshop modifications.” But no such policy applies to the scope of protection. It is sufficient to avoid infringement if the accused product is of a design which produces a “different overall impression.” There is no policy requirement that the difference be “clear.” If a design differs, that is enough – an informed user can discriminate.

Actually the Judge did not expressly, in his detailed consideration of infringement, import any requirement of “clearly differ”. It probably, however, underlay some of his reasoning.

I move on to mention another point. The right conferred applies to any sort of product even though the registration contains an indication of the type of article for which it is intended, see Art. 36(2). Where the alleged infringement is a quite different sort of product from that indicated as being the intended type, there may be problems about identifying the attributes of the informed user – is he a user of the kind of article such as the alleged infringement or a user of the kind of intended article? Or both? But none of that applies here.

Here the “design corpus” of which the informed user is taken be aware are other sprayers generally known – not just sprayers for air fresheners or even those of the kind purchased by ordinary consumers. There was no dispute about this.

The “informed user” test makes sense: a user who has experience of other similar articles will be reasonably discriminatory – able to appreciate enough detail to decide whether a design creates an overall impression which has individual character and whether an alleged infringement produces a different overall impression.

It follows that the informed user is not the same as the “average consumer” of trade mark law. The ECJ describes the attributes of this notional figure in Lloyd Schuhfabrik Meyer v Klijsen Handel, Case C-342/97 [1999] ECR I-3819:

The informed user of design law is more discriminating. Whilst I do not say that imperfect recollection has no part to play in judging what the overall impression of design is, it cannot be decisive. The Judge placed more emphasis than I think is right on an “imperfect recollection” test or something like it. He accepted Mr Wyand’s submission that “the overall impression of a design is what sticks in the mind after [my emphasis] it has been carefully viewed” [57]. I would say that what matters is what strikes the mind of the informed user when it is carefully viewed.

I think the Higher Provisional Court in Vienna, in holding that P&G’s design is not infringed by the Air-Wick product (decision of 6^th December 2006, overruling a lower court decision granting an interim injunction) was right when it said:

Policy considerations point the same way. The main point of protection of a trade mark is to prevent consumer confusion or deception. The possibility of imperfect recollection plays a significant part in that. The point of protecting a design is to protect that design as a design. So what matters is the overall impression created by it: will the user buy it, consider it or appreciate it for its individual design? That involves the user looking at the article, not half-remembering it. The motivation is different from purchasing or otherwise relying on a trade mark as a guarantee of origin.

So the informed user is alert to design issues and is better informed than the average consumer in trade mark law. Things which may infringe a registered trade mark may not infringe a corresponding registered design. I cannot think of any instance where the reverse might be so.

Another thing is also clear. Where shapes are, to some extent, required to be the way they are by reason of function, the informed user is taken to know that. That is what Art 6(2), (for validity) and Art. 10(2) (for scope of protection) require. Take an aspect of this case. Both products have a trigger and something of a “pistol grip”. There is some constraint on design freedom for this – the product must be grippable so that the index finger can pull the trigger, the trigger must be shaped to fit the finger and have sufficient space behind it for it to be pulled. That is a given. The informed user must take those requirements into account when assessing overall impression.

Thus Art. 10(2) is a narrowing provision. Smaller differences will be enough to create a different overall impression where freedom of design is limited.

In this connection I have no doubt that the “freedom of the designer” referred to is not that of a particular party – it is the degree of choice a designer would have in creating his design, not particular constraints on a particular party. The Judge so held at [43]. Mr Carr did not challenge that, and rightly so.

The Judge considered the nature of the informed user at [30-41]. He considered a formulation by HHJ Fysh in Woodhouse v Architectural Lighting [2006] RPC 1 and several decisions of the invalidity division of OHIM. I set the passages out for convenience here, though in the end I do not think they are saying anything different from the Austrian Court or what I have concluded above:

Honda Motor Co Ltd v Kwang Yang Motor Co Ltd (OHIM Ref: ICD 000001006; 30 August 2006, an internal combustion engine, intended principally for lawnmowers):

Once one has identified the notional “informed user” correctly and what he would know about the design corpus, one asks whether the accused product produces “a different overall impression” to such a person?

This test is inherently rather imprecise: an article may reasonably seem to one man to create “a different overall impression” and yet to another to do so. It is always so with the scope of rights in a visual work. You need to cover not only exact imitations, but also things which come “too close”. Whatever words you choose, you are bound to leave a considerable margin for the judgment of the tribunal. Thus in copyright the test for infringement is doing a forbidden act “in relation to the work as a whole or any substantial part of it” (s.16.(3) of the Copyright Designs and Patents Act 1988) and for infringement of a UK registered design the test used to be doing a forbidden act in respect of “an article ,,, to which [the registered design] or a design not substantially different from it has been applied.” The test of “substantiality” leaves just the same sort of margin as the “overall impression” test. Whether it is the same test or not I leave for others to consider.

Having said that, however, there are some general observations that can be made:

For the reasons I have given above, the test is “different” not “clearly different.”

The notional informed user is “fairly familiar” with design issues, as discussed above.

Next is not a proposition of law but a statement about the way people (and thus the notional informed user) perceive things. It is simply that if a new design is markedly different from anything that has gone before, it is likely to have a greater overall visual impact than if it is “surrounded by kindred prior art.” (HHJ Fysh’s pithy phrase in Woodhouse at [58]). It follows that the “overall impression” created by such a design will be more significant and the room for differences which do not create a substantially different overall impression is greater. So protection for a striking novel product will be correspondingly greater than for a product which is incrementally different from the prior art, though different enough to have its own individual character and thus be validly registered.

On the other hand it does not follow, in a case of markedly new design (or indeed any design) that it is sufficient to ask “is the alleged infringement closer to the registered design or to the prior art”, if the former infringement, if the latter not. The tests remains “is the overall impression different?”

It is legitimate to compare the registered design and the alleged infringement with a reasonable degree of care. The court must “don the spectacles of the informed user” to adapt the hackneyed but convenient metaphor of patent law. The possibility of imperfect recollection has a limited part to play in this exercise.

The court must identify the “overall impression” of the registered design with care. True it is that it is difficult to put into language, and it is helpful to use pictures as part of the identification, but the exercise must be done.

In this exercise the level of generality to which the court must descend is important. Here, for instance, it would be too general to say that the overall impression of the registered design is “a canister fitted with a trigger spray device on the top.” The appropriate level of generality is that which would be taken by the notional informed user.

The court should then do the same exercise for the alleged infringement.

Finally the court should ask whether the overall impression of each is different. This is almost the equivalent to asking whether they are the same – the difference is nuanced, probably, involving a question of onus and no more.

There was no dispute as to these. It must be shown that the Judge has gone wrong in principle, see, e.g. Designers Guild v Russell Williams [2001] FSR 113 where Lord Hoffmann said, speaking of the closely analogous question of substantiality in relation to copyright infringement:

Mr Carr contends that the Judge did go wrong in principle: that he did so in the following way. He first correctly identified the overall impression of the P&G design. He also at some points identified the overall impression of the Air Wick product. But in the end he failed to compare the overall impression of the P&G design as he had found it to be, with the overall impression of the Air Wick product.

The Judge having made that error, Mr Carr submits that we should form our own overall impression of the design and the accused sprayer and that we should conclude that the impressions are different.

Putting it another way, Mr Carr submits that on a proper analysis the Judge actually found that the overall impression of each was different and he ought to have concluded that there was no infringement.

Before going on to consider this argument in detail, I should record that Mr Carr, under a little pressure from the Court, abandoned his point about decoration, rejected by the Judge at [71-73]. He was right to do so. The registration is evidently for a shape. The proper comparison is with the shape of the alleged infringement. Graphics on that (or on the physical embodiment of the design) are irrelevant.

The Judge first described the design thus:

Mr Carr submits that this description fairly embodies the essence of the design, its overall impression. If he had applied that as the overall impression he would have found the Air Wick Product to give a different impression.

The first time the Judge refers to the impression created by the Air Wick design was at [20]. He said:

If that is right (and Mr Carr submits it is), then surely it creates a “different overall impression.”

Further, submits Mr Carr, when the Judge came to consider the question of the overall impression he failed to apply his own earlier description. Instead he identified a list of features which were stated at too high a level of generality. The Judge said:

This, Mr Carr submits, does not capture the overall impression the judge had found at [8-9]. The Air Wick product does not have a “smooth and dynamic feel” or an “elegant sense of movement”. Nor does it have “flowing lines”. There is no curved canister. The error the judge made was to generalise too much. So for instance, item (iv) was described in [8] in more detail – enough to convey the impression made: “The curves of the can and the head merge around the neck giving the product an integrated feel.” Similarly at [8] the Judge referred to the “curves of the canister” but his item (v), said to be a “dominant feature” is just a cylindrical body – which would have seemed wholly unremarkable to an informed (indeed any) user.

So, submits, Mr Carr, it is not right to identify “dominant features” at such a level of generality that the overall impression is not captured.

The Judge at [67] identified differences between the registered design and the Air Wick Product but went on to say at [67]:

In saying that “the registered design is of far greater quality and more integrated”, Mr Carr submits that the Judge was in fact finding that there was a different overall impression. And that he should have so held.

Mr Carr further submits that the Judge was wrong about item (ii). It is not a dominant feature of the design that the trigger is within the footprint of the base – if you just look at the design pictures that is not a thing that sticks out at all. Moreover being within the footprint is a true technical consideration – if it sticks out there will be problems about packing the canister in a box. So it is an aspect which the informed user would discount when deciding what the overall impression is.

Mr Carr further submitted that that there were real, significant differences between the Air Wick product and the registered design. Most importantly:

The Air Wick product has something of a “hammer head”. The top “lozenge” has depth. It is shaped something like a Foreign Legionnaire’s kepi. You could not say the same about the top of the registered design. As the Austrian court put it:

The registered design has a much thinner, more elegant, “neck”. That is indeed so, particularly in the front or back views.

The registered design top ellipse is significantly larger than that of the Air-Wick product.

The triggers, given the design constraint that they have to be triggers, could hardly be more different.

The “cape” of the registered design flows elegantly over the curved top of the canister. That of the Air Wick product just sits on the top and is cut off at much the same level as the front. There is no integration.

Moreover, he submitted, there were elements of functionality which the Judge failed to consider:

I have already commented on the trigger/footprint point.

Once you have decided on a trigger mechanism you will need some sort of flaring so that the trigger can be gripped without the product slipping from the hand. Both the design and Air Wick have flared necks which achieve this, although the Air Wick does not do it as well. One should, for functional reasons, concentrate on the detail rather than the existence of the flare. If one does that, the flares are different – that of the Air-Wick product being significantly less marked.

The direction of the spray nozzle, angled up, is functional – if, as you naturally will, you hold the canister vertically you need the spray to go upwards. The general line needs to be angled up. So the fact of an upwards inclination should be discounted.

As to the Judge’s drawing “some comfort” (at [70]) from interlocutory decisions on parallel proceedings in other European countries in favour of P&G, Mr Carr submitted that they were of minor significance. Moreover, the position had changed in that the Austrian court of appeal had reversed the finding of the court below, as I have already recorded.

Mr Wyand emphasised the purpose of protection as set out in recital 7:

Next, and this was Mr Wyand’s main point, he drew attention to how different the P&G design was from anything that had gone before. And he was right about that. The attempt to invalidate the design by reference to the prior art to my mind only served to emphasise just how innovative the P&G was. We decided to dismiss the appeal on validity without calling upon Mr Wyand. The Judge dealt with this shortly at [74-77]. He was right to do so. Any attempt to re-state his reasons would be superfluous.

Mr Wyand went on to emphasise how little there was in reality of design restraint by reason of function. The very fact of the large difference between the registered design and anything in the prior art – the “existing design corpus” proved this. If all spray canisters had to have a generally similar appearance, it would not have been possible for the P&G design to be so different from anything that had gone before.

That too is right. A large departure from the prior design corpus is indeed an indication of design freedom. Where a departure is small that may be for two reasons: technical restraint or simply lack of imagination by prior designers. The present design is a case of the latter. Mr Carr’s points about design restraint for technical reasons are rather a long way from showing that a trigger operated sprayer must have anything like the general appearance of the P&G design. They are too general. Take for instance the point about the nozzle pointing upwards and being seen to do that. That in no way requires more than that the nozzle does in fact point up and that its housing should in some way indicate that. It need not be elliptical, or square, or any particular shape to do any of that. Similarly his point about the grip – any sort of grip will do. You could even have a shape like a real revolver.

Given all that freedom, Mr Wyand submitted that the scope of protection should be wide – that it should cover what gives a “slightly cheaper, or slightly coarser impression.” It followed, he submitted, that the Judge cannot be faulted in his conclusion about same overall impression. And that is all the more so given the principles governing an appeal about that type of decision.

I have come to the conclusion that the Judge did err in principle, essentially for the reasons advanced by Mr Carr, excluding his submissions about functional design restraint. I think the Judge erred in the following ways:

In failing to apply the “overall impression” of the registered design he in effect had found at [8-9];

In failing to apply the overall impression of the accused product he had found (or summarised) at [20];

In failing, at the point where he was considering infringement, to state what the overall impression of the alleged infringement was;

In applying by implication a requirement that the accused product should give the informed user a clearly different impression.

In applying a “stick in the mind” test rather than “what would impress now” test.

In approaching the “dominant features” of the design at too general a level, a level such as not to convey in words the overall impression which would be given to an informed observer.

The Judge was concerned about the possibility that “a poor quality imitation would escape infringement.” I am not so concerned. We are here considering monopolies in designs, not trade marks. A “poor quality” imitation if it does not convey the same impression as the “original” will fail on its own design merits, or rather the lack of them. If it conveys the “same impression” then it can hardly be a “poor quality imitation” and will succeed for the same reason as the “original.”

Accordingly I am free to form my own view. I think the impression which would be given to the informed user by the Air-Wick product is different from that of the registered design. I say that for the reasons advanced by Mr Carr which I need not repeat here. The Austrian court put it well:

The similarities between the products are at too general a level for one fairly to say that they would produce on the informed user the same overall impression. On the contrary, that user would get a different overall impression.

Accordingly I would allow the appeal on infringement, but dismiss the appeal on validity.

I agree.

I also agree that the appeal on infringement should be allowed, but the appeal on validity should be dismissed for the reasons given by Jacob LJ.