Working Guidelines

by Luis-Alfonso DURAN, Reporter General,
Jochen E. BüHLING, Deputy Reporter General and
Ian KARET, Deputy Reporter General
Dariusz SZLEPER and Thierry CALAME,
Assistants to the Reporter General

Question Q181

Conditions for registration and scope of protection of non-conventional trademarks

I. Introduction

The use of trademarks in modern society is in a constant state of evolution. As technologies and advertising practices have changed and developed, so the types of sign used by businesses to promote and identify their goods and services have expanded beyond "traditional" trademark categories, such as words and devices, to encompass "non-conventional" marks, such as sounds, fragrances, colours, three-dimensional shapes, moving characters and holograms. As product marketing reaches new levels of sophistication, such "non-conventional" marks are increasingly recognised and understood by consumers.

In line with these trends, brand owners have in recent years increasingly attempted to
obtain registered trademark protection for "non-conventional" marks. Applications for
registration of colours or three-dimensional shapes have in particular become increasingly common. However, national laws have responded differently to the legal questions and administrative demands posed by attempts to register "non-conventional" marks.

For example, while it is widely recognised that the primary function of a trademark is to identify the origin or source of the goods or services in respect of which it is registered, national laws have differed in their acceptance of the extent to which "non-conventional" marks can fulfil this function. National laws have also been inconsistent in their approach to the need for "non-conventional" trademarks to be accurately recorded and reproduced on trademark registers in order to maintain an acceptable degree of legal certainty for third parties and have taken varying stances on the question of whether brand owners should be put to proof of de facto distinctiveness, acquired through use, before applications for "non-conventional" marks are allowed to proceed to registration.

II. WIPO Standing Committee on the Law of Trademarks, Industrial Designs
and Geographical Indications

The WIPO Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications has recently begun a programme of work considering inter alia the protection of "non-conventional" signs as trademarks. At its Eighth Session (Geneva, May 2002), the Committee launched discussions on various proposals for the harmonisation of substantive trademark law, including proposals for the revision of the definition of what may constitute a registrable trademark under national laws (SCT/8/3, "Suggestions for the Further Development of International Trademark Law", 26 April 2002).

The fifth principle for discussion set out in SCT/8/3 ("Definition of a mark") proposed that "[p]rovisions could be proposed to give a more complete and broader definition of a mark, for example by going beyond current definitions such as 'visible signs'". SCT/8/3 suggested that "provisions could build on Article 15(1) of the TRIPs Agreement in providing that Members 'shall' require, as a condition of registration, that signs be visually perceptible, or capable of being represented graphically, depicted or described by written notation, diagram or any other means". SCT/8/3 also proposed that provisions could "expressly include hologram marks, sound marks and olfactory marks".
The differing approaches taken by national laws to the registrability of "non-conventional" marks were apparent from the preliminary general remarks submitted in response to the Committee's proposals (reported at SCT/8/7, "Report Adopted by the Standing Committee", 14 November 2002). On the one hand, certain national delegations considered it "premature" to consider the registrability of "non-conventional" marks, in particular sounds and smells. On the other hand, a number of national delegations indicated that their national laws already adopted broad definitions of what may constitute a trademark, encompassing a varying range of "non-conventional" marks.

At its Ninth Session (Geneva, November 2002), the WIPO Standing Committee presented
a questionnaire exploring national practice on the questions of substantive trademark
law discussed in response to SCT/8/3. That questionnaire (SCT/9/3, "Further Development
of International Trademark Law and Convergence of Trademark Practices", 1 October
2002) is currently under revision as a result of comments provided through the SCT
Electronic Forum.

III. AIPPI Resolutions

a) Trademark Registrability

AIPPI has considered the general question of trademark registrability on various
occasions.

In Q23 (The notion of the trademark / Study on the Unification of the Right of
Trademarks, Yearbook 1963, New Series No. 13, 1st Part, 66th Year, page 83),
the AIPPI 1963 Berlin Congress adopted the principle that "a mark is a sign capable
of distinguishing the products or services of a person or a group of persons"
and that "[t]he distinctive character of a mark, in respect of the goods or services
designated by it, arises from the nature of the sign or of the use which has been
made of it".

In Q68 (Economic Significance, Functions and Purpose of the Trademark, Yearbook
1979/I, pages 463-465), AIPPI confirmed that "trademark law in every country
lays down the 'normative functions' of the trademark, being the function of indicating
origin and the function of identification" and that "when used in trade, the
mark additionally possesses 'economic functions' which may vary according to its
use and the type of trademark". AIPPI noted that "among these economic functions,
the following should especially be mentioned: the quality function, the advertising
function [and] the competitive function".

In response to Q92C/96 (Absolute Grounds of Refusal of Registration of Trademarks
/ What may constitute a Registrable Trademark?, Yearbook 1989/II, pages
309-317), AIPPI observed that "most countries agree that the main function of the
trademark is to distinguish goods and services of one undertaking from the goods
and services of another undertaking, and that the distinctive character is relative
in the sense that it must be evaluated in relation to the goods and services to
which the trademark is to be applied".

In response to Q96, AIPPI concluded that "the decisive criteria" for assessing the
registrability of a sign as a trademark should be its "distinctiveness". AIPPI noted
that the majority of national laws provide for a general definition of a trademark
and enumerate signs which are capable of constituting a registered trademark,
usually by way of illustration. Commenting that the giving in national laws of examples of signs which may be registrable as trademarks is convenient, but should
not be exhaustive, AIPPI concluded that "all signs should indeed be capable of
being registered as trademarks, provided that they are inherently distinctive or
have become distinctive in respect of the designated goods or services".
It should be noted that AIPPI has in the past also set up two Special Committees
(Q92E, Harmonisation of Substantive Provisions of Trademark Law and Q45 B,
The Role of Trade marks, Trade names and Geographical Indications) to consider
substantive provisions of trademark law. Those Special Committees have now
been dissolved.

b) Registrability of "non-conventional" Marks

In Q23, Q92C/96 and Q148, AIPPI has also addressed the more specific question
of the registrability of certain types of "non-conventional" marks.
In Q23, the AIPPI 1963 Berlin Congress affirmed that combinations of colours
should be treated as capable of constituting a trademark "on the sole condition
they are distinctive or have become distinctive for the products or services which
they designate".

In response to Q96, AIPPI also commented on the registrability of colours per se,
sounds and smells. Following its response to Q23, AIPPI concluded that a colour
per se should be registrable when it is or has become distinctive. AIPPI also concluded that sound trademarks should be registrable "at least if they can be represented by symbols", the registration protecting "the sound thereby represented".
However, in respect of smells, AIPPI noted that no national legislation at that time
provided for the registration of olfactory trademarks and concluded that "the
rather restricted interest in the registration of olfactory trademarks does not justify the complicated administrative and legal problems involved in such registrations".

AIPPI has also commented in some detail in Q96 and Q148 on the registrability
of three-dimensional signs. Looking at the broad question of whether three-dimensional
signs should be permitted as trademarks, AIPPI concluded in response to Q96 that three-dimensional signs are registrable "irrespective of their nature" and suggested various administrative procedures for the filing of applications to register three-dimensional signs. In response to Q148 (Three-dimensional marks: the borderline between trademarks and industrial designs, Yearbook 2000/II, pages 261-262), in which AIPPI explored the relationship between industrial design and trademark protection for three-dimensional signs, AIPPI observed that "[i]t is commercially important to recognize and protect intellectual property rights embodied in or represented by three-dimensional objects, compositions, or shapes … and both trademark rights and industrial design rights can be applied to achieve that objective".

Turning to the particular requirements for registration of three-dimensional signs
and noting that "it is accepted that 3D shapes should be protected and registered
as trademarks, as is already the case under many national laws, provided they
perform the trademark function", AIPPI resolved in response to Q148 that "3D shapes are protectable both as industrial designs and as trademarks, provided that the usual requirements for each modality of protection are satisfied".

However, in addition to those primary requirements, AIPPI also resolved that
"there may be other requirements which should be met in order to obtain and
maintain trademark protection". AIPPI resolved that "to be protected as a trademark,
[a] 3D shape must be distinctive and not solely functional or necessary, and
it must also satisfy other conditions generally applicable to trademarks". Noting
that "to be distinctive, [a] 3D shape must indicate product or service source to the
relevant consuming public", AIPPI further resolved that "the standard of required
distinctiveness may not be the same in every country or territory" and that "[i]t is
left to the appropriate legislature or courts in each jurisdiction to determine
whether a 3D shape can be considered to be inherently distinctive for purposes of
trademark protection, or whether the 3D shape can be protected under trademark
law only after it has acquired distinctiveness through use". However, AIPPI considered
that "protection of a 3D shape as a trademark should not be precluded by
the fact that the shape is used in combination with a word mark or another type of
sign" and that "the use of [a] 3D shape with [a] word mark or other type of sign can
lead to acquired distinctiveness for the 3D shape".

Finally, looking at issues of functionality and technical result, AIPPI resolved that
"[t]rade mark law should not protect 3D shapes that are solely functional or necessary,
namely those which solely: (a) result from the nature of the goods themselves,
or (b) are necessary to obtain a technical result relating to the nature of
the goods or services covered by the mark".

IV. Issues arising for particular Forms of "non-conventional" Trademark

a) Registrability of Colours per se and Combinations of Colours per
se

Applications to register a colour per se or a combination of colours per se (by
which expression is meant two or more colours displayed together) are complicated
by the fact that colours do not have an existence independent from the
products to which they are applied. As the Advocate-General to the European
Court of Justice has recently said in his Opinion in Libertel Groep BV v Benelux
Merkenbureau (Case C-104/01, 12 November 2002), concerning a Benelux application
to register the colour orange in respect of goods and services in international
classes 9 and 35-38, colour can only exist as an "attribute" of an object;
what a consumer sees when he perceives an object to which a colour has been
applied is not the colour, but the object assuming that colour. This difficulty is compounded in applications to register colours per se and combinations of colours per
se in relation to services, as such services have no concrete form on which the
colour can be used. At most, the colours can only be affixed to documents or other
items used in the provision of the services in question.

These issues impact both on the question of whether colours per se and combinations
of colours per se are capable of fulfilling the functions of a trademark and
whether they are capable of sufficiently accurate graphical representation to satisfy
the need to preserve legal certainty for third parties. Even within the European
Union, which is subject to the harmonising Trade Marks Directive (89/104/EEC), certain Member States have explicitly accepted that colours per se are capable of being registered as trademarks, while others have expressly excluded colours per se from registration, accepting only colours which are applied to specific shapes. National laws also impose different conditions for the filing of applications for colour marks, with some jurisdictions requiring a sample of the colour to be submitted as part of the application and others requiring the filing of a colorimetric reference or other colour code, for example a Pantone? reference.

The position may also be further complicated if a colour is associated with or used
on a material which has a specific texture and which therefore creates a particular
effect.

There also remains the question of whether colours per se and combinations of
colours per se can ever be inherently distinctive or should only be accepted for
registration on the basis of de facto distinctiveness acquired through use. Certain
national laws have, for example, treated colours per se as inherently distinctive
where they are unusual for the goods or services applied for. Others have rejected
this position, influenced in particular by a perceived need to keep basic colours
free for general use in the light of the limited number of colours available. Proving
the acquired distinctiveness of a colour per se can also often be complicated by
the fact that the colour is generally always used in conjunction with a shape or
word marks which may also be treated by consumers as indicators of the origin of
the products or services in question.

b) Three-dimensional Marks

Although applications to register three-dimensional signs are increasingly common,
there remain significant questions as to the extent to which three-dimensional
signs are capable of distinguishing the goods of one undertaking from
those of other undertakings or can be considered to have a distinctive character,
either inherently or acquired through use. There are a limited number of shapes
which can fulfil certain functions (for example, acting as a container) and, as with
colour marks, consumers may be less used to understanding three-dimensional
shapes as an indicator of origin than other forms of trademark. It may also be difficult to prove that a three-dimensional mark has acquired distinctiveness through
use where that mark is always used in conjunction with additional, distinctive material such as a product name or house mark.

The treatment of three-dimensional signs within the EU has been the subject of
an important decision of the European Court of Justice (Philips -v- Remington,
Case C-299/99, 18 June 2002). In Philips -v- Remington, the European Court of
Justice held that it is not necessary for a three-dimensional sign to contain some
form of "capricious addition" in order to be considered capable of distinguishing
the product of the proprietor of the trademark from those of other undertakings.
However, a number of jurisdictions have required a three-dimensional sign to
have an arbitrary or fanciful element if it is to be considered capable of distinguishing or to have a distinctive character. In addition, a number of jurisdictions require that the trademark proprietor must prove de facto distinctiveness as a precondition to the protection of three-dimensional signs; for example, in Wal-Mart Stores Inc -v- Samara Bros Inc (22 March 2000), the US Supreme Court has held that unregistered trade dress protection will only be available under the US Lanham Act where the product design in question has acquired secondary meaning
through use.

Three dimensional trademarks are also often excluded from registration because
of technical or functional aspects of their design. As a matter of policy, there are
concerns that trademark rights, which may exist indefinitely, should not be applied
so as to give perpetual monopoly rights in a product the protection of which is
more suited to other forms of intellectual property right, such as industrial designs
or patents. For example, under US law a three-dimensional sign may only be reg-istered where it is primarily non-functional, in that it serves no utilitarian need or is
only de facto functional (the shape performs a function, but is not dictated primarily
by that function) as opposed to de jure functional (the shape is essential to the
operation of the product): under US law, the existence of patent protection for the
product in question may be relevant to such functionality objections. In the EU,
the European Court of Justice has held in Philips -v- Remington that three-dimensional
sign consisting exclusively of the shape of a product will not be permitted for registration if it is shown that the essential functional features of that shape are attributable only to a technical result. This may be the case even if there are other shapes which would achieve the same result.

c) Registrability of Smells

While certain jurisdictions, such as the US, have accepted that smells may in appropriate circumstances be capable of registration as trademarks, applications to
register smells as trademarks often encounter considerable difficulties.
This is particularly so where a graphical representation of the mark is required in
the application for registration. While it has been suggested in certain jurisdictions
that a precise description of the mark in words should be taken as a sufficient alternative to a graphical representation, even this requirement may not be capable
of satisfaction where it is difficult independently to "benchmark" the exact qualities
of the smell in question. The perception of smells is subjective. While chemical
formulae may describe the components used to create the smell, they do not describe
the smell itself and descriptions in words often presuppose that the reader
is familiar with the smell or aspects of the smell in question. Even actual product
samples may degenerate over time. The European Court of Justice in Ralf Sieckmann
-v- Deutsches Patent- und Markenamt (Case 273/00, 12 December 2002)
has recently ruled that signs which are not in themselves capable of being perceived
visibly can be registered as trademarks in the EU provided that they are
capable of visible representation. In order to allow third parties to determine the
precise nature of the mark in question, that representation should be "clear, precise,
self-contained, easily accessible, intelligible, durable and objective". However,
the European Court of Justice ruled that the smell mark in question in that case
was not adequately represented by a chemical formula, description in words or
product sample.

Applications to register smell marks may also be rejected on the basis that the
smell in question is not perceptible independently of the goods of which it constitutes part. This is a particular problem for applications to register the smell of perfumes, in which the smell may be considered to constitute the product itself,
rather than being indicative of the origin of the goods in question. As such, the
smell fails to perform a trademark function and has no distinctive character.
d) Registrability of Sounds As with smell marks, the requirement to provide a graphical representation of a mark on filing of a trademark application is often a particular problem for sound marks. There are many different ways in which a sound may be recorded: for example, by an onomatopoeic or other description of the sound in words, by musical notation, by graphical representation in the form of a sonogram or spectrogram or by an audio recording of the sound. However, written descriptions and
musical notation may not provide a sufficiently clear and unambiguous representation
of the sound in question for third parties to be able to assess the exact nature
of the mark applied for. For example, a sound mark recorded by means of musical notation may differ if different instruments are used or, even using the
same instruments, from one performance to the next.

Introducing a requirement to file an audio recording of the mark applied raises
complex practical problems for trademark registries in relation to the storage of
such audio versions, publication of sound mark applications and the public availability for inspection of the relevant entries on the register. Even if such facilities are available, however, the perception of sound marks by an individual listener remains subjective. This has the potential to influence not only the scope of protection of sound marks as assessed in infringement claims, but also the process of
examination on relative grounds in jurisdictions where examination for prior rights
is provided for in national law. Within the EU, the decision of the European Court
of Justice in Ralf Sieckmann -v- Deutsches Patent- und Markenamt appears to
preclude the filing of audio recordings or graphical representations of sounds in
the form of sonograms or spectrograms in the absence of a further, adequate visible
representation: audio recordings are not "graphical representations" and
sonogram or spectrogram readings are unlikely to be considered sufficiently clear
or intelligible to persons viewing the register.

e) Registrability of moving Images and Holograms

Certain jurisdictions have recently begun to accept for registration trademark applications for marks consisting of moving images. For example, the OHIM has accepted
a number of applications for registration of "moving image design marks", being published in the form of stills showing the different various stages in the progression of the moving image, accompanied by a description in words of the progression of that sequence. The Australian Trade Marks Registry has gone further,
accepting for registration a mark consisting of "a visual representation" of a particular character "in static pose or in animation", without limitation of the mark to the particular poses shown in the pictures of the character supplied as part of the
application for registration.

Similarly, the French Institut National de la Propriété Industrielle has accepted
holograms for registration. Other jurisdictions have, however, rejected the proposition that holograms should be capable of registration as trademarks on the basis that they change colour and shape depending upon the way in which they are
viewed. While certain jurisdictions, such as the UK, may be prepared to accept a
hologram for registration where all possible variants of the hologram's appearance
can be pictorially represented on the application form, this may not be practicable
for complex holograms consisting of multiple views from a large number of
different angles.

The registration of moving images and of holograms as trademarks again raises
issues as to the degree of fixed and defined graphical representation which is to
be required by national laws in order to provide sufficient legal certainty for third
parties. While it is possible that moves towards more sophisticated means of filing
and publishing trademark applications (for example, on-line filing and publication)
may provide a means to meet some of the technical demands involved in the
acceptance of such forms of trademarks, it is by no means certain that all registries
and consumers will have the necessary technology to be able to implement these advances.

f) Scope of Protection

As greater numbers of "non-conventional" trademarks proceed to registration, so
the number of decisions on the scope of their protection should also increase.

However, all of the issues discussed above in relation to the registrability of "nonconventional" marks have their corollary in the assessment of the scope of their
protection. For example, while it might be argued that the scope of protection of
"non-conventional" marks should not be assessed any differently from "traditional"
forms of trademark registration, any doubt as to the exact definition or distinctiveness of a registered mark may lead a national court to conclude that the registration in question should be granted only a narrow scope of protection.

Some jurisdictions appear to have taken a purposive approach to the enforcement
of registrations for "non-conventional" marks, influenced in particular by the
need to give effect to such registrations in a meaningful manner. The Northern Ireland
Court of Appeal in BP Amoco Plc v John Kelly Ltd has, for example, held that
the assessment of infringement of a pure colour trademark registration should be
effected on a mark-for-mark basis, ignoring any additional get-up used by the alleged
infringer on the product in question. The Northern Ireland Court of Appeal
reached this conclusion on the basis that any approach other than a mark-formark
comparison would undermine the value of trademark registration for a colour per se.

However, other national courts appear to favour a stricter approach. For example,
an Italian Court has ruled that a trademark registration for a single colour per se
is weak and would only be infringed by a slavish imitation of the registered mark
(Aquatherm GmbH v Wavin PsA, 19 June 2001).

V. Questions:

Groups are invited to answer the following questions under their national laws.

1. How is the scope of what constitutes a registrable trademark defined? Does that definition list specific examples of permissible types of trademark?

2. What categories of signs are capable of registration as a trademark? Are any categories of signs excluded from registrability as a trademark?

3. Where applications to register a colour per se or a combination of colours per se may be accepted for registration:

3.1 in relation to goods, is such an application acceptable if it does not specify the outline, contours or proportions in which or the shape or object to which the colour is to be applied;

3.2 is such an application registrable in relation to services and, if so, under what conditions;

3.3 is an applicant required to file a sample of the colour(s) applied for and/or a colorimetric reference or other colour code;

3.4 can the mark be considered inherently distinctive in relation to certain goods or services;

3.5 will the mark only be accepted for registration after de facto distinctiveness acquired through use has been shown;

3.6 are certain colours denied registration on the basis that there is a need to keep them free for general use; and

3.7 to what extent do other visual aspects of the way the mark is used affect the assessment of the distinctiveness of the mark?

4. Where applications to register three-dimensional signs may be accepted for registration:

4.1 what form of representation of the three-dimensional sign is accepted as part of the application for registration;

4.2 what are the criteria for assessing whether the three-dimensional sign can be considered capable of distinguishing the goods or services of one undertaking from those of another;

4.3 what are the criteria for assessing whether the three-dimensional sign can be considered inherently distinctive in relation to certain goods or services;

4.4 will the mark only be accepted for registration after de facto distinctiveness acquired through use has been shown;

4.5 are certain shapes denied registration on the basis that there is a need to keep them free for general use;

4.6 to what extent do other visual aspects of the way the mark is used affect the assessment of the distinctiveness of the mark; and

4.7 to what extent do technical or functional considerations prevent registration?

5. Where applications to register smells may be accepted for registration:

5.1 what form of representation or description of the smell is accepted as part of the application for registration;

5.2 how is the capacity of the smell to function as a trademark and/or its distinctiveness assessed; and

5.3 how is the mark made available to the public on publication of the trademark application and thereafter?

6. Where applications to register a sound mark may be accepted for registration:

6.1 what form of graphical representation of the mark applied for is required as part of the application for registration;

6.2 can audio recordings of a sound mark be filed as part of the application for registration; and

6.3 if such audio recordings are filed as part of the application for registration, how are these recordings made available to the public on publication of the trademark application and thereafter?

7. Where applications to register moving images or holograms may be accepted for registration:

7.1 what form of representation of the moving image or hologram is accepted as part of the application for registration; and

7.2 how is the moving image or hologram made available to the public on publication of the trademark application and thereafter?

8. How is the scope of protection of "non-conventional" trademarks assessed? In particular, are registrations for "non-conventional" trademarks treated any differently than registrations for "traditional" forms of trademark either in substantive or procedural terms?

9. To what extent is assessment of the registrability of "non-conventional" trademarks influenced by issues relating to enforcement and the scope of protection which should be afforded to those marks?

10. To what extent is the assessment of the registrability of "non-conventional" trademarks influenced by the fact that there may be a limited number of certain "non-conventional" marks (such as colours or shapes) available for general use?

11. To what extent is the assessment of the registrability of "non-conventional" trademarks influenced by the fact that consumers may be less used to understanding "non-conventional" marks as an indicator of origin?

As well as stating the laws of their respective countries on the above-listed questions, Groups are invited to make any proposals for harmonisation of national laws which they consider desirable and to offer any further comments or observations of interest.

Note:

It will be helpful and appreciated if the Groups follow the order of the questions in their Reports and cite the questions and numbers for each answer.