view that the Association has failed to establish infringement of the remaining device marks.
[19] The appellant, in defence, relied on section 34(2)(b) of the Act which in essence provides that a registered
trade mark is not infringed by the use by any person of any bona fide description or indication of the
geographical origin of his services.2 This defence represents the other side of the coin of the requirement
that the infringing use has to be trade mark use. A bona fide description or indication of the geographical origin
of an alleged infringer's services amounts to nontrade mark use and whether one considers it as part of the
trade mark owner's cause of action or as a defence does not make much difference. Since I have already
found that the appellant's use of its trade or brand name and the domain name amounted to trade mark use,
this defence has to fail.
The counterapplication geographical origin
[20] The appellant launched a counterapplication for the expungement of the Association's registrations by relying
on section 10(2)(b) of the Act. It provides that a mark may not be registered and a registered trade mark is
Page 417 of [2010] 2 All SA 409 (SCA)
liable to be removed from the register if it consists exclusively of a sign or an indication which may serve, in
trade, to designate the geographical origin of the services.3 The factual basis of the argument was that
View Parallel Citation
Century City has become a geographical location that designates the geographical origin of services.
[21] It is necessary to consider the history of this provision more closely since both sides relied on judgments in
support of their respective arguments concerning its interpretation on the supposition that our law and, for
instance, English law on trade marks have always been the same, and that general propositions in older
trade mark cases are still applicable under the present legislative regime.
[22] The Patents, Designs and Trade Marks Act of 1883 (46 and 47 Vict c 57) did not deal with trade marks with a
geographical connotation. It did, however, list the matters which trade marks had to consist of or contain.
Trade mark registrations then were limited to names of individuals or firms represented in "some particular
and distinctive manner", a signature, a distinctive device and the like, as well as a "fancy word or words not in
common use" (section 64(1)(c)). The 1888 amendment added "an invented word or words" to the list and
replaced "fancy word or words not in common use" with "a word or words having no reference to the
character or quality of the goods, and not being a geographical name". The reasons for these provisions were
said to be that the vocabulary of the English language is common property which belongs to all and that no
one should be permitted to prevent other members of the community from using "for purposes of description"
a word that has reference to character or quality of goods and, one assumes, geographical names (In re
Eastman Photographic Materials Co Ltd v The ControllerGeneral of Patents, Designs and Trademarks 1898 AC 571
(HL) per Lord Herschell). Fry LJ referred to the perpetual struggle to enclose and appropriate as private
View Parallel Citation
property certain little strips of the great open common of the English language (In re Dunn's Trademarks
(1889) 41 ChD 439). It is apparent from the use of the phrase "for purposes of description" that the concept
of trade mark use had not yet been understood fully.
Page 418 of [2010] 2 All SA 409 (SCA)
[23] The Trade Marks Act of 1905 (5 Edw VII c 15) was not much different although in regard to geographical
names it prohibited the registration of any name if it was "according to its ordinary signification a geographical
name" (section 9(4)). This was introduced to overcome the objection to the registration of geographical
names that were not generally known. Courts nevertheless struggled with its application (TA Blanco White
and Robin Jacob Kerly's Law of Trade Marks and Trade Names (11ed) (1983) at 83). The Act was amended to
create a twopart register. The law was then consolidated in the Trade Marks Act of 1939 (1 and 2 Geo VI c
22). In relation to a registration in Part A of the register (section 9(1)), which dealt with inherently distinctive
marks, this Act was identical to the 1905 Act but in relation to marks that were capable of becoming
distinctive, which were registrable in Part B, a similar limitation did not exist.
[24] South African legislation did not in all respects follow suit. The Patents, Designs, Trade Marks and Copyright
Act 9 of 1916 permitted the registration of "a distinctive word or words not reasonably required for use in the
trade" and "any other distinctive mark" (section 99). The Act, accordingly, did not prohibit the registration of
geographical names per se and consequently any such name, which was distinctive and not reasonably
required for use in the trade, could have been registered. In addition, section 126 provided that a trade mark
registration could not interfere with the bona fide use by a person of his own place of business. This Act was
replaced by the Trade Marks Act 62 of 1963, and although it was largely based on the 1939 British Act it
retained the provisions of section 99 the 1916 Act but limited them to Part A registrations. In other words,
words not reasonably required for use in the trade were capable of registration; there was no specific
reference to geographical names; and section 46 was in this regard in the same terms as the mentioned
section 126.
[25] The 1963 Act was materially amended by Act 46 of 1971. It provided that in order to be registrable in part A of
the register, a trade mark had to contain or consist of a "distinctive" mark ( section 10(1)). "Distinctive" was