Skip to content

THOUGHTS ON TRADE MARK PRIORITY IN AMERICAN, BRITISH, AND SOUTH AFRICAN LAW

23 August 2016
– 47 Minute Read

DOWNLOAD ARTICLE

THOUGHTS ON TRADE MARK PRIORITY IN AMERICAN, BRITISH, AND SOUTH AFRICAN LAW *By Wim Alberts
Introduction
The determination of priority of rights is one of the most fundamental issues in trade mark law. Since the introduction of a system of registration there is, in particular, the potential for a conflict between rights derived from use, and those derived from registration, or the lodgment of an application. What would, at first glance, seem to be a matter merely requiring a simple calculation, can often turn out to be a complex task. Certain guidelines have however been set out in decisions, and these are discussed below.
Opposition of applications
It has been said, as far as the United States is concerned, that priority and ownership is determined, not by a race to the Patent and Trademark Office, but “a race to the market place.” (Holsum Bakeries, Inc. v General Baking Co., 228 F. Supp 962 (E.D.La. 1964)). It was also said (Mccarthy Trade Marks and Unfair Competition (2002) 16-6) that:

“The United States, unlike many civil law nations, has a rule of priority that is based on first-to-use not first-to-register.”
On the other hand, the filing of an intent-to-use application can provide priority (section 7(c) of the Lanham Act of 1946). The filing of such an application can be seen to provide rights that trump that of a prior user. In a sense such an applicant, B, can thus prevail, although the applicant for registration, A, has actual use prior to that of B. In the conventional situation the question would simply be who has the earliest date of use. An intent-to-use application can be used though to establish priority even if the use concerned does not predate the earliest date of use of the other party. Eventually use is required, of course. Under the British and South African systems priority can, however, be based on an earlier filing date, without any use.
In South Africa, in instances where an earlier application is opposed by a later applicant, the latter would, in terms of section 10(16) of the Act, have the burden of proof of establishing existing rights at the time of filing. In relation to previous legislation, it was said (by the authors Webster and Page South African Law of Trade Marks (1986) 127) that an arbitrary rule based on an earlier filing date will prejudice the later applicant. If forced to oppose an application, he would have to prove that the prior applicant is not entitled to registration. As a rival applicant, he only need show that his claim to registration was stronger, and could prevail if he has use which, although not sufficient to found an opposition only on prior use, would be sufficient to establish a stronger claim. It was said that if he was an objector, he would not succeed. It seems that with section 10(16) the matter has come full circle. Now, the later applicant must prove that the prior applicant is not entitled to registration, but although not a rival applicant, but an objector, he can, indeed, rely on the use of a mark to an extent that would not be sufficient for an opposition just on the basis of prior use, but which, nevertheless, establishes that registration would be contrary to the rights thus built up.
Where a prior user relies on use that has created a reputation as the basis for an opposition, the provisions of section 10(12) will be applicable. No specific case has yet been decided on this section’s import. However, it is likely that the approach set out in Oils International (Pty) Limited v WM Penn Oils Limited 1965 (3) SA 64 (T) will be followed. This requires the existence of a reputation. In the situation where section 10(16) applies, a difference exists, in that it presupposes that the opponent has also filed an application, albeit with a later filing date. The opposition must be based on “existing rights” as at the filing date of the earlier application. Commentators have suggested that “existing rights” could include use of a mark to an extent that would not be sufficient for an opposition under section 10(12), but which, nevertheless, establishes that registration would be contrary to the rights thus built up. There is clearly a lesser burden of proof involved. The inevitable conclusion, if the commentators’ view is accepted, is that the filing of an application reduces the quantum of use required. This can be related to the decision in the Oils International case in two respects. Firstly, it was said (page 72 D – E) that the applicant there took the important step of filing an application. This view can be aligned with the effect of the filing of an application as discussed above. Secondly, the legislature seems to have drawn a distinction between a section 10(16) and 10(12) opposition. In the latter case, there is, per definition, at the date of filing, the likelihood of confusion. In the case of section 10(16), this will not necessarily be the position. However, with time, this could become the position. In this sense it is perhaps fair to say that this type of protection is what the court in the Oils International (Pty) case seemingly accepted as being, at least, logical. More in particular, the contention by the opponent that whilst it did not (yet) have a reputation, it was sufficient to succeed with an opposition if acts effected by the filing date were such that they would “fructify” into a reputation later, so that there would then be confusion. The court reacted (page 69 F) as follows:

“Although that contention goes further, in favour of the objector, and any decided case to which we were referred, or of which I am aware, there seems to me to be logic in it.”
It thus seems that the approach contained in section 10(16) can be related to some, although obiter, dicta in earlier cases.
The relevant date for determining the merits of an opposition.
It is necessary to comment on the above issue. Often the matter is approached simply from the standpoint of whether the opponent had a reputation on the filing date. However, if the mark for which registration is applied for has been in use, then the matter should be considered at an earlier date. Kerly Law of Trade Marks (2001) 259 indicates that this is in line with the general principles pertaining to passing off, namely that the date for proving the existence of a reputation and goodwill is the date of commencement of the activities concerned. In this situation the opposition will only succeed if the claimed prior user has use that predates the earliest of the registrant’s date of application or date of first use. In other words, use prior to the application date is not decisive. It is submitted that South African law should follow the same approach.
The correctness of Kerly’s approach can be illustrated with the following example. A commences use in 1980. B commences use in 2000. A files in 2002. At the date of A’s filing, B does have use which predates the application date of A. However, it would violate the basic principle of priority to disregard the extensive prior use that A has, and to allow B to oppose A’s application on the basis of his more limited and later use (see however the discussion of geographical priority below). The position as at the filing date can thus be merely the tip of the iceberg.
It is interesting to contrast the above position with that in the United States. In terms of the latter, the position is that where an opposition is based on a registration, the date of first use of the opponent is not relevant. In such circumstances, a comparison of the respective dates of first use of the parties is thus not made (Lincoln Logs, Ltd v Lincoln Pre-Cut Log Homes 23 USPQ2d 1701 1704 (Fed. Cir. 1992)). Priority of use only features when an application is opposed on the basis of prior use, not a prior registration. It would not be a defence for A, the prior user and subsequent applicant, to rely on his use prior to the filing date of B, the prior registrant. From a British and South African perspective, this is a startling result. The solution to this dilemma is said to be that if the opposer has no right to its registration, the applicant should institute cancellation proceedings, which would, however, not produce the correct result If the mark has been registered for more than five years. On what basis does this apparent unfair principle rest? Callmann (Unfair Competition Trade Marks and Monopolies (1997) 25 – 349; 25 – 350) explains the position as follows:

“[T]he priority of the registrant over the applicant is assumed, and is not in issue. The registration on which the opposer relies is entitled to the benefit of all [section] 7(b) presumptions of validity, ownership, and exclusive right to use…If the applicant were allowed to assert priority of use as against a registration, he would in effect be collaterally attacking the validity and ownership of the registration…”
With regard to a cancellation action for marks that are older than five years, the author then (page 25 – 350 ) states:

“In such cases, therefore, if the prospect of confusion is established, the opposer can rest his case by relying on his previously issued registration and the statutory effects created thereby. Priority of trademark use will be wholly irrelevant to the enquiry.”
The situation is thus problematic largely due to the fact that a registration on the Principal Register becomes, in general, incontestable after five years. In South Africa there is no similar provision for incontestability. It would thus be possible to expunge a registered mark that is relied on to oppose an application. South African law is more flexible and less formalistic in comparison to American law on this point. It is submitted that it is competent for the Registrar or a court to decide in favour of the applicant that has prior use without him having to file an expungement application. This follows from the clear principle of priority of our law, as evidenced specifically by section 36(1).
Geographical priority
The above is a rather complex aspect that requires close attention. In the conventional situation A, the prior applicant and user, is a “clear winner”, and there is no obvious challenger to his superior right to obtain registration. Often there would be prior use in the whole of the country. This would imply that he can rely on common law rights to prevent passing off anywhere in the country, on condition that he has a reputation in the particular territory. Once registration has been obtained, he can rely on statutory protection as well, and prevent infringement in areas even if he does not have a reputation there. Seen in a certain light, rights are obtained in an area where there is no commercial presence. The question that arises here is to what extent the statutory position truly deviates from the accepted view of the protection afforded by the common law.
In terms of the decision in Dawn Donut Co. v Hart’s Food Stores, Inc., 267 F 2d 358 (2d Cir. 1959) the owner of a federal registration in the United States does retain priority, but the obtainment of relief is influenced by the fact that there is no likelihood of confusion until there is a likelihood of entry by the owner into that area. The basis for this view is that until that time consumers in the particular geographical area will not, factually, be susceptible to confusion as both products will not be available simultaneously. This position is not the same as that in South Africa, and relief is granted here on a regular basis in territories far removed from the actual business areas of the proprietor. The point is, however, that there is some tension in relation to the protection provided by the registration of a mark insofar as an area where he is not actually trading is concerned.
In South Africa, if an opponent in opposition proceedings has use in only one area, he can, nevertheless, oppose an application for the whole country. The American position is the same (McCarthy Trademarks and Unfair Competition 20-103). In South Africa, the other side of the coin is that if an opponent has prior but restricted use, it may be possible that his opposition will meet with limited success, in that the registration will be granted, with only an exclusion for the specific geographical area of his prior use (Webster and Page South African Law of Trade Marks (1997) 9-5). Does this imply that the prior applicant prevails over the prior user simply because the latter has not filed an own application?
It is relevant here to have regard to the American decision of Weiner King, Inc. v Wiener King Corp., 615 F 2d 512 (CCPA 1980) which related to a concurrent registration dispute. Here the subsequent user, B, conducted business in 20 states, the senior user, A, only in one. The court remarked (page 522) that its function is to weigh the relevant considerations. These included the interests of the prior user who remained a small, locally oriented business, and those of a subsequent user, whose intent is to expand nationwide, and who has done so. Registration was granted to B for the remainder of the United States, with the exception of the particular locations where A had conducted business. In Nark, Inc. v Noah’s, Inc., 560 F Supp 1253 (E.D. Mo 1983) a similar approach was followed, on the basis of, amongst others, the view that the senior user has relinquished the right to nationwide protection.
In South Africa, a practical example could be the following. A commences use in 2003, in Upington, in the Northern Cape. His business has, ever since, been restricted to that area. In 2000 B commences use in the whole of the Gauteng and Free State provinces, on an extensive basis, and files an application in 2002. If, as was said in the Weiner King case, the relevant considerations are balanced, it seems inequitable for A to be able to oppose B’s application. However, it is clear that it is not because of the inherent merit of B’s earlier filing date, but rather an evaluation of what is fair in the particular circumstances. What is the position where A has a registration? B would be forced to follow the honest concurrent use route to registration. The limitations this would entail could imply that a dynamic concern with the ability to do business in the whole of the country, would be restrained, unless the registration of A is limited geographically. Such an approach would indirectly involve a finding of geographical priority, however.
In terms of some systems, an opposer with limited prior use cannot oppose an application at all. Reference can be made here to the European Trade Mark Regulation Council Regulation 40/94/EEC, in particular article 8(4), read with article 52(1)(c), which deals with prior use as a basis for opposition or invalidation, respectively:

“Upon opposition by the proprietor of a non-registered trade mark or of another sign used in the course of trade of more than mere local significance, the trade mark applied for shall not be registered where, and to the extent that, pursuant to the law of the Member State governing that sign,
(a) rights to that sign were acquired prior to the date of application for registration of the Community trade mark, or of the date of priority claimed for the application for registration of the Community trade mark;

(b) that sign confers on its proprietor the right to prohibit the use of a subsequent mark.”
Article 8(4) read with article 52(1)(c) provides for the opposition of an application, or the expungement of a registration, on the basis of rights derived from the use of a mark which are of more than mere local significance. The following comment was made in regard to the latter concept in Compass Publishing BV v Compass Logistics Ltd [2004] RPC 809 825 – 826 (own emphasis):

“In other words, there is no reason to believe that Art. 8 (4) is primarily designed to protect the owner of the earlier unregistered right. He does not need such protection. If that is right, the primary purpose of Art. 8 (4) is to prevent the registration of or invalidate granted CTMs when there is an unregistered mark of sufficient width that it would significantly impair the ability of the CTM to deliver the Community-wide rights which the Regulation is seeking to encourage… In summary, it seems to me that a mark should be considered as having mere local significance within the meaning of Art. 8(4) if its geographical spread is restricted to substantially less than the whole of the European Union and that, from the perspective of the Community market in the services or goods in question, the mark is of little significance.”
It has been stated (Morcom et.al. The Modern Law of Trade Marks 532) in relation to the above provision, that the requirement that the use must be of more than mere local significance implies that the owners of small businesses will not be in a position to oppose Community applications.
The obtainment of a registration is, of course, not always the end of the matter, and there is a complicating factor that must be considered in this regard. This is namely the rights of a senior user in a particular area, who is a junior user in a national context. The following example can be used. A commences use in the Eastern Cape in 1995. B commences use in the Western Cape in 1998. A files in 2002. Received wisdom would be that B cannot oppose A’s application as A has prior use. He might have a claim to honest concurrent use later on. In this regard it is necessary to consider the Sidewalk Cafes (Pty) Ltd t/a Diggers Grill v Diggers Steakhouse (Pty) Ltd 1990 (1) SA 192 (T) case, which dealt with an expungement application. The facts briefly are that A had used his mark, in the then Transvaal, since 1982, and B commenced use in Natal in 1984. At the stage that A filed an application later that year, B already had substantial use in that province. The registration of A was rectified to exclude the province of Natal.
The court accepted, without deciding, that B would have been entitled to relief in terms of section 17(2) of the previous Act, which deals with concurrent rights. But the court seems to have treated the dispute more as one relating to priority. It was stated (page 198 I) for instance, that:

“In my view [B] became proprietor of the trade mark, at least in Natal…”
Furthermore, it was held (page 197 I, own emphasis) that:

“Mr Ashton submitted, and I agree with him, that because of [B’s] vested interest in the trade mark, described above, [A] was not entitled to claim exclusive proprietorship in the whole country, and to obtain exclusive rights to the trade mark and thereby depriving [B] of his vested rights in the trade mark.”
In the process of coming to an understanding of the approach followed by the court, it is important to have regard to the fact that the court referred to the Broadway Pen Corporation v Wechsler & Co (Pty) Ltd 1963 (4) SA 434 (T) and Oils International (Pty) Ltd v Wm Penn Oils Ltd decisions. These cases revolve around the idea of an exclusive, not concurrent proprietor of a mark. The conclusion is unavoidable. The court in effect held that the use that B had, at the date of A’s application, provided him with rights to the province of Natal. Put differently, A’s prima facie right to a national registration, being the national senior user, was limited by the rights of the senior user in Natal. This fact provides support for the statement that the decision recognizes the concept of geographical priority, although it is apparently in conflict with the principle of qui prior est tempore potior est iure, which is generally accepted in our trade mark law. It can be argued that in terms of traditional principles, the court should have had regard, not to the position as at A’s date of application, but as at A’s date of first use.
If the principle of geographical priority is, however, seen as being part of our law, the obvious question that is raised is whether the position prevailing in opposition proceedings should not be adapted to accommodate this approach? The argument that may be used in the above regard relates to section 10(12). This section refers to the opposition of a mark whose use is likely to cause confusion. It would appear that in the particular area where B had use, confusion would, indeed, be likely if A were to use his mark there.
The concept of a national junior user having priority rights in a particular geographical area seems to find application in English law. Particular reference can be made to Saxon Trade Mark [2003] FSR 704 where the following was stated (page 716 paragraph 32, own emphasis) in relation to section 5(4)(a) of the Trade Marks Act, which provides for an opposition on the basis of prior user rights:

“Mr Foley appears to have construed the section as if it is only concerned with cases where the use of the mark by the proprietor starts after the use of the same or a similar mark by someone else. I do not think that this is what the section says. For the prohibition to bite, all that needs to be shown is that, at the time of the application to register, the normal use of the mark by the proprietor would be liable to be prevented by passing off proceedings brought by someone else. It may well be that in most cases this will only arise when the other party had commenced using his mark before the proprietor, but it is not inevitably so and the section does not require it to be so. The fact that the convenient title “proprietor of an earlier [right] is used to designate the other party does not limit the scope of the section.”
The court then gave the following example:

“Consider, for example, a case in which one proprietor uses a mark on a retail clothing business in Manchester and the other uses it on a similar business in Plymouth. They commence trading at the same time. Their trades do not compete because of geographical separation. Assume the Manchester trader registers the mark. Normal and fair use of it would include use in Plymouth. That would be liable to give rise to a cause of action in passing off…For that reason the Manchester trader would fall foul of s. 5(4)(a) even though the Plymouth trader commenced use of the mark at the same time. For that same reason the Plymouth trader could not register the mark.”
It seems that the position is that where A has use prior to B’s application, he can oppose B’s application. Where a party does not have prior use (nationally) can it still, on the strength of the Saxon case, oppose another party’s application insofar as it relates to its area of use? For argument’s sake, B commences use, on a national scale, after A, but B has use prior to A in the area concerned. B can prevent use of the mark by A on the basis of passing off in the particular area, and, hence, oppose the application. This would recognize the principle of geographical priority.
Whether this approach is correct, is not clear. For one, it would be out of kilter with the approach adopted in section 11(3), although it is conceded that the wording of the sections differ. Section 11(3) also relates to the entitlement to relief on the basis of passing off, but it is linked to the earlier of the other party’s date of first use, or filing date. In other words, the fact that, in the above example, B commenced use after A could be seen as disqualifying him from opposing A’s mark.
In South Africa the principle of geographical priority does, however, not find general application. The position will only change if the decision in the Diggers Grill case is accepted as a principle of priority that can be reconciled with the general tenets of South African trade mark law.
Concurrent user rights
In the United States, section 2(d) of the Lanham Act allows a concurrent registration in the event of concurrent lawful use prior to the filing date of an application or registration. This relatively clear principle has caused considerable difficulty in its implementation, in particular in relation to geographical disputes. Several important principles were stated in the leading decision of In re Beatrice Foods Co.,. 429 F. 2d 466 (CCPA 2003). Firstly, the general rule is that the prior user of a mark is prima facie entitled to a registration covering the entire country. The second principle stated by the court is that where the prior user does not apply for registration before registration is granted to another, there may be valid grounds, based on a policy of rewarding those who first seek federal registration, for permitting the prior registrant to retain the nationwide protection of the Act, restricted only by the territory of the prior user, and for limiting the prior user’s registration to the area of actual use. Thirdly, the court stated that in regard to the two above scenarios where prior use is protected in a given area, it may be possible for a party to obtain protection for areas wider than its actual use. It is now necessary to deal with the following three aspects.
Firstly, with regard to the date for determining use, the position in the United States is that, as indicated, section 2(d) of the Lanham Act allows a concurrent registration in the event of concurrent lawful use prior to the filing date of an application or registration. The application of this provision can be illustrated with reference to the following example. A commences use in 1995, and obtains federal registration in 2000, with a filing date of 1997. B commences use in 1996. He will be able to obtain concurrent registration. Actual use prior to the applicant or registrant’s date of first use will not necessarily be of assistance. This is the situation where A files an intent-to-use application, and B commences use thereafter. After B’s date of first use, A commences use. B has prior use, but cannot rely on section 2(d). The decisive date is thus the date of filing. Any use thereafter cannot form the basis for a registration. English and South African law follow a different approach. In these countries the important date for purposes of honest concurrent use is the own filing date, as the sufficiency of use is judged retrospectively from that date. In both countries the provisions relating to honest concurrent use are based on the approach that rights are established after the date on which a priority right is vested in the opponent. For instance, in the case of section 10(14) and 14(1) of the South African Act, the application is per definition filed after the filing date of the registration, yet that is explicitly “condoned” by the legislature. Under American law this would not be possible.
It is, secondly, appropriate to make some brief comments on the difference between American and South African law insofar as the In re Beatrice Foods Co., decision is concerned. This ruling is to the effect that the prior user of a mark is prima facie entitled to a registration covering the whole United States. This would seem to be true for South Africa, subject to the qualification that a prior application can, of course, trump use that commenced after the filing date. In the decision it was also said that there may be valid grounds, based on a policy of rewarding those who first seek registration, for permitting the prior registrant to retain the nationwide protection of the Act, limited only by the prior user’s area of actual use. In South Africa, it would seem that the fact of prior use would not, in itself, give priority over an earlier applicant, that is, an application made prior to the date of first use of the former. The broader meaning attached to the concept of priority in South Africa thus becomes evident from this comparison.
In the United States, the Weiner King and Nark cases granted “priority” to the junior but extensive user in preference to the static senior user. This was in the context of concurrent use proceedings. The mechanism that would be used in South Africa, section 14(1), does not provide a basis for the same ruling. Section 14(1) presupposes an existing rights holder, whose rights will not be diminished by the later “concurrent” application. In other words, an “add on” system. In contrast, American concurrent registration proceedings in reality mask a priority dispute, and a finding as to which party’s registration should have the most extensive geographical reach. In this situation the national senior user’s rights can, indeed, be diminished. The approach of South African law is thus different, and a registration can only be limited in the context of expungement proceedings.
Thirdly, the issue of knowledge must be considered. The substantive law regarding the meaning of the concept “honest concurrent use,” is in general the same under the previous and current British and South African legislation. Insofar as knowledge of a mark is concerned, the position is that knowledge is an important factor to be considered when balancing considerations for and against registration, but it is not per se exclusionary. It would seem to follow, however, that, on balance, it is still best to create a situation where the infringing party cannot raise the issue of lack of knowledge for whatever purpose.
The first and obvious advantage of informing the infringer of the rights of the proprietor, and of the proprietor’s unambiguous objection to the infringing use, is to remove the possibility of the infringer relying on the special defences of acquiescence and estoppel.
The second advantage must be understood in the context of the practical and logistical facets involved in obtaining registration of a mark, having particular regard to the position in South Africa. The decision in the Diggers Grill case, which is of central importance to the subject of this research, must be considered here again. This case establishes that litigation, as soon as a registration ensues, may not always be sufficient. Prior to registration, A could not succeed with passing off proceedings in Natal, as his reputation was limited to the Transvaal. The common law did thus not provide assistance. In such circumstances a company’s trade mark attorneys would no doubt advise the party to immediately file a trade mark application to overcome this impediment. During this time he would, however, be without a remedy, and depending on the time that the mark takes to be registered, it could imply that the potential defendant could in the interim build up rights of such an extent as to be able to rely on honest concurrent use.
Having regard to the facts of the matter, it appears that up to the stage that A’s attorneys wrote to B, the latter’s use would undoubtly have been honest, as there was no awareness of A’s mark at the stage of the adoption thereof; B’s mark was also selected without reference to A’s mark; and there was the bona fide belief, for argument’s sake, following the obtainment of legal advice, that he was not infringing A’s rights. Consequently, if honesty is seen to relate to knowledge of a mark, and the reason for the choice of a mark, B’s position would be strong.
The letter of demand sent by A’s attorneys changed the position as far as knowledge of A’s mark was concerned. The question that must be considered here is the relevance of the fact that the letter did not mention the trade mark application filed by A, a fact to which the court attached weight (page 196 I). The implication probably was that B was not aware of the (potential) rights that A would obtain, in particular their nationwide scope. In the absence of such knowledge, it does seem reasonable, from B’s perspective, to then have continued to use his mark. In contrast, if the letter raised the spectre of statutory infringement, he may have elected to discontinue use.
Received wisdom in practice is, indeed, not to mention a pending application, for fear that the defendant will oppose the application. It may be necessary, however, having regard to the Diggers Grill decision, to revisit this approach. The better view seems to be that a comprehensive letter of demand should be sent as soon as a party becomes aware of another’s use, so that, amongst others, the principle of volenti non fit iniuria can apply. It is thus submitted that if, after a detailed notification of rights, imparting knowledge of a trade mark application that will mature into registration, a defendant elects to continue to use the mark, he must be taken to have assumed the risk that a court may rule that the use that followed the letter cannot found a claim for registration on the basis of honest concurrent use. There would then be use in the face of strong opposition which makes it clear that, although pro tempore, the dominant rights holder would not act against the defendant, he would as soon as he has a legal basis to do so. In particular, the principles contained in the Bali Trade Mark (no. 2) 1978] FSR 193 case would protect the dominant rights holder so that the defendant cannot, if an opposition is filed, rely on use during the period of opposition in support of a later application by it on the basis of honest concurrent use.
In given instances, if the use of the junior user has already been extensive, a letter of demand can, of course, not change matters. This was the situation in the Diggers Grill case, where the court considered the relevant date to be A’s application date, not his date of first use. The situation from this perspective is that it is always advisable to file an application as soon as possible, so as to “arrest” the establishment of rights. In the court’s interpretation, B had, at the stage that A filed his application, use of four months, moreover, such use was on an extensive basis. Therefore, if A filed an application in June, that is a month after becoming aware of B’s use, the ruling might have been in favour of A, as B’s use would not have established sufficient rights. By way of repetition, if A’s date of first use was taken into account, A would have had priority of more than 19 months.
It is, however, necessary to sound a word of caution. In practice it frequently happens that some time elapses before action is taken. In fact, it would appear to be the normal type of situation that practitioners encounter. The system of enforcement of trade mark rights would be thrown in turmoil if every business that has used a mark in a particular geographical area would be able to attack the registration of the proprietor on the basis of rights in that area. To this extent, it may perhaps be necessary to see the Diggers Grill decision to be of a sui generis nature.
Prior use defence
In the United States, section 33(b)(5) of the Lanham Act provides protection for use prior to the filing date of a mark, if the mark has been used without knowledge of the proprietor’s use, prior to the date of constructive use or registration date, on condition, however, that the right to use the mark shall only apply in the area of prior use. It is not of relevance, in the context of section 33(b)(5), that the junior user may be in a position to obtain its own concurrent federal registration in proceedings before the Patent and Trademark Office. If a (national) junior user has common law rights in a particular area, it can only object to the senior user’s registration in concurrent use proceedings before the Trademark Trial and Appeal Board, or in a court action.
It was held, in Tie Rack Enterprises, Inc. v Tie Rak Stores, 168 USPQ 441 (TTAB 2003) that the defence contained in section 33(b)(5) cannot have effect in a territory where the junior user commenced use after the date of the senior user’s registration. After registration, the junior user can thus not extend the area of his use. Although the wording of section 33(b)(5) refers to “registration”, it has been held in Thrifty Rent-A-Car System, Inc v Thrift Cars, Inc., 4 USPQ2d 1709 (1st Cir. 1987) that the date of application is decisive. After this date, the zone of permissible use is frozen.
The position, normaliter, would be that the senior user is the federal registrant. However, the registrant can in fact be the junior user. If the registration concerned is contestable, the senior user can challenge the registrant’s exclusive rights in the territory of use. If the registration has become incontestable, reliance can be placed on the wording of section 15, which delineates exceptions to incontestability, insofar as the acquisition of valid rights are concerned.
The 1994 British Act maintains protection for a prior user in section 11(3). This section provides protection on the same basis as that in the previous Act, in the sense that use must predate the registrant’s filing date or date of first use, whichever is the earlier. This is termed an earlier right, and is protected in the case of use in a particular “locality,” if use in that locality can be prevented by, in particular, passing off.
Consideration was also given to the issue as to when the particular earlier right must exist. In terms of its wording, section 11(3) appears to require merely that the earlier right should apply at the time of the infringement and that for the section to apply, it need only be protected at that time. But it would appear to be more correct if the approach is linked to the definition of earlier right so as to qualify it as being a right recognised by law. Therefore, in the context of the provision as a whole, it is submitted that the goodwill must have been established at the earlier of the application date or the date of first use by the proprietor.
Section 36(1) of the current South African Act provides protection for a user who has used a mark prior to the earliest of the registrant’s date of first use, or filing date, whichever is the earlier. The application of the provision to infringement in terms of section 34(1)(a) of the Act can require a consideration of the scope of protection the prior user would enjoy. It seems that there is some support for the approach of granting protection wider than the goods in relation to which the mark was actually used. In relation to section 34(1)(b), it was suggested that prior use must have taken place in relation to both the same or similar goods. In the case of section 34(1)(c), it was argued that the concept of prior use does not essentially find application.
In the case of the protection of prior user rights relating to a well-known marks, the courts have rejected the defence in circumstances where it was considered that the purpose of the acquisition of a business was not to distinguish the business from that of others, but the converse, namely to use a mark confusingly similar to that of the proprietor of the well-known mark. It was argued that transactions where companies purchase whole businesses, including their trade mark, or purchase only a particular trade mark are entered into on a regular basis. To distinguish when the defence of prior use will not succeed is, on a practical level, thus a difficult task. However, the approach of the courts may be aligned with general principles of unlawful competition which attach considerable importance to the motive of a competitor.
In the case of the prior user that relies on honest concurrent use, section 36(1) states that nothing in the Act shall be interpreted to allow the proprietor (if such use is proven) to object to registration in terms of section 14(1). The proprietor would thus seemingly be able to challenge only the veracity of a claim to prior use by the applicant. In terms of section 36(2), specific provision is made for well-known marks in terms of the Paris Convention.
Insofar as the relevant date is concerned, the most striking difference between American, English and South African law is the date for the determination of the prior user’s right to continue use. Section 33(b)(5) of the Lanham Act provides that use prior to the filing date concerned is protected. The following example can be utilised. A commences use in 1990 in Texas. B commences use in 2003, in California. In 2004, A files a federal application. B will be able to rely on the prior use defence. In other words, although B commenced use after A did, B would be entitled to protection. The inference that can be drawn is that American law places a high premium on the filing of federal trade marks. In other words, if a filing is made as soon as the mark is to be introduced to the market, it will prevent a regional senior, but national junior user, from relying on section 33(b)(5).
The importance of knowledge in the American system must always be borne in mind. This is apparent from decisions such as that in United We Stand America v United We Stand New York. 128 F3d 86 (2nd Cir. 1997). Here the slogan “United We Stand America” was used by A in a presidential election campaign in August 1992. An application for it was also filed. B broke away from the campaign and commenced use of the slogan in October 1992. Insofar as the relevance of B’s use prior to A’s filing date is concerned, the court stated (page 93, own emphasis) the following:

“The owner’s right to bar infringing use does not depend on registration prior to the infringer’s use. As [A] was the first user of the Mark, and [B] was well aware of [A’s] prior use before [B] used it, the date of [A’s] registration is irrelevant to the dispute.”
In this instance, B had use prior to A’s filing date, but the fact of knowledge of A’s mark excluded protection. If the fact of knowledge is removed from the equation, it would appear not to be correct to state that the fact of use by B prior to A’s registration is in effect irrelevant. This pronouncement, as a general proposition, is not in line with the wording of section 33(b)(5), which does make use prior to the filing date relevant.
In contrast to the above position in the United States, the British and South African legislation, in sections 11(3) and 36(1) respectively, require use prior to the earliest of the registrant’s filing date or date of first use. Mere use prior to the filing date of the registration is thus not sufficient. Seen from this perspective, American law is thus more lenient towards the junior user. British and South African law is focused, on the other hand, on the senior user, and is in fact more strict towards the claimed senior user. In other words, to obtain protection, he would have to prove use prior to the filing date of the registrant, and he would also have to prove use prior to the date of first use, something that is not required of the American “prior” user.
Substantial use is required to rely on the defence of prior use in South Africa, and the decision in Williams Ltd v Massey Ltd. [1911] RPC 512 which is to this effect, will apply. This principle can be applied from a different perspective in section 36(1). The latter section, of course, refers to the use of the apparent prior user. The mechanism for determining the ability of A, the alleged infringer and claimed prior user, to have a defence involves, amongst others, the date of first use of the registrant, B. The situation that can be envisaged is where B commences use on a modest scale, not amounting to substantial use. On the face of it, B is, however, the prior user. At a later stage A commences use on a substantial scale. B thereafter files an application. It would seem to follow that if B’s use has not been substantial, A would not be called upon to present a defence as, in view of the lack of sufficient prior use by B, the registrant, the only relevant date on the table is B’s date of filing, which postdates A’s date of first use. It is submitted that there is no valid reason to differentiate between the quantum of use required for a plaintiff or defendant.
In the context of the current discussion, the issue of geographical priority, which is, perhaps, the most problematic and pervasive issue dealt with in this work, surfaces again. Under American law, the common law position will apply, namely that the mark must have been used in a remote area, in good faith (see Alberts Thesis 46)?????. As is clear from the above, the common law envisages that the senior user will not be able to prevent the use of his trade mark in all parts of the country. In the context of registered rights, the position would appear to be different. Section 33(b)(5) namely states that the registration of a mark brings about exclusive rights to the use of the mark. From this it seems that the proprietor may indeed protect his mark in the whole country. Section 33(b)(5), however, protects prior use in the stated circumstances, which would imply that the effect of a registration is not necessarily to provide protection for the whole United States.
The position in English law is that prior user rights are protected in a particular “locality”. It was, however, submitted by writers that the section would also include a user on a national scale. Section 11(3) refers to the protection of an earlier right in a locality, if, or to the extent that, its use in that locality is protected by virtue of any rule of law, in particular, the law of passing-off. The first question that must thus be answered is whether the person in fact has prior user. Only then does the second question arise, whether the person has established the requirements for passing off.
The Saxon case has seemingly accepted the principle of geographical priority, namely the right to oppose even if the use does not predate the filing date of the application being opposed. This would be a requirement in terms of section 11(3), and an anterior question. In the Saxon case, the court seemed to focus only on the passing off position, and not the underlying condition of use prior to the opponent’s date of application or date of first use, whichever is the earlier. In this sense the requirements for the prior use defence and an opposition would not be the same.
The requirement for priority in an opposition matter is usually taken to be use prior to the date of application or first use of applicant, whichever is the earlier. As was stated above, this formulation is echoed by section 36(1) of the South African Act. However, it was noted that the Diggers Grill decision could be interpreted as recognizing the principle of geographical priority. In other words, if B uses his mark in an area prior to A, the registrant, B would have rights, and could oppose the application. In the case of section 36(1), there appears to be no scope for such an interpretation. The registration itself would have to be attacked, or the particular area excised, depending on the factual circumstances, for instance, where there is limited use in a small area. It is appropriate to remark as this stage that in practice an attack on a registration with the aim of obtaining a geographical delimitation is fairly rare, and it would appear that the courts would prefer not to follow this route. Clients are accordingly usually advised to follow the strategy of applying for registration on the basis of honest concurrent use. In such an event, a more flexible approach would be followed, and the use made by the applicant for registration, prior to his application, not prior to the date of first use or registration of the registrant, would be relevant.
Lastly, the question must be considered whether bona fide prior use, adopting a South African perspective, would necessarily provide a licence to use the mark vis-à-vis the later registrant in the whole of the country. This raises to an extent the question of the territorial nature of domestic trade mark rights. With regard to the common law, in particular passing off, it is well-established that protection for the prior user will be limited to the actual area within which a reputation is established. This is evident from decisions such as Scott and Leisure Research and Design (Pty) Ltd v Watermaid (Pty) Ltd 1985 (1) SA 211 (C), wherein it was held that a trader could have a reputation in a particular part of the country, but not in another. The retention of (only) common law rights have in fact been recognised by the courts. In Nino’s Italian Coffee & Sandwich Bar CC v Nino’s Coffee Bar & Restaurant CC 1998 (3) SA 656 (C) the court, for instance, stated (the view is seemingly based on a passage (page 12-79) in Webster and Page South African Law of Trade Marks (1997)) the following (page 675 H, own emphasis) in relation to section 36(1):

“The underlying purpose of this section is to prevent a proprietor of a trade mark from exercising his rights merely on the basis of priority of registration and it preserves whatever common-law rights there may be antecedent to the rights of the registered proprietor.”
Common law rights are thus protected. Such rights are typically related to the particular area where a reputation exists. It would seem to follow that the prior use by A in the relevant area of use would be preserved, but there would not automatically be a blanket licence to extend use. The fact of prior use by A does also not, ipso facto, invalidate B’s registration. Normaliter, A would thus have to file an expungement application to remove the entry from the Register. If the entry is not removed, it is obviously valid, and A’s position would thus hinge on the status quo at the relevant date. This result may seem incorrect, but not necessarily in the situation where A has used his mark on a relatively small scale, only in one particular location, and B has more extensive use in a number of geographical locations.
What is the relevant date for determining A’s right to use? If the issue is seen to be the prevention of priority merely on the basis of a registration, B’s filing date would, strictly speaking, not be relevant, and from this perspective, the cut-off date for expansion by A, and the date after which such use would be infringement, is B’s date of first use. It is submitted, however, that the situation must rather be judged on the basis of general principles of priority, in terms of which either a filing date or date of first use, can be relevant. Put differently, in regard to A’s use prior to B’s filing date or B’s date of first use, A has “absolute” priority in terms of section 36(1). In relation to use thereafter, A would be in the same position as any other party, and extended geographical use by A after B’s priority date, in a broad sense, would amount to infringement, as it would not be protected by section 36(1). The guiding principle should be the extent of A’s reputation, in a geographical sense, at the relevant date.
It should be pointed out that even if an expungement application is filed, which would be in terms of section 10(12) read with section 24(1), there is the possibility, on the strength of the Diggers Grill case, that a court will carve out from the registration only the particular geographical area where the prior user has been trading.
It is noteworthy that it is determined in the United Kingdom, by section 11(3) of the Trade Marks Act, that a registered mark is not infringed by the use of a mark by a local prior user in the area where the latter has built up rights. The effect of section 11(3) in this regard has been stated (Morcom et.al. Modern Law of Trade Marks 216, own emphasis) to be the following:

“It is fairly clear that the owner of the ‘earlier right’ would not be permitted to expand the area of use outside the particular locality, and still avoid infringement”.
Insofar as the United States is concerned, it was held in Tie Rack Enterprises, Inc. v Tie Rak Stores, that the defence contained in section 33(b)(5) cannot have effect in a territory where the junior user commenced use after the date of the senior user’s registration. After registration, the junior user can thus not extend the area of his use. In this regard the provisions of section 22, which provides for constructive knowledge of a registration, is relevant. This section has the effect of freezing the zone of permissible use of the junior user.
It is of interest, lastly, to have regard to Australian legislation. Section 124(1) of the Trade Marks Act 1995 provides protection on the same general priority basis as the South African Act’s section 36(1). However, it is specifically provided that if the mark concerned has been used only in a particular area of Australia, the protection shall apply only in that area.
Various systems thus seem to adopt the approach that protection for the prior user, or user prior to the filing date concerned, is limited to the actual area of use at the relevant date.
Summary
In conclusion, it appears that the first use of a mark will be relevant in opposition proceedings in the United States, except in the case where an intent-to-use application has been filed. In the United Kingdom and South Africa, the date of first use can be trumped by a filing date that is earlier than the date of first use. In the United States, in the case of concurrent use, use prior to the other party’s application date is relevant, whilst in the United Kingdom and South Africa, use after the other party’s application date while still suffice. In the event of prior use as a defence, the American position is that use prior to the filing date concerned is relevant. In the British and South African system use prior to the earliest of the registration’s application date, or date of first use, is required.