top of page
  • Vincent Teh

Render unto Caesar...but which one?

Render unto Caesar that which are Caesar’s. But what happens when you are faced with two Caesars? That was the question that the Court of Appeal of Singapore had to answer in the recent case of Ceramiche Caesar SpA v Caesarstone Sdot-Yam Ltd [2017] SGCA 30, Court of Appeal, Singapore


The Respondent, Caesarstone Sdot-Yam Ltd (the “Respondent”), applied to register its trade mark “CAESARSTONE” in Singapore under Class 19 for various stone tiles and stone floor/wall coverings. The Respondent’s application was then opposed by Ceramiche Caesar SpA (the “Appellant”) who is the registered proprietor of the trade mark “CAESAR” for goods that are similar to the ones claimed by the Respondent’s application. The Trade Mark Registry (the “Registry”) allowed the opposition on the grounds of confusion under s.8(2)(b) of the Trade Marks Act (Cap 332, 2005 Rev Ed) and that the use of the Respondent’s mark would indicate a connection with the Appellant and its well-known mark under s.8(4)(a) and s.8(4)(b)(i). Unhappy with the Registry’s ruling, the Respondent appealed to the High Court. Finding for the Respondent, the Judge ruled that both grounds are unfounded and allowed the appeal in Caesarstone Sdot-YamLtd v Ceramiche Caesar SpA [2016] 2 SLR 1129.

The Appellant then brought the present appeal to the Court of Appeal (the “Court”). The Appellant submitted that:-

  1. the decision of the Registry ought not to be disturbed unless there has been a “material error of fact or law” (i.e. the appropriate threshold for appellate intervention), and in the present case the Registry did not make any material error in coming to its decision;

  2. the High Court has erred in finding that the grounds under s.8(2)(b) of the Trade Marks Act (i.e. likelihood of confusion) are not made out; and

  3. the High Court has erred in finding that the grounds under s.8(4)(a) read with s.8(4)(b)(i) of the Trade Marks Act (i.e. well-known mark) are not made out.

In relation the appropriate threshold for appellate intervention, the Appellant was of the view that the case law (e.g. Future Enterprises Pte Ltd v McDonald’s Corp [2007] 2 SLR(R) 845) and “normative grounds” support a high threshold for appellate intervention, i.e. the court should not intervene unless the Registrar has made a “material error of fact or law” in coming to its decision. The Court disagreed, even going as far as to call the decision of Future Enterprises (which is itself a Court of Appeal decision) ‘wrong’. The Court held that by virtue of O 87 r 4(2) of the Rules of Court (Cap 322, R 5, 2014 Rev Ed), which provides that any appeal from the Registry shall be ‘by way of rehearing’, the courts need not constrain itself to only intervening where there is a ‘material error of principle’ as stated in Future Enterprise. As for the “normative grounds”, the Court disagreed that any deference should be accorded to the Registrar’s decision. Although it is trite principle that that an appellate court should not second-guess the exercise of discretion by a lower court or tribunal, the nature of a trade mark opposition does not require the Registrar to exercise his discretion. While the Court accepted that there is some element of subjectivity in the Registrar’s decision (e.g. the determination of whether the marks are similar), such subjectivity does not amount to the exercise of discretion. Rather, what is involved is a matter of legal correctness and entitlement, which is well within the appellate court’s jurisdiction to question.

As regards the likelihood of confusion under s.8(2)(b) of the Trade Marks Act, the COURT found the marks “CAESAR” and “CAESARSTONE” to be visually, orally and conceptually similar. The court arrived at this decision because the word “caesar” is considered to be the dominant feature in both marks, and the addition of a simple device and the descriptive word “stone” in the Appellant’s mark did not do much to distinguish it from the Respondent’s. Unsurprisingly, the Court also found that the goods represented by both marks to be similar.

Having found the marks and the goods they represent to be similar, the Court then turned to consider the nature of the goods and their mode of purchase. It was found that the consumers of goods in Class 19 are generally indifferent towards the trade marks that are attached to these goods. Rather, the consumers tend to focus on the price and attributes of the products in making their purchase decisions. Accordingly, it is tempting to assume that confusion would not be an issue, since the trade marks did not play a role in the consumers’ purchase decision (which was the position of the High Court). However, the Court rejected this line of argument on 3 grounds.

Firstly, the Court held that just because a consumer is indifferent towards the mark that is attached to a particular product, it does not mean that there can be no likelihood of confusion. The Court gave a simple example to illustrate this point. Suppose a consumer had done a search on the internet for tiles that are of a particular price and attributes, having found the tiles that he wanted, he took note of the trade mark (e.g. CAESAR). He then visited the hardware store and was faced with the Respondent’s CAESARSTONE tiles. Due to imperfect recollection he mistook the Respondent’s tiles for the Appellant’s. Secondly, the main function of a trade mark is to allow the consumers to distinguish the product/services of one person from those of another. As such, the issue of confusion is really an extension of that function, i.e. a trademark that is confusingly similar to another is also incapable of differentiating itself from the other, and thus, must not be registered. The fact that the consumers are indifferent to the trade marks have no bearing on whether it can properly function as a trademark (and to be registered) which incidentally is the main contention of this opposition. Thirdly, for policy reasons, the finding of a likelihood of confusion cannot be excluded by virtue of consumer indifference towards the competing marks as it would undermine the entire trade mark regime.

Having rejected the above argument, the Court went on to find that due to consumer indifference towards the competing marks, the consumers would naturally pay less attention to the differences between the marks, thus, leading to a higher likelihood of confusion. Surprisingly, the Court also found that the fact that the goods in question are generally expensive and personal would not serve to lower the likelihood of confusion, even though the consumers are likely to exercise more care in buying these goods. Further, the Court also found that even as most consumers would consult specialists when purchasing the goods, this did not negate the likelihood of confusion. The Court reasoned that the indifference of the consumers to the trade marks attached to these goods will mean that they would only exercise more care in relation to the price and attributes of these goods, and similarly, their discussions with the specialists are likely to be centred around the price and attributes as well. Accordingly, the Court held that there exists a likelihood of confusion on the part of the public.

As for the grounds of s.8(4)(a) read with s.8(4)(b)(i) of the Trade Marks Act, the Court rejected the Appellant’s claims that its trade mark is well-known in Singapore. As such, the Court upheld the finding of the High Court in that the grounds of s.8(4)(a) read with s.8(4)(b)(i) of the Trade Marks Act are not made out.

In light of the above the Court allowed the appeal and the Respondent’s CAESARSTONE Mark was not allowed to proceed to registration.


It is clear from the Rules of Court that any appeal from the Registry would be a rehearing, and the courts need not constrain itself to only intervening where there is a ‘material error of principle’ (which is the usual position in many other jurisdictions). Although the Court did not explicitly overrule Future Enterprise, the present decision has done much to clarify the position of the law with regard to appropriate threshold for appeals from the Trade Mark Registry. The Court's decision on the issue of confusion vis-à-vis consumer indifference may seem counter intuitive at first. This is especially true since the products are expensive and usually bought by the consumers with specialists advice (both of which are normally considered to have a limiting effect on the likelihood of confusion). However, if one were to look at the facts of the case closely and bearing in mind the salient example of the proactive buyer as provided by the Court, then it becomes clear that the finding of confusion despite consumers indifference was a right one.

Practical significance

In cases relating to trade mark confusion, it is very common for one party to rely on consumer indifference towards the trade marks in order to refute the likelihood of confusion. However, this case has turn the whole argument on its head, as the Court found that consumer indifference has no bearing on the test for the likelihood of confusion. On the contrary, under certain circumstances, consumer indifference towards the marks is likely to increase the risks of confusion. The case also shows that there must not be a formulaic approach to the consideration of the likelihood of confusion. The fact that a product is highly priced and are often bought in the presence of specialist need not always point to a lower likelihood of confusion. Accordingly, this case is likely to change the ways that the issue of confusion is argued from now on.

54 views0 comments
bottom of page