Ajanta Pharma v. Theon Pharmaceuticals #TM Infringement

I write to discuss a recent Bombay HC Justice G.S. Patel Order on trade mark infringement.

Though the core reasoning involved is rather straightforward, a few of the questions and arguments raised deserve our attention. Though these aspects might not have been relevant to the final result in the current fact scenario, they are very much relevant in the larger scheme of jurisprudence.

Prior to delving into the trickier aspects of the order, let us set aside the less convoluted portion (though more fundamental to the final result) of the reasoning…

Basic Relevant Facts

Ajanta registered a mark called “FERANTA”.

Intas used a mark called “INTAS FERINTAS”.

Ajanta claimed trade mark infringement and sought an injunction order.

Court held in favour of Intas.

Ratio Deconstruction 

The basic analysis, as stated earlier, is quite straightforward.

J. Patel held that the marks, in his eyes, weren’t “deceptively similar” and hence, no other question was required to be considered.

Now, one of the arguments put forth by the defendants (Intas), to show absence of infringement, attempted to show that the mark was named in accordance with standard industry practice.

The standard industry practice of naming products, in the current case, involved combining a key ingredient’s name with that of the corporation.  For example, FER (Iron) + INTAS.

J.Patel, while not emphasizing on this argument, does not invalidate it either. He posits it as an argument of the defendant and though he does not explicitly accept it, the nature of its treatment seems to suggest he did (Para 17).

I, on the other hand, am unsure about accepting this argument as a valid defence against infringement.

On one hand, I find the argument problematic because:

Irrespective of whether a particular product is named according to industry practice or otherwise, a “deceptively similar” mark should be disallowed on account of its likelihood of causing confusion among consumers.

Why should leeching off of goodwill be justified on the basis of “industry practice”?

It is to be noted that intention behind infringement in TM law is irrelevant and the irrelevance of the honesty in adoption (of the mark by the allegedly infringing party) was explicitly accepted as well (Para 23). The “industry practice” argument seems to be attempting to make intentions relevant and therefore, when the overall defence of honest adoption has been rendered irrelevant, I don’t see why a subset of the argument should be accepted.

But on the other hand, there might be some merit in it:

The argument could be seen as distinct from the intention. An argument could be made that as a consequence of the existence of a standard industry practice, consumers would be cognizant of the practice and would, as a result, be more careful while distinguishing products.

But please note that the above analysis is entirely mine and J. Patel hasn’t clarified upon the same.

Furthermore, please note that this was not the sole basis of holding in favour of intas. The judgment, very much, goes into a thorough analysis of comparing the two marks and provides sound reasoning for holding the marks to be sufficiently dissimilar. We need not go into these reasons as they are factual in nature and provide us with little insight into the law.

Finally, before we move onto the more, so to speak, directly inconsequential aspects of the judgment, we need to take note of another holding. It is kind of intuitive, but still worthy of a short note.

The excerpt puts it quite clearly:

“But I do not think it is possible to invert this paradigm, and suggest that because the goods are either the same or similar, therefore one must work backwards to conclude that the marks must be similar, or held to be similar, and therefore, infringement and passing off must both be presumed. Identity or similarity of marks must be the first test, not the assumption, and we proceed to the goods or services from there, not the other way around. That is self-evident.”

Also, it was explicitly stated that each prong of analysis is distinct from the other. So, if it is determined that the marks are similar, it has no consequence on the threshold for determining if the goods are similar. Each condition has be fulfilled independently and the analysis of one, does not affect the other.

Obiter Deconstruction

Despite the above mentioned holding, the judgment discussed the following also (and so will we!):

1. Distinction between 29(1) and S.29(2) of the TM Act.
2. For essential conditions of infringement to be fulfilled, does the infringing good have to fall within the broad class under which the (allegedly) infringed good has been registered, or does it also have to fall within the specific sub-classification as well?
3. Appropriateness of Judges as determinants of “likelihood of confusion”.

Distinction between S.29 (1) and S. 29 (2)

The defendant used an elaborate argument to suggest that for infringement to be proven, not only “similarity” in the marks had to be shown, but also “likelihood of confusion”. Note at the outset itself, that this an inconsequential argument as “similarity” alone was not established. But anyway, let us have a look.

Before moving onto his argument, let us have a quick look at the relevant provisions of S. 29:

“29. Infringement of registered trade marks.—

(1) A registered trade mark is infringed by a person who, not being a registered proprietor or a person using by way of permitted use, uses in the course of trade, a mark which is identical with, or deceptively similar to, the trade mark in relation to goods or services in respect of which the trade mark is registered and in such manner as to render the use of the mark likely to be taken as being used as a trade mark.

(2) A registered trade mark is infringed by a person who, not being a registered proprietor or a person using by way of permitted use, uses in the course of trade, a mark which because of—

(a) its identity with the registered trade mark and the similarity of the goods or services covered by such registered trade mark; or

(b) its similarity to the registered trade mark and the identity or similarity of the goods or services covered by such registered trade mark; or

(c) its identity with the registered trade mark and the identity of the goods or services covered by such registered trade mark,

is likely to cause confusion on the part of the public, or which is likely to have an association with the registered trade mark

… “

Venkateswaran on Trade Marks & Passing Off states that 29(1) lays down the general proposition of law, while 29(2)-29(9) merely enumerate upon the general proposition.

But the defendant argued for a different interpretation.

Here is the relevant excerpt:

“…

12. That takes us to what I think is the central defence. Mr. Dhond says this is not a case under Section 29(1) of the Trade Marks Act, 1999 at all, but more correctly under Section 29(2) and this makes all the difference. This is how he constructs the defence: the goods are admittedly different, and the slightest difference in formulation or prescribed use is sufficient. It is not possible to conflate uses because Class 05 has a wide sweep. Within that class, there are permissible differentiations, and if a party has only ever used its mark on, say, a pharmaceutical, and never on a food supplement, there is no question of confusion or its likelihood. Ajanta’s claim falls under Section 29(2)(b), says Mr. Dhond.

…

14. Section 29(1), Mr. Dhond submits, is mark-centric. It speaks of marks that are identical, or a rival mark that is ‘deceptively similar’, and, by definition, ‘deceptive similarity’ takes within its sweep the likelihood of confusion. This was, indeed, the state of the earlier law before the 1999 Act. Section 29(2) is a later evolution of the law. Its concern is effect-centric. It speaks not of deceptive similarity but of the likelihood of confusion and it contemplates three situations: (1) identity of marks plus similarity of goods and services; (2) similarity of marks plus similarity or identity of goods and services; and (3) identity of marks plus identity of services. Here, the two marks are not identical. That is not even Ajanta’s case. That eliminates 29(2)(a) and 29(2)(c), and leaves only 29(2)(b): where the marks are allegedly similar, and the goods are identical or similar. The test in 29(2) is not of deceptive similarity, but of likelihood of confusion; and, therefore, if this likelihood of confusion is not established, no injunction can be granted. Section 29(2) is a bulwark against unreasonable and unfair monopolization. In the simplest terms: in a Section 29(2) case, mere similarity of marks is insufficient. It must be accompanied by a similarity or identity of goods; and even that will not completely fulfil the requirements of Section 29(2), for a likelihood of confusion must be shown.
15. That construct from Mr. Dhond might be one that stops a little short. The ending words of Section 29(2) — “or which is likely to have an association with the registered trade mark” — probably speak to deception, and, therefore, read as a whole, the test might still be of deceptive similarity…”

I have quite a few problems with the counsel’s argument.

Firstly, the contention seems to suggest that the standard for determining “deceptive similarity” and “likelihood of confusion” are distinct, but I don’t see where this is coming from given that the definition under S.2(1)(h) of the TM Act effectively equates the two.

Secondly, the argument seems to be contradicting itself. On one hand, the counsel argues that “deceptive similarity” “takes within its sweep the likelihood of confusion” and on the other, argues that “the test in 29(2) is not of deceptive similarity, but of likelihood of confusion; and, therefore, if this likelihood of confusion is not established, no injunction can be granted.”

Unfortunately, J. Patel doesn’t deal with this entire analysis, despite calling it the “central defence”, on account of there being no “similarity” in the first place. So, the question of whether the additional standard of “likelihood of confusion” arises or not was not gone into. Nevertheless, he says that, “probably”, the standard in both the provisions is that of “deceptive similarity”.

I would have loved it if J. Patel could have clarified upon this though.

Matching of Overall Class Sufficient?

The question, as stated earlier, is this:

For essential conditions of infringement to be fulfilled, is it enough if the (allegedly) infringing good falls within the broad NICE Agreement class under which the (allegedly) infringed good has been registered, or does it also have to fall within the specific sub-classification as well?

Again, J. Patel doesn’t go too deep into this. Here is the excerpt:

“One must guard as much against an approach that is overbroad, and sweeps in all manner of goods easily distinguished, and one that is much too narrow”

The take away from this analysis would be that a straight jacket formula has not been preferred. A straightforward formula that could have been adopted was deeming the sub-condition for infringement to have been fulfilled if the allegedly infringing goods fell within the overall NICE Agreement class, but it wasn’t adopted and a more flexible approach seems to have been preferred.

Appropriateness of Judges as Determinants of “likelihood of confusion”

Again, he spoke about this only in passing.

He refers to the act of judges appropriating the voice of the consumers, in the context of adjudicating upon “likelihood of confusion”, as “fallible”, “unreliable”, “paternalistic” and “condescending”.

He talked about how, up until now, judges had resorted to using their own judgment for determining “likelihood of confusion” citing the lack of resources (time and otherwise) to collect data from the public, but now, he opined that the advent of modern methods have rendered the resource crunch nugatory and consequently, suggested imposition of a higher threshold upon plaintiffs.

Chief References

 Venkateswaran on Trade Marks & Passing Off, The Order.