[SpicyIP] - When Orders don’t Speak: Analysing Tapas Chatterjee vs Controller

[Bharathwaj Ramakrishnan]

The recent judgment of the Division Bench in Tapas Chatterjee v. Assistant Controller of Patents and Designs (Tapas) had caught the headlines for taking the Indian Patent Office (IPO) to task for issuing an unreasoned order or in the words of the Court a ‘laconic’ order (‘Laconic’ means –using or involving the use of a minimum of words: concise to the point of seeming rude or mysterious).  The word itself has an interesting etymology tracing its origins to the land of Laconia, which was once ruled by the Spartans, known for their terse and concise (laconic) speech.

Regular readers and keen followers of Indian patent law will be aware that this is not the first time the IPO has been taken to task by High Courts for issuing unreasoned or inadequately reasoned or ‘laconic’ orders. The Division Bench’s judgment is not new in that regard, yet it gains force from the fact that the slap on the wrist is now coming from the Division Bench of a High Court.

This post will first focus on the judgment, its reasons to set-aside the SB and the controller’s order, DB’s observations on Section 3(d), and finally on the crucial temporal gap that is possibly present in our current understanding of unreasoned orders.

A Brief Overview of the Invention:

The invention in question was a process for extracting Potassium Sulphate and other valuable products from Distillery Spent Wash, ensuring a Zero Liquid Discharge (ZLD). Spent Wash is a waste product produced by distilleries that manufacture alcohol. This untreated waste product/affluent, although environmentally harmful, is also potentially a rich source of nutrients that could be processed and utilised for plants, agriculture, and other useful materials (For more on this, see here and here). The invention, as disclosed in the patent application, was directed towards a process for “recovery of potassium sulphate, magnesium sulphate, activated carbon and other value-added products that can be used either as fuel or as cattle feed from molasses-based alcohol distillery effluent.”

Process as claimed in claim 1

What’s the Lis?

Two questions arose before the DB while examining the patent application:

1. Was the process as claimed in the patent application a “known process” and hence barred by  Section 3(d)?
2. Did the claimed process lack inventive step in light of the prior arts cited?

The controller had to examine 4 prior arts and a handful of objections raised by ICMR in their pre-grant proceedings. In the end, the controller concluded that the patent application was hit by Section 3(d) and lacked inventive step without clearly articulating the reasons why he thought the same. 

The Single Bench, as the DB had noted, did not remand the case back to the controller (a normal course of such appeals) but decided to examine the matter on its own. SB, in a more detailed examination of the issue at hand, affirmed the position of the controller, concluding that the process as claimed in claim 1 was a set of combination of known processes based on prior art and hence hit by Section 3d. Secondly, SB also concluded that the invention disclosed lacks inventive step in light of the prior arts D1 and D2. Not surprising since the SB had concluded that the entire process, as claimed in claim 1, was a combination of known processes. 

Division Bench Intervenes:

The DB issued a detailed order disagreeing with the Single Bench and criticising the controller for not providing reasons in the first place, while acknowledging the Single Judge’s effort to supply the reasoning the controller omitted. The DB still disagreed with both the approach and the conclusions on questions of fact and law reached by the SB.

1. DB, on a detailed analysis of the prior arts with respect to the subject invention, noted differences in the process employed in the subject invention and the processes disclosed in prior arts. 
2. Since the patent in question was a process patent, these differences became crucial. 
3. The DB reasoned that the controller, by rejecting the novelty objection raised by the opponent, had accepted the premise that the process disclosed in the subject invention is distinct from the processes disclosed in the prior art. 
4. Hence, the controller following from point 3 must have identified the differences present in the subject invention with respect to the prior art. 
5. Following point 4, he must have shown how those differences would have been obvious to a person skilled in the art in light of the prior arts cited and general technical knowledge available before the priority date. Yet DB concluded he failed to do so. 
6. Single Bench sought to remedy the situation by conducting its own analysis on points 4 and 5, but DB disagreed with SB’s finding of fact (with respect to the difference between the prior art and the subject invention). Further concluding that the Single Bench incorrectly applied the inventive step test provided in Roche v. Cipla (Roche) by starting the analysis from Step 4 of a 5-step test without explaining why. 
7. DB disagreed on applying Section 3(d) to the facts of the case, stating that known process = same or substantially similar process, and went as far as saying that a person skilled in the art is a  “stranger” with respect to Section 3(d) and finding the conclusions of the controller on Section 3(d) as “mechanical”. 
8. Finally, DB, following its findings, decided in favour of Tapas Chatterjee on the Section 3(d) objection and remanded the matter back to the IPO to decide on the question of inventive step by applying the 5-step test provided in Roche.

Section 3(d) and identifying “known process”

I wish to highlight an observation of DB with respect to Section 3(d), briefly discussed in point 7 above. In para 22.5, DB observed that a known process would mean that the process disclosed in the prior art and the process claimed in the patent application to be the same or substantially similar. It further claimed that PSITA is persona non grata with respect to Section 3(d).

DB’s position on Section 3(d) and “known process” is a bit puzzling to me for the following reasons:

1. The judgment seeks to confine Section 3(d) to a strict identity test, stating that the earlier and later processes “must be the same,”. Yet in the same paragraph, it acknowledges that processes can be “substantially similar,” permitting differences between the claimed process and the process disclosed in the prior art.
2. DB further clarifies that PSITA is persona non grata in the context of Section 3(d). This observation is puzzling, as a patent application is a techno-legal document addressed to PSITA. Any determination of whether the process claimed is a “known process” has to be from the perspective of PSITA. If PSITA is absent in a Section 3(d) analysis, how would the prior art be construed to identify the process in the prior art and compare it with the claimed process?
3. Finally, is the DB reducing Section 3(d) known process analysis to a novelty test by limiting the analysis to whether the process is ‘known’ from a single prior art, if yes, why?

Temporal Gap and Unreasoned Orders:

IPO has been criticised previously for issuing unreasoned orders. In some of my previous posts, I have discussed how structural issues, such as performance metrics, training and other institutional factors, create conditions for issuing such unreasoned orders. We have also discussed simmering internal disputes and controversies (see here and here) within the IPO, which might prevent the IPO from effectively tackling these issues. Yet, the judgment of the DB highlights a major but overlooked issue: a temporal gap associated with our discussion surrounding unreasoned orders.

It must be kept in mind that the impugned order in this case was issued on 20 December 2021—nearly four years ago, with the SB delivering judgment in March 2023 (15 months later). Thus, such orders are brought to light only after an appeal to the High Court and are only made visible to us after a time lag, thus possibly revealing a temporal gap.

This is akin to observing the light of distant stars in faraway galaxies: the light carries valuable information about the stars and their planetary systems, yet it reflects a state that existed long ago. In a similar manner, judicial decisions expose instances of unreasoned orders that may be the outcome of institutional practices from the past rather than the present. This temporal gap underscores a structural limitation—we lack mechanisms to obtain a contemporaneous view of how the Indian Patent Office issues orders, leaving much of its functioning to be understood only in retrospect, after a judicial decision. This is also problematic in another aspect: not all orders are appealed, leaving us with a narrower sample that only indicates the existence of the problem, rather than its scope or extent. It also highlights how orders of the IPO do not receive as much scrutiny as decisions of a Court. Orders are hence issued on a regular basis by the IPO with minimal scrutiny and discussion among the public. 

Thanks to Swaraj and Praharsh for this input on the post!

Please click here to view the post on SpicyIP and leave a comment.