The jurisprudence surrounding Section 2(1)(ja) of the Patents Act, 1970, which defines an “inventive step” as a feature involving a technical advance or economic significance that makes an invention not obvious to a person skilled in the art, has long been shaped by the F. Hoffmann-La Roche Ltd. vs Cipla Ltd. [2016 (65) PTC 1 (Del)] decision. The Roche test, now regarded as the gold standard by the Courts, lays down a structured five-step method to assess obviousness. Yet, as the recent Tapas Chatterjee vs Assistant Controller of Patents & Anr. [LPA 836/2023] decision reveals, the Patent Office often bypasses this structured evaluation, issuing unreasoned refusals that fail to explain how a “person skilled in the art” (PSITA) could have arrived at the claimed invention.

The Division Bench of the Delhi High Courtdismantled both the order passed by the Controller that refused the Appellant’s patent application and the Single Judge’s affirming judgment, exposing contradictions and lapses in how the inventive step was assessed.

The Appellant had filed patent application number 201911036748 for a process entitled “Recovery of Potassium Sulphate and other valuable products from Spent Wash leading to ZLD System”. The process sought to recover potassium sulphate and other by-products from distillery effluents, promoting zero liquid discharge (ZLD).

A pre-grant opposition was filed by CSIR under Section 25(1) on multiple grounds, including lack of novelty [25(1)(b)], lack of inventive step [25(1)(e)], non-patentability under Section 3(d) and 3(a) [25(1)(f)] and insufficiency of disclosure [25(1)(g)]. The Controller rejected the application, holding that while the invention was novel, it lacked an inventive step and was hit by Section 3(d). The Single Judge of the Delhi High Court upheld this order [in CA (Comm IPD-PAT) 18/2022], affirming the findings on both counts.

The Division Bench found the Controller’s reasoning “thoroughly unsatisfactory” and “a mockery” of adjudication. It observed that though the Controller had acknowledged novelty of the subject invention, since not all features of the invention were disclosed in the cited prior art documents D1 and D2, he paradoxically held that no inventive step was involved. This led to what the Court called an “inherent contradiction” in reasoning: “If there were features in the subject invention not disclosed either in D1 or D2, and the AC [Controller] nonetheless felt that no inventive step was involved, the AC was required to identify those features and demonstrate how they would not require a person skilled in the art to employ any inventive step. The AC has not done so”.

In effect, the Controller’s order failed to articulate the most fundamental part of the Section 2(1)(ja) inquiry: how exactly a PSITA would move from prior art to the claimed invention. Instead, the order merely concluded without explanation that such a transition would be “obvious”. The Bench condemned this as a clear case of non-application of mind, noting that the Controller had vaguely referred to “D1 alone, D2 alone, D1-D2 in combination, D3 in combination with D1 or D2 or both,” without any analytical exercise to show what combination, if any, would lead to the claimed process.

Notably, both the Single Judge and the Division Bench referred to the Roche decision, but the Division Bench found that the Single Judge had misapplied the very test he cited. As the Bench noted, Roche prescribes five sequential steps:

1. Identify the person skilled in the art.
2. Identify the inventive concept embodied in the patent.
3. Impute to the PSITA what was common general knowledge at the priority date.
4. Identify the differences between the cited matter and the alleged invention.
5. Decide whether those differences would have been obvious to the PSITA.

The Single Judge, however, began his analysis at Step 4, skipping the crucial identification of the PSITA and the inventive concept. The Division Bench found this approach legally untenable: “The learned Single Judge could not have commenced applying the principles in Hoffmann from Step 4. The error in starting from Step 4 is self-evident. It bypasses the person skilled in the art who, statutory and legally, is the person from whose point of view the aspect of inventive step and obviousness has to be determined”.

In doing so, the Single Judge’s method collapsed the very scaffolding of the obviousness inquiry. The Bench emphasized that identifying the PSITA and the inventive concept of both the prior art and the claimed invention is essential to determine the distance between them. Without that comparative framework, any finding of obviousness becomes speculative.

The Bench clarified that inventiveness may exist even in the manner in which a particular objective, or end, is achieved. Starting from the same reagent, the same final product may be achieved by two sets of processes, each inventive in its own way, if the processes are different, and one cannot be said to be “taught” by the other. At times, inventiveness resides in simplicity.

Further, the Bench ruled that Section 3(d) was wrongly invoked since the claimed process yielded distinct products such as magnesium sulphate and high molecular weight organic compounds which were not disclosed in the cited prior art documents. The Bench observed that the Controller had held that the process of the impugned application did not involve any new reactant nor results in a new product, and therefore, the claims were not allowable under Section 3(d). The Single Judge, in turn, had agreed that since all individual steps of the claimed process were already known and produced the same end-product (potassium sulphate) the invention was hit by Section 3(d). The Bench held that the Controller and the Single Judge had both invoked Section 3(d) without demonstrating that the claimed process was a “known process” or a “mere use” of one.

The Bench explained that the essential precondition for the applicability of Section 3(d) is identity or at least substantial similarity between the earlier and later processes. Further, the Court clarified that the PSITA plays no role in assessing Section 3(d): “The person skilled in the art, we may note, is a stranger to Section 3(d). He is persona non grata, so far as applicability of Section 3(d) is concerned”.

It was noted that the mere fact that potassium sulphate may be one of the products that finally emerges from the process that the Appellant seeks to patent, as well as from D1, cannot result in the two processes being the same.

In view of the foregoing, the Bench quashed both the impugned decisions and remanded the matter to the Patent Office for fresh consideration. The Bench directed that the Roche test be properly applied and stressed adherence to Steps 1, 2, and 5: identifying the PSITA, defining the inventive concept, and assessing obviousness without hindsight.