- INTRODUCTION
In the landmark decision of Gayatri Balasamy v. ISG Novasoft Technologies Ltd1., the Supreme Court by a majority (4:1), led by then Chief Justice Sanjeev Khanna, held that the statutory power to "set aside" an award inherently includes a limited power to "modify" it. The reference became necessary in light of the Supreme Court's decision in NHAI v. M. Hakeem2, which had barred courts from modifying arbitral awards under Section 34 of the Arbitration and Conciliation Act, 1996 ("Act"). Dissenting with the majority, Justice K.V. Viswanathan, has cautioned against such a power based on the text and legislative intent of the Act.
This article critically examines the Supreme Court's ruling, contending that the majority's reasoning, particularly its conclusion that the greater power to annul an award includes the lesser power to modify it, is fraught with uncertainties for India's pro-arbitration regime. Drawing on the dissenting view and comparative perspectives from the United Kingdom, Singapore, and UNCITRAL Model Law jurisdictions, the article explores how this expansive interpretation of Section 34 of the Act, could undermine the finality of arbitral awards, blur the line between limited judicial review and appellate functions, and create practical difficulties and delays in enforcement proceedings. It also discusses why existing mechanisms in the Act (such as Section 33 for corrections and Section 34(4) for remittal) are better suited for remedying manifest errors, without compromising the integrity of the arbitral process.
- THE MAJORITY'S REASONING: FROM SETTING ASIDE TO LIMITED MODIFICATION
Both the majority and minority judgments accept the possibility of partially setting aside an award. Applying the principles of severability and omne majus continet in se minus ("the greater power contains the lesser"), the court recognizes severing the offending portion of the award as a sub-species of the power to set aside the award in its entirety.
The court sought to confine modification powers to three types of cases: (a) severable illegality (by partial setting aside), (b) typographical and clerical errors, and (c) post-award interest. However, it committed two significant errors. First, it conflated the powers of partial setting aside and modification by implicitly recognizing partial setting aside as a subset of modification. Second, it failed to explicitly confine the types of cases to these three scenarios, leaving the question open-ended for subsequent judgments to potentially expand the scope. At one point (Para 44), the majority discusses the scope of modification as including anything that falls from the tribunal's decision on a question of law.
Although the court was careful to disallow modifications of the substance of an award on the ground that it would result in significant hardship, its open-ended permission to tinker with peripheral aspects of an award creates significant uncertainty. If the court circumscribes the power to such minor aspects, one may ask whether remission would not be more appropriate – a power that the Act specifically provides. The court observed that if precise modification is possible, the court will undertake it; if reconsideration is required, it will remit the matter to the tribunal under Section 34(4); and if the illegality is such that the tribunal cannot fix it on remission, the award or part thereof will be set aside.
- UNDERMINING FINALITY AND PARTY AUTONOMY
While the majority's intent to spare parties the ordeal of re-arbitration is understandable, its solution - judicial tinkering with the award, comes at the cost of finality, a hallmark of arbitration. Parties choose arbitration with the expectation that the arbitrators' decision will be the final word on their dispute, without prolonged appeals, as opposed to a decree granted by a civil court which can be appealed on merits. Even a limited opening for court-driven changes undermines this core expectation of finality.
The dissent emphasized that the Act nowhere authorizes courts to modify an award, unlike the old Arbitration Act, 1940, which explicitly did, indicating a deliberate legislative choice to restrict judicial intervention. Notably, the terms 'modification' or 'modify' do not appear even once in the IACA. The draft Arbitration and Conciliation (Amendment) Bill, 2024, consciously uses the word 'modify' (by inserting section 17(da)(a)) to empower the arbitral tribunal, not the court, to confirm, modify, or vacate ad-interim measures. This demonstrates that the power to modify must emanate from legislative mandate. Moreover, the 2024 Bill refrains from granting modification power to the court under Section 34.
The language in Section 34 of the Act, permits only setting aside (and, at most, remanding to the tribunal) and nothing more. Reading in a power to alter awards, contradicts the Act's plain text and the Model Law's ethos of minimal interference. Even the 2015, 2019, and 2021 amendments to the Act did not introduce any court power of modification, confirming the Parliament's intent and choice not to grant it.
Allowing a court to rewrite parts of an award, even to fix apparent errors, is a clear departure from the contractual understanding and legislative intent. It also risks encouraging losing parties to treat Section 34 of the Act as an opportunity for a de facto appeal, hoping for a court-engineered revision rather than the all-or-nothing outcome of a set-aside. This, in turn, means more post-award litigation and diminished finality.
- BLURRING JUDICIAL REVIEW AND APPELLATE OVERREACH
Section 34 of the Act was never intended to operate as an appeal on merits. The majority insists that its limited modification remedy will not entail reconsidering evidence or findings. Yet, in practice, any modification of an award risks sliding into a merits-based review. To "modify" an award, a judge must inevitably engage with the substance of the decision, which would mean going into the merits of the case, a power which the Court does not have under Section 34.
As the dissent observed, the power to modify "involves the court entering into an adjudication on the merits, qualitative of an appellate function," which lies outside the scope of Section 34 of the Act. Unlike an appellate court under the Civil Procedure Code, 1908 (which can vary a trial court's decree), a court under Section 34 of the Act was never given such broad authority. This provision of the Act is increasingly being used as an appellate provision, which must be heavily discouraged by the Courts.
The majority's invocation of the maxim "omne majus continet in se minus" is, with respect, misplaced. Annulment and modification are fundamentally different. One nullifies the award, the other changes its content. Justice Viswanathan draws a clear line between severance (removing an offending part of an award, which the Act allows in certain cases) and modification (changing the award's substance). Severance is a passive remedy to uphold what can be saved, whereas modification actively inserts the court into the merits. By authoring a portion of the award, even if limited, the court begins to perform the arbitrator's role, which is not envisaged under the Act.
While the majority likened some modifications to mere corrections of clerical errors, even determining what counts as a "manifest error" can draw judges into re-examining the arbitrator's reasoning. Reducing a high interest rate or excising a disputed cost component entails a judgment call on the arbitrator's discretion, fairness or correctness - matters that the parties entrusted to the arbitral tribunal. The thin line between a permissible fix and an impermissible revaluation of the case facts can easily blur. This is precisely the scenario that the narrow scope of Section 34 of the Act was designed to avoid.
- LIMITED JUDICIAL INTERVENTION UNDER SECTION 5 OF THE ACT
Section 5 of the Act contains a special provision barring judicial intervention except as provided for under the Act itself.
The objective of this section is to ensure that the Act aims at speedy and efficient dispute resolution. As recently as in the seven-judge bench judgment in Re: Interplay3 between arbitration agreements under the Act and the Indian Stamp Act 1889, the Supreme Court observed that the primary objective of the Act was to minimize judicial intervention in arbitral proceedings. Judicial authorities are generally not allowed to interfere in arbitration unless the Act specifically allows it. There are two main issues with Section 5 of the Act and the judgment in question:
- The judgment allows courts to modify arbitral awards, even though the Act does not expressly provide this power;
- The Court by interpreting Section 34 of the Act to include a power to modify awards, tries to justify judicial intervention as being inherent to Section 34 of the Act, suggesting that this power to modify, does not violate the principle of limited court involvement.
This approach, however, runs counter to the legislative intent of the Act and opens the door to circumventing the Section 5 bar by holding that such powers are inherent in other sections by reading them in.
- EXISTING MECHANISMS FOR CORRECTING ERRORS
Critics of the ruling point out that the IACA already provides avenues to address manifest errors in awards without compromising finality. Section 33 of the Act empowers the arbitral tribunal itself to correct clerical or computational mistakes in the award (or give an interpretation of it) upon a party's request, typically within 30 days of the award. Likewise, Section 34(4) allows a court, if asked by a party, to suspend the proceedings and remit the matter back to the tribunal so that the arbitrators can cure any defect that has prompted the challenge. These mechanisms enable errors to be fixed by the original decision-maker—the arbitrator, rather than by the court.
The dissent highlights that these existing provisions are the proper route for handling obvious errors. If an award contains a calculative mistake or technical flaw, the tribunal can be reconvened under Section 34(4) to set it right. This preserves the arbitral process: the arbitrators who heard the case correct their own award, maintaining consistency with party autonomy and the one-stop adjudication intent. This also ensures that the arbitral tribunal takes its job seriously, without becoming complacent in the assumption that any errors it commits can be rectified by the court under the guise of modification.
In the dissent's view, therefore, a judicial modification power was unnecessary. Parties are not left without recourse: if an award truly cannot stand due to an incurable defect, it can be set aside and, as Section 43(4) assures, the parties may commence fresh arbitration to finally resolve their dispute. That outcome, albeit inefficient, respects the limited role of courts. By contrast, the majority's approach of court-led award revision bypasses the Act's built-in error-correction framework and blurs the line between review and re-adjudication.
- ENFORCEMENT RISKS UNDER THE NEW YORK CONVENTION
A further concern is how a modified award will be viewed when enforcement is sought abroad under the New York Convention. The majority dismissed this as a non-issue, reasoning that the Convention defers to the law of the seat in determining an award's validity. Since Indian law (per the majority) permits limited modifications, a court's modification simply becomes part of the arbitral award, which is enforceable under the Convention.
While disagreeing with the majority opinion on this aspect, Justice Viswanathan warns that without explicit statutory guidance, a foreign court might well conclude that an Indian award altered by a judge is no longer the original arbitral award and thus falls outside the Convention's coverage. Notably, other jurisdictions that allow courts to modify awards have statutory provisions to deem the court's changes as part of the award itself. For example, the UK Arbitration Act's Section 71 ensures that if a court varies an award, the variation is treated as part of the award for all enforcement purposes. Singapore and New Zealand have similar statutory clarity.
The dissent warns that, in the absence of such a provision, a foreign court may well take the view that an award which has been substantively altered by a Court is no longer the "original" arbitral award contemplated by the Convention. This could lead to the refusal of enforcement on the ground that the document presented is not an "award" within the meaning of the Convention. Different foreign courts may take divergent approaches to the enforcement of Indian awards that have been modified by courts. Some may accept the modified awards as enforceable, while others may reject it, leading to unpredictability and increased litigation risk for parties seeking to enforce Indian awards internationally. These practicalities are yet to be tested.
This ambiguity could make India an unviable seat for international arbitration. Parties may worry that a court-adjusted award introduces uncertainty and potential resistance in foreign courts. The majority's ruling, unless buttressed by legislative fixes, inadvertently puts the enforceability of Indian awards at risk on the global stage - precisely the opposite of the pro-enforcement bias that the New York Convention regime is meant to foster.
- PRACTICAL CHALLENGES
The majority's doctrinal leap also translates into a host of day-to-day complexities for practitioners. Case assessment becomes a moving target. A continuum of possible judicial tweaks must be factored in, ranging from a nominal interest adjustment to a drastic recalibration of damages - making risk matrices and settlement discussions far less predictable for businesses. Moreover, the process is likely to become more time-consuming and difficult for the parties to navigate.
Allowing such modifications under Section 34 of the Act would pose practical challenges For instance, a limited challenge to part of the award, seeking modification on any of the three specified grounds, may risk stalling execution and enforcement proceedings. This process could take several years, including parties filing several rounds of appeal challenging the modification order, thereby resulting in delay in enforcement or event execution of arbitral awards.
The lack of statutory clarity creates unpredictability for parties seeking to enforce Indian awards abroad. Parties may face protracted litigation in foreign courts as respondents challenge the enforceability of modified awards, arguing that such awards do not comply with the New York Convention. This unpredictability increases costs and delays, undermining the efficiency and finality that arbitration is intended to provide.
By introducing the power to modify under Section 34 of the Act, the Supreme Court has effectively opened up for debate what was otherwise a largely settled jurisprudential question. Given the malleable nature of the power to now modify an award, it is likely that parties would in addition to challenging an award, also in alternative seek its modification. Such an approach poses a serious risk that a Section 34 court could turn into an appellate court, something that was never intended by the legislature.
- CONCLUSION
The Supreme Court's decision in Gyatri Balasamy marks a significant shift in Indian arbitration law by allowing courts a limited power to modify arbitral awards under Section 34 of the Act. While the majority aims to prevent manifest injustice and avoid unnecessary re-arbitration, this approach risks becoming a double-edged sword, undermining the finality, part autonomy and efficiency that are the hallmarks of arbitration. The dissenting opinion warns that such judicial intervention could blur the line between review and appeal, create enforcement risks under the New York Convention, and introduce practical uncertainties for parties. Existing mechanisms within the Act already provide for error correction without compromising the arbitral process. The key takeaway is that, unless exercised with utmost restraint and supported by legislative clarity, the power to modify arbitral awards could erode the very advantages that make arbitration a preferred mode of dispute resolution. The future of Indian arbitration will hinge on how courts judiciously wield this new authority.
The authors thank Kedar Manoj Ammanji and Aditya Bhargava students from NLSIU, Bangalore for their valuable research assistance and insightful observations, which contributed to the preparation of this article.
Footnotes
1 2025 INSC 605
2 (2021) 9 SCC 1
3 2023 INSC 1066
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
