Empirical insights into the legal quicksand: the case of patent illegality in Indian arbitration

Abstract

This paper delves into the intriguing evolution of patent illegality in Indian arbitration law, a concept formalized by the Supreme Court in the Saw Pipes case and later refined by the 2015 Amendment to the Arbitration and Conciliation Act, 1996. Analysing 699 cases from 2016 to 2023 reveals a striking 29.38 per cent success rate for arguments based on patent illegality. Despite the Supreme Court’s intent to restrict its scope, lower courts often delve into case merits under this guise. Key rulings like Ssangyong Engineering and Delhi  Airport  Metro  Express highlight the judiciary’s struggle to maintain this balance. The study critiques the inconsistent application of standards under patent illegality, leading to contradictory verdicts and threatening arbitration finality. Overall, this analysis underscores the critical need for clearer judicial guidelines to ensure a balanced approach that respects both the autonomy of arbitral awards and the supervisory role of the judiciary. The findings suggest that current judicial practices may inadvertently expand the scope of patent illegality, thus necessitating reforms to align with the intended legislative framework.

Introduction

Arbitration is, in principle and practice, a mode of alternate dispute resolution, where parties agree to resolve their disputes outside the court. However, it is well acknowledged that courts must have a certain supervisory jurisdiction over the arbitral awards rendered.¹ Section 34 of the Arbitration and Conciliation Act 1996 (‘the Act’) encapsulates the grounds under which a domestic award can be set aside. One such ground under Section 34 is stipulated in subsection (2A) which gives judges the discretion to set aside an arbitral award if it is vitiated by ‘patent illegality’ appearing on the face of it.² Before it was formalized under Section 34(2A), patent illegality was used as a subground under public policy.³ Much like the ground of public policy, patent illegality has been criticized for being vague and allowing parties to unnecessarily challenge and get awards set aside.⁴ Despite being hailed as a narrow ground, the disturbing tendency amongst courts to use patent illegality to get into the substantive merits of the case has been observed by the Supreme Court.⁵

In the empirical study this paper undertakes, we see that in the 699 cases analysed between 2016 and 2023, there is a 29.38 per cent success rate while arguing patent illegality. Despite the staggering number, there has been very little literature surrounding this subject matter. Most of the scholarship dealing with patent illegality can be divided under two heads, individual analysis of case law⁶ and doctrinal analysis of patent illegality before the 2015 Amendment to the Act.⁷ Thus, there has been little engagement with the various subgrounds under patent illegality or a comprehensive assessment of high court decisions to observe how the ground has been interpreted. It is in this broad context that this paper argues that the evolving jurisprudence of patent illegality in Indian arbitration law, marked by the Supreme Court’s initial articulation in the Saw Pipes case and subsequent refinements, has led to inconsistent judicial interpretations and threatens the finality of arbitral awards, necessitating clearer guidelines to ensure balanced legal oversight.

The paper first traces the historical development of patent illegality, focussing on the Supreme Court’s landmark Saw Pipes case and the impact of the 2015 Amendment. Second, it presents an empirical analysis of 699 cases from 2016 to 2023, revealing trends and other observations. Third, individual subgrounds under patent illegality are analysed to understand the inconsistent application of standards that invariably lead to a high success rate as a defence. Finally, the conclusion emphasizes the need for clearer judicial guidelines to ensure the finality of arbitral awards while providing necessary legal scrutiny.

History of patent illegality: Saw Pipes, Ssangyong Engg., and Beyond

The Supreme Court in Oil & Natural Gas Corporation Ltd v. Saw Pipes introduced the ground of patent illegality.⁸ It was observed by the Court that the jurisdiction of the arbitrator flows from the contract between the parties and the Act itself. Thus, the tribunal cannot undertake any action that violates the procedures prescribed in the Act or the contract.⁹ However, the Supreme Court went one step further to state that while traversing through its jurisdiction, the arbitrator cannot also go beyond other substantive laws.¹⁰ The rationale put forth was that it cannot be the legislature’s intention to allow an arbitral tribunal to contravene provisions of other statutes as that would be contrary to the basic concept of justice.¹¹ The test thus put forth was that the error of law must be apparent on the face of the award.¹² The Court concluded that any finding in violation of the contract, the Arbitration Act and substantive provisions of law would be patently illegal. It must be noted that patent illegality was introduced as one of the four categories of violations under public policy and not as a separate ground.¹³

While Saw Pipes may seem like a novel creation, the Supreme Court claimed it to be a mere culmination of years of jurisprudence. In Champsey Bhara v The Jivraj Balloo Spinning, the Bombay High Court in 1923 held that an award based on an erroneous legal proposition could be set aside.¹⁴ Similarly, the Supreme Court in M/S. Alopi Parshad & Sons v Union Of India observed that if upon scrutiny it was evident that the award is based on an erroneous legal application, then the same can be set aside.¹⁵ The Supreme Court in Saw Pipes relied upon multiple precedents such as Alopi Parshad and Champsey Bhara to justify their conclusion.¹⁶ The issue that arises here is that all the cases cited by the Supreme Court relied on the Arbitration Act, 1940. Section 16(1)(c) of the erstwhile Act clearly allowed the courts to reject enforcement of awards wherein ‘an objection to the legality of the award is apparent upon the face of it’.¹⁷ However, no such provision existed in the 1996 Act for the Court to put reliance upon such cases and reach a similar conclusion. Thus, while Saw Pipes gave the illusion of sticking to the legislative mandate and relevant precedents, the creation of ‘patent illegality’ as a ground was clearly by judicial fiat. FS Nariman, in his speech at the inaugural session of Legal Reforms in Infrastructure, made a similar criticism of the case. He remarked that Saw Pipes failed to incorporate the intent of legislature which was to shift away from the 1940 Act and minimize judicial interference.¹⁸

The point is further buttressed by the fact that the 167th Law Commission Report in 2002 specifically proposed the insertion of a new provision (Section 34A), which would have included the ground of ‘substantial error of law apparent on the face of the award’.¹⁹ This recommendation was made after recognition of the fact that ‘no provision is made in section 34 if there is an error of law apparent on the face of the award’.²⁰ However, it is pertinent to note that this recommendation was never adopted into law, leaving a legislative vacuum on this issue. Consequently, Saw Pipes ended up creating the ground of patent illegality by judicial fiat, filling the gap through judicial interpretation rather than legislative action.

Subsequent developments—Associated Builders and Ssangyong

The judgment in Saw Pipes stayed largely consistent for a decade with subsequent benches reiterating its conclusions.²¹ It was finally in Associate Builders v Delhi Development Authority that the Supreme Court provided further clarity over the scope of patent illegality.²² The Court explained that the three kinds of violations mentioned in Saw Pipes were merely violations of the 1996 Act.²³  First, a contravention of the substantive law of India is merely a manifestation of Section 28(1)(a), which states that if the seat of arbitration is in India, then the dispute should be decided in ‘accordance with the substantive law for the time being in force in India’.²⁴  Second, adherence to the contract is a requirement stipulated by Section 28(3) of the 1996 Act.²⁵ Section 28(3) states that the ‘arbitral tribunal shall decide in accordance with the terms of the contract’.²⁶ It was clarified by the Court that as long as the arbitral tribunal construes the contract in a reasonable manner, no violation can be made out. It is only when the construction is of such nature that ‘no fair-minded or reasonable person’ could reach would the award be set aside.²⁷ And third, obviously, a violation of the 1996 Act itself would amount to patent illegality.²⁸ Thus, an attempt was made to root all the requirements of patent illegality within the 1996 Act and not create an independent basis.

At about the same time another controversy had emerged. In Phulchand Exports v OOO Patriots, the Supreme Court extended the scope of public policy as understood in Saw Pipes to international arbitrations.²⁹ Although this decision was overruled in Shri Lal Mahal,³⁰ the 246th Law Commission Report recommended amendments to remove any ambiguities. It was suggested that Section 34(2A) be inserted in the Act to clarify that the application of patent illegality was limited to domestic awards.³¹ Keeping this recommendation in mind, the 2015 Amendment to the 1996 Act inserted Section 34(2A).³² It contained a proviso that stated that ‘an award shall not be set aside merely on the ground of an erroneous application of the law or by re-appreciating evidence’.³³ The Report specifically mentioned that the extent of review under patent illegality is determined by the standard established by the Supreme Court in Saw Pipes.³⁴ Hence, we witnessed a full circle where Saw Pipes invoked the 1940 Act to establish the concept of patent illegality and later, the legislature drew from Saw Pipes to introduce it into the 1996 Act.

The stage was set for the Supreme Court in Ssangyong Engineering v National Highways Authority of India to clarify the scope of the 2015 Amendment. The Supreme Court made three important observations to that effect. First, a mere violation of substantive law that does not form a part of public policy would not constitute a ground to set aside an award.³⁵ However, failure to give reasons for the award as stipulated by Section 31(3) of the 1996 Act would still amount to patent illegality.³⁶  Second, under the ground of patent illegality, the courts cannot step into the role of an appellate forum and reappreciate evidence.³⁷ And third, the scope of patent illegality would include (i) when the arbitrator construes the contract in a manner that no fair-minded or reasonable person would; (ii) when the arbitrator goes beyond their jurisdiction delineated by the contract; and (iii) when the finding suffers from ‘perversity’.³⁸ The test of perversity was borrowed from Associate Builders, wherein an award was said to be perverse when it either considered no evidence for its findings, considered irrelevant evidence for its findings, or reached a conclusion by ignoring vital evidence.³⁹ Therefore, the test underwent a paradigm shift after the Saw Pipes case.

Post Ssangyong

The Supreme Court in Delhi Airport Metro Express v Delhi Metro Rail Corporation noted a disturbing tendency among courts. It was observed that under the garb of assessing patent illegality, lower courts were appreciating factual aspects of a case.⁴⁰ The Court reiterated the scope clarified in Ssangyong and warned courts from wandering away from the contours of the demarcated standard.⁴¹ The Supreme Court, in its most recent pronouncement on the matter in the year 2023, affirmed that the legal principles enunciated in the seminal cases of Associate Builders (to the extent not overruled by the subsequent decision in Ssangyong) and Ssangyong remain authoritative and constitute binding precedent for the jurisprudence concerning the doctrine of patent illegality.⁴² However, as observed in Batliboi Environmental Engineers, awards rendered prior to the effective date of the 2015 Amendment will only be tested on the standard elucidated in Associate Builders.⁴³

Empirical Analysis and Observations

With this history in mind, this paper undertakes an empirical study of patent illegality cases in India to assess whether the High Courts are adhering to the restrictive interpretation endorsed by the Supreme Court.

Data and methodology

This paper is based on an observational study of patent illegality cases from the period of January 2016 to December 2023. This period is specifically to account for the 2015? Amendment of the Arbitration and Conciliation Act 1996, which introduced Section 34(2A). The cases have been selected from the case reporter, Supreme Court Cases (SCC),⁴⁴ and filters have been applied to restrict the analysis only to High Court and Supreme Court cases. The keywords used in the search are ‘arbitration’ and ‘patent illegality’. This keyword search produced 1,277 cases, which were then manually read through by the authors and discerned for relevancy. The data collected from the relevant 699 cases was then hand-coded for seven variables (although not all of them are relied upon for this paper). All the seven variables for which the data is hand-coded are attached in Annexure A.

Observational analysis

The first obvious observation is that out of 699 cases analysed, patent illegality was successfully argued in 205 cases, indicating a 29.32 per cent success rate. While no similar study has been conducted in India for us to have an effective benchmark for comparison, Dr Monique Sasson has conducted an empirical study of national courts and their enforcement of international commercial arbitration awards.⁴⁵ Since most countries do not have a patent illegality ground for set-aside, we shall look at the defence of public policy for comparison. This study of 1,093 cases revealed that public policy as a defence succeeded 19 per cent of the time.⁴⁶ Thus, 29.32 per cent is a significantly higher rate of success and merits further exploration.

Figure 1 shows that the success rate has been largely consistent in the last four years, hovering around the 30 per cent mark. Except for 2019, there is no big outlier that is skewing the overall data.

The second observation as evident through Fig. 2 is that the High Courts of Delhi, Bombay and Madras dealt with 76.39 per cent of patent illegality cases. This is not surprising considering these cities are the commercial hubs of India and account for a significant portion of the GDP of the country.⁴⁷ Even the Khaitan Domestic Arbitration Report 2024 concluded that Mumbai and Delhi were the preferred seats for domestic arbitration.⁴⁸

However, the surprising result from this data is the absurd variance in success rates across the forums. For example, while the Delhi High Court has only an 18.34 per cent success rate while arguing patent illegality, the High Courts of Bombay and Madras have a success rate of 46.66 and 37.73 per cent, respectively! While courts usually rely on different philosophies and precedents while adjudicating matters, this variance cannot merely be attributed to those factors. Until the scholarship can conclusively explain this phenomenon, it appears that, statistically, litigants have a higher likelihood of success with patent illegality before the Bombay or Madras High Court as opposed to the Delhi High Court.

The third and final observation is that litigants often argue multiple grounds under patent illegality, as shown in Fig. 3. One can see that the most commonly used argument is the arbitrator incorrectly interpreting the contract, with the arbitrator exceeding their jurisdiction being the least commonly used.

Understandably, perversity and contractual interpretation are the two most commonly used grounds as it is easier to allege that there existed another meaning of the terms of the contract or that the arbitrator has failed to appreciate some evidence. These grounds will be analysed individually in the subsequent section.

Analysing the subgrounds under patent illegality

This section will go over the three most used subgrounds of patent illegality to see why it has such a high success rate in domestic arbitrations. A caveat may be drawn here that there might be other reasons why such a success rate is being achieved. However, the purpose of this paper is not to get into the institutional deficiencies of Indian arbitration⁴⁹ but rather to analyse the shortcomings in the jurisprudence around patent illegality.

Contractual interpretation

In Ssangyong Engineering, the Supreme Court laid out that an arbitrator construing the contract in a manner that no fair-minded or reasonable person would do would amount to patent illegality.⁵⁰ Since this case in 2019, the usage of this ground has increased every year, possibly owing to its vagueness and ambiguity (see Fig. 4). To that end, this section will first demonstrate how the standard to assess this prong of patent illegality has oscillated between ‘possible’ and ‘plausible’ interpretation; second, showcase how such varying standards have led to contradictory verdicts; And third, argue how such an approach has shifted the court’s role from a supervisory to an appellate forum.

Possible or plausible: the legal tug of war

‘Plausible’ and ‘possible’, both revolve around the idea of something being within the realm of potentiality, but they carry different nuances. ‘Plausible’ suggests that something appears believable or convincing based on the available evidence or circumstances, whereas ‘possible’ simply indicates that something could happen or be true without assessing its credibility or likelihood.⁵¹ Thus a pictorial representation of their position on a spectrum is as follows:

In Ssangyong, the Supreme Court observed that the interpretation of contractual terms is entirely the arbitrator’s domain unless ‘the arbitrator construes the contract in a manner that no fair-minded or reasonable person would; in short, that the arbitrator’s view is not even a possible view to take’. This view was reiterated by the three-judge bench in Patel  Engg. Ltd. v  North  Eastern  Electric  Power  Corp. Ltd.⁵² This approach can be traced back to the Supreme Court in Sudarsan  Trading  v  Govt. of  Kerala back in 1989. In the case of Sudarsan, the court emphasized that judicial intervention in arbitral awards was unwarranted as long as the arbitrator’s interpretation was a possible one.⁵³ This principle is rooted in the recognition that the arbitrator holds the final authority in resolving disputes between parties. Consequently, a well-reasoned award cannot be set aside merely because the court might have arrived at a different conclusion based on the same facts.⁵⁴ Thus, the test appeared to be of ‘possible’ interpretation. However, around 2007, the Supreme Court in Boc  India  Ltd  v  Bhagwati  Oxygen concluded that ‘when the Arbitrator had taken a plausible view on interpretation of contract, it is not open to the court to set aside the award’.⁵⁵ This led to a trend wherein High Courts began scrutinizing whether the arbitrators’ views were ‘plausible’.⁵⁶ Some courts even went further to combine both standards as a single test. For example, in Delhi  Jal  Board  v  M/S  Dev  Raj Katari, the Delhi High Court described the arbitrators’ view as the ‘possible and plausible view’.⁵⁷

We know from the earlier portion of the section that ‘possible’ and ‘plausible’ mean two different things. While no court has undertaken the exercise of demarcating the said difference, several decisions have hinted at it. In Rashtriya Ispat Nigam Ltd v M/S Dewan Chand Ram Saran, the Supreme Court, while looking at the arbitrator’s interpretation, remarked that ‘the view taken by the arbitrator was clearly a possible if not a plausible one’. Hence, there was an understanding regarding the distinction between ‘possible’ and ‘plausible’. Even more recently, a three-judge Supreme Court bench in Atlanta Limited v Union of India observed that:

As long as the Arbitrator has taken a possible view, which may be a plausible view, simply because a different view from that taken in the Award, is possible based on the same evidence, would also not be a ground to interfere in the Award.

Thus, a view can be both possible and plausible. However, the standard is only for the former and not the latter. This oscillation between the two standards has allowed courts to justify a higher or lower level of scrutiny over the awards. The next subsection will demonstrate how the level of interference by the courts has been inconsistent.

Judicial inconsistencies

The ambiguity in understanding patent illegality has given rise to a number of contradictory judgments, making the debate murkier. In Fiberfill Engineers v Indian Oil Corp. Ltd., the Delhi High Court set aside an award on patent illegality because the arbitrator had not rendered a finding on whether a delay caused by a party had led to damage or loss.⁵⁸ On the contrary, in Delhi Paper Products Co. (P) Ltd. v Union of India, the very same court reprimanded the arbitrator for getting into an independent enquiry of loss and damages and held that this enquiry made the award patently illegal.⁵⁹ The Court in this case described the enquiry as ‘remote and not a direct loss’.⁶⁰

Another situation, where the inconsistency becomes more prominent, is for issues of interpretation of specific clauses. In National Highway Authority of India v Suresh Chandra, the Delhi High Court was dealing with the challenge to the arbitrator’s interpretation of the ‘force majeure’ clause.⁶¹ The arbitrator in this case interpreted ‘complete blockade of the road’ to mean that it covers the situation in the instant case wherein owing to damage caused to the river bridge, there was a considerable reduction in traffic volume.⁶² The Court held that such an interpretation is patently illegal since only a literal interpretation of ‘complete blockade’ could be taken by the arbitrator.⁶³ The same court, four months later, in a judgment delivered in Union of India v Tayal & Co., took an entirely different approach. This case also involved the interpretation of a force majeure clause.⁶⁴ The arbitrator had concluded that power cuts in Haryana and Himachal Pradesh could fall within the scope of the clause.⁶⁵ While the Court held that the clause was wrongly interpreted, it went on to hold that a mere wrongful interpretation of a contractual clause would not amount to patent illegality.⁶⁶ Thus, while one case held that a wrongful interpretation would render the award liable to be set aside, the same court, in a different case, adjudicated that erroneous interpretation would fall outside the scope of patent illegality. This paper argues that such outcomes are possible because of the inability of the courts to settle on the standard of scrutiny (possible v plausible).

Supervisory or appellate courts: from oversight to overrule

The role of courts under Section 34 of the Act has always been understood as supervisory in nature. The Supreme Court, in McDermott International Inc. v Burn Standard Co. Ltd., observed that any review of an award under Section 34 would be supervisory, ie, pertaining to ensuring fairness and following principles of natural justice.⁶⁷ The reason why the role of the courts was kept to a minimum was to ensure party autonomy and independence.⁶⁸ This view was further reiterated by the Delhi High Court in IRCON International Ltd. v Patil Rail Infrastructure Pvt. Ltd., which held that when deciding cases under Section 34, the court does not sit in appeal over the award and decide on matters of interpretation as long as such interpretation is possible.⁶⁹

What does it mean to say that the role of the court is supervisory and not appellate? In Puri Investments v Young Friends & Co., when enunciating what a supervisory jurisdiction entails, the Court held that there can be no reappreciation of evidence or revisiting of the facts of the case.⁷⁰ In Central Council for Research in Ayurvedic Sciences v Bikartan Das, the Supreme Court held that while exercising supervisory jurisdiction, as opposed to an appellate one, the court could not review findings of fact by the lower court or tribunal, even if they were erroneous.⁷¹ This standard, however, has not been adhered to by courts, with them constantly assuming appellate jurisdiction and going into an analysis of facts. In Baba Builders v Ircon International Ltd., the Court undertook a factual enquiry, looking at past letters, notices, and contractual clauses between the parties to hold that the action of the parties was not under a clause for modification (as held by the arbitrator) but under a clause for termination.⁷² Thus, even though the conclusion reached by the arbitrator was a plausible one, the Court disagreed by actively undertaking a factual analysis. Similarly, in National Highway Authority of India v Suresh Chandra, the Court again delved into an analysis of the intention of the parties and held that the interpretation of a contractual clause was contrary to its intent.⁷³ The construction here was set aside, not because it was contrary to the terms of the contract, but to the intention of the parties to contract. Such precedents have allowed courts more leeway to hold an award to be patently illegal and also seems to depart from the standards of a supervisory role. This enables litigants to repeatedly challenge awards under this ground.

Perversity

The Supreme Court in Ssangyong remarked that

[w]hat is important to note is that a decision which is perverse, as understood in paragraphs 31 and 32 of Associate Builders (supra), while no longer being a ground for challenge under ‘public policy of India’, would certainly amount to a patent illegality appearing on the face of the award.⁷⁴

This is interesting since the paragraphs referred to here did not discuss the issue of patent illegality. In Associate Builders, paragraphs 31 and 32 are discussed under the subheading ‘Fundamental Policy of Indian Law’ rather than in the context of patent illegality. Associate Builders’ point was that an award which does not consider relevant/vital evidence or which is based on no evidence would be against the fundamental policy of Indian law.⁷⁵ There is thus no basis provided by Ssangyong as to why perversity should be brought within the scope of patent illegality. In fact, the cases cited by Associate Builders to substantiate perversity are all administrative law cases.⁷⁶ Another important point to consider is that the 246th Law Commission Report specifically mentioned that the extent of review under patent illegality is determined by the standard established by the Supreme Court in Saw Pipes.⁷⁷ Thus, the 2015 Amendment only intended to give effect to Saw Pipe’s understanding of patent illegality that did not contain perversity. Nonetheless, it is a ground often used by courts and will therefore be analysed in this section.

Perversity, as explained above, is when the arbitrator considers no evidence for their findings, considers irrelevant evidence for their findings, or reaches a conclusion by ignoring vital evidence. As summarized in Shamlal Jotwani v Veena Jotwani, if there is some acceptable evidence on record, no matter how limited, the conclusions would not be deemed perverse and the findings would not be disturbed.⁷⁸ The proviso of Section 34(2A) specifically mentions that the test of patent illegality does not entail a reappreciation of evidence. How do we reconcile the ratio of Ssangyong and the proviso? Trade International v Avon Healthcare provides some answers. The Delhi High Court in this case stated that interference by the courts can only take place when there is an error apparent on the face of the record.⁷⁹ The courts cannot go into the merits and reappreciate facts and evidence.⁸⁰ Unfortunately, an assessment of the cases demonstrates that courts have not adhered to this narrow framework.

Out of the 205 instances wherein an award was set aside under patent illegality, 92 cases were of perversity. What is particularly striking is that perversity has been used to set aside awards in 92 out of the 236 cases where it was raised as an argument. This gives it a high success rate of 38.98 per cent. This section shall attempt to discern this phenomenon. Figure 5 provides a break down of the various sub-grounds argued under perversity. Perversity can be broadly divided into two categories for the purposes of this paper, ie, failing to consider vital evidence or making a decision based on no evidence . While there exist other types of cases, their number is too few to have a tangible impact on the outcome of the success rate.

Ignoring vital evidence

This sub-ground under perversity is closely tied to Section 31(3) of the Act. In Bachawat’s Law of Arbitration and Conciliation, it is said that while courts cannot reappreciate evidence, the arbitrator (owing to Section 31) is still obligated to provide reasons for considering one party’s evidence over another.⁸¹ The arbitrator must ‘give its reasons on consideration of the relevant materials while the irrelevant material may be ignored’.⁸² Thus, there is a realistic expectation from the arbitrators, recognizing the difference between court and arbitral proceedings. In Pt Munshi Ram & Associates (P) Ltd. v DDA, the Delhi High Court opined that an arbitrator is only mandated to deal with ‘substantial objections’ raised by the parties.⁸³ Hence, there is sufficient confusion about the point at which lack of vital evidence becomes an unreasoned award under the Act. As a matter of fact, courts frequently combine these two reasons for setting aside an award. For example, in Damodar Valley Corpn. v Reliance Infrastructure Ltd., the Calcutta High Court concluded that the ‘award granted is patently illegal, perverse, and contrary to Section 31(3) of the Act, and is accordingly set aside’.⁸⁴

The standard of ‘ignoring vital evidence’ currently enables the party that loses in an arbitration proceeding to contest the outcome by alleging that their presented evidence was pivotal and that the arbitrators disregarded it. This trend contributes to an increase in the number of cases brought before the courts seeking a re-evaluation of evidence in blatant disregard of the proviso outlined in Section 34(2A). This trend is evident in the data, as challenges based on perceived perversity rank as the second most commonly argued ground falling under the category of patent illegality. The issue lies not only in the successful challenges to these cases but also in the mere existence of this practice, which needlessly prompts parties to challenge awards, thereby burdening the courts further. The BN Srikrishna Committee highlighted the tendency of Indian Courts to have an interventionist approach, which makes India an ‘arbitration-unfriendly’ jurisdiction.⁸⁵ The interventionist attitude of the courts was also reflected in State of Chhattisgarh v SAL Udyog, wherein the Supreme Court held that patent illegality does not have to be specifically argued but can be invoked by courts on their own.⁸⁶ Recognizing these shortcomings, some US Courts have deliberately removed this as a ground to challenge arbitral awards. In Success Systems v Maddy Petroleum Equipment, the United States District Court ruled that

if manifest disregard of the evidence were an available ground for vacatur, courts would inevitably be asked to second-guess the arbitrators’ factual decisions and, in effect, to retry the case, thereby depriving the parties of the simplicity for which they bargained⁸⁷

The Court further stated that even if manifest disregard of evidence was a valid basis for nullifying an arbitral award, the Court could not apply that criterion.⁸⁸ Such limitations arise because, as is common in arbitration proceedings, the parties opted not to incur the costs associated with transcribing or recording the testimony provided before the arbitrator⁸⁹. Consequently, there exists no documented record of the testimony presented to the arbitrators. Without a comprehensive record of the proceedings, it would be unfeasible for this Court to assess the evidence or the parties’ assertions concerning the proceedings before the arbitrator.

An illustration of the above discussion can be found in Hemant D. Shah v Chittaranjan D. Shah. This case revolved around the issue of whether one of the directors had actually retired.⁹⁰ The respondents contended that the director could not complete the cross examination owing to severe anaemia.⁹¹ Even the arbitrator in the award had acknowledged that both sides were aware of the witness’s medical condition and thus, in his absence, reliance would be placed on other evidence and witnesses.⁹² The counsel for the petitioner also did not contest the respondent’s health condition, as indicated through the minutes of the meeting.⁹³ Despite there being significant jurisprudence regarding accepting a sick person’s statement without cross examination,⁹⁴ the judge ruled that the arbitrator ought to have made an adverse inference.⁹⁵ The inability to do so by the arbitrator was perverse and overlooked vital evidence.⁹⁶ This case sums up the interventionist approach of the courts. Despite there being adequate jurisprudence and evidence supporting the arbitrator’s stance, the Bombay High Court still set the award aside for patent illegality.

Based on no evidence

In Associate Builders, the Supreme Court, commenting on this subground, opined that an arbitral tribunal awarding damages without any justification would be against judicial conscience and hence perverse.⁹⁷ As argued earlier, Associate Builders always referred to perversity under the sub-heading of ‘fundamental policy of Indian law’, which falls under public policy. Yet the confusion continues to persist. In Raipur Alloys and Steel Ltd. v Environmental Engineers, the Bombay High Court held that an award based on no evidence would not only be patently illegal (under perversity) but also against the fundamental policy of Indian law.⁹⁸ The categorization of perversity under patent illegality seems redundant, given the availability of alternative recourses under different subsections of Section 34. Since public policy is a ground for challenge under Section 48, particularly in international commercial arbitrations, it is more appropriate to align perversity within this framework to ensure consistency and uniform application.

Another problem that persists within this subground is that the courts often reappreciate evidence to comment on its adequacy. In Sanman Rice Mills v Punjab State Civil Supplies, the Court concluded that merely relying on the witness statement was not enough, and the arbitrator should have looked at documentary evidence as well.⁹⁹ Thus, an attempt was made to convert the ground of ‘no evidence’ to ‘not enough evidence’. Adequacy of evidence has also been used as a ground for setting aside the award in several cases.¹⁰⁰ However, this was not the intent behind the ground at all, as has been explained above. In John Distilleries v Brihan Maharashtra Sugar Syndicate, the Bombay High Court explicitly observed that:

As long as there is some evidence to sustain a finding, the challenge court under Section 34 of the Act does not interfere with the award.  Sufficiency of evidence is something for the arbitral forum alone to rule on.¹⁰¹

The Orissa High Court in GMR Kamalanga Energy v. SEPCO Electric Power remarked that the sufficiency of evidence is entirely the arbitrator’s domain.¹⁰² The court’s jurisdiction only extends to determining whether the arbitrator’s actions shock the conscience of the court.¹⁰³ However, since these are merely conflicting High Court decisions, parties across the country have another piece of ammunition in their arsenal to challenge arbitral awards in India by arguing that the other side did not ‘sufficiently’ prove their case.

Violation of law

The proviso to Section 34(2A) clearly specifies that an award cannot be set aside on the grounds of patent illegality merely because there has been an erroneous interpretation of law by the arbitral tribunal.¹⁰⁴ Despite this clear expression, High Courts have consistently departed from the proviso to consider violations of substantive law as leading to patent illegality and have set aside the award under section 34(2A). The success rate for arguing this sub-ground is the highest, at 51.5 per cent. While it has only been argued in about one-fifth of the cases, its high success rate surely merits a deeper analysis of the court’s rationale. It is in this context that this section will delve into how courts have tried to circumvent the proviso and have gone on to consider violations of substantive law as a ground for patent illegality.

Use of Section 28(1)(a)

Section 28(1)(a) of the Act states that in cases where the arbitration takes place in India (except for international commercial arbitration), the arbitral tribunal must resolve the dispute based on the current substantive laws of India.¹⁰⁵ The 167th Law Commission Report highlighted the fact that the Act does not provide a remedy for this provision.¹⁰⁶ Moreover, the 246th Law Commission Report while commenting on the amendment to Section 28(3) stated that:

The amendment to section 28(3) has similarly been proposed solely in order to remove the basis for the decision of the Supreme Court in ONGC vs. Saw Pipes Ltd, (2003) 5 SCC 705 – and in order that any contravention of a term of the contract by the tribunal should not ipso jure result in rendering the award becoming capable of being set aside. The Commission believes no similar amendment is necessary to section 28 (1) given the express restriction of the public policy ground.¹⁰⁷

Thus, there was an implicit understanding that mere erroneous interpretation of law would not lead to the award being set aside. The interpretation would have to amount to a violation of public policy. Furthermore, two rules of statutory interpretation would kick in for this conflict within the provisions. First, the proviso to Section 34(2A) being a specific provision will prevail over the general provision of Section 28(1)(a).¹⁰⁸ And second, if two provisions in a statute cannot be harmoniously constructed, the provision enacted later should prevail (in this case, the proviso).¹⁰⁹

However, Courts in India have creatively used this Section to circumvent the proviso to section 34(2A) and hold that violations of substantive law of India by the arbitral tribunal can validly be investigated by courts. In Amazing Research Laboratories v. Krishna Pharma, the Delhi High Court opined that the arbitral award was passed in contravention of the Indian Contract Act.¹¹⁰ Citing Section 28(1)(a), the Court held that since the provision necessitates adherence to the substantive Indian law, this award would be patently illegal.¹¹¹ Interestingly, no reference was made to Section 34(2A) or any other judicial precedents to reach this conclusion.¹¹² Again, in Zakir Hussain v Sunshine Agrisystem, the Court observed that the award had misapplied various sections of bailment under the Indian Contract Act.¹¹³ Relying on Associate Builders, the Court held that a misapplication of substantive laws in an arbitral award would amount to patent illegality.¹¹⁴ This conclusion is extremely misplaced because as observed above, Ssangyong had explicitly stated that post the 2015 Amendment, ‘a mere contravention of the substantive law of India, by itself, is no longer a ground available to set aside an arbitral award.’¹¹⁵

Thus, there has been a clear disregard of the strict statutory wording and Supreme Court precedents. Two years before these High Court decisions, the Supreme Court had already addressed the alleged gap in the law. In Delhi Airport Metro Express v DMRC, the Supreme Court stated that Associate Builder’s interpretation of Section 28(1) to include patent illegality is no longer good law.¹¹⁶ Thus, these decisions undermine the work done at the judicial and legislative level to minimize interference in arbitral awards and gives parties a locus to challenge awards based on the interpretation of substantive laws.

From (incorrect) statutory basis to legal whims

While some courts have resorted to Section 28(1)(a) as a justification to look at contravention of substantive law, others have blatantly held that a violation of law is patent illegality without any basis. For example, in New India Assurance v Khanna Paper Mills, the Delhi High Court set aside an award on the grounds of patent illegality because the arbitrator had erroneously interpreted the IRDA Regulations.¹¹⁷ Similarly, in Jatin Pratap Desai v A.C. Chokshi Share Broker, the Bombay High Court held that the nonadherence to the SEBI Bye-laws and Guidelines amounted to patent illegality.¹¹⁸ Recently, the Madhya Pradesh High Court went even one step further. In Apollo Real Estate LLP v Arun Waghmare, the award was held to be patently illegal because the arbitrator relied on an overturned judgment to interpret Section 32 of the Indian Contract Act.¹¹⁹ This creates further confusion on the scope of Section 34(2A), and makes it more ambiguous.

Violation of The Arbitration Act: is patent illegality the right recourse?

In Associate Builders, it was held that a violation of the Arbitration and Conciliation Act, 1996 would also amount to patent illegality.¹²⁰ However, in Ssangyong, the Supreme Court stated that except for Section 31(3) of the Act, no other provisions of any substantive law in India can be used to set aside an arbitral award under patent illegality.¹²¹ However, as seen in Fig. 6, out of the 61 cases that have been set aside owing to the arbitrator not adhering to the laws of India, 39 cases (almost 64 per cent) are violations of the Act itself. Within the 39 cases, 30 of them are violations of Sections 31(3) and 31(7). Thus, these two provisions will be primarily discussed in this section.

Section 31(7)

Section 31(7)(a) of the Act states that unless otherwise agreed by the parties, the arbitrator may include interest for the sum awarded, at any such rate as he deems reasonable.¹²² Section 31(7)(b), on the other hand, states that a sum awarded by the arbitrator, unless the award otherwise directs, will carry a default interest rate of 2 per cent higher than the prevailing rate on the date of the award.¹²³ Thus, these two provisions serve two very different purposes: while the former gives the arbitrator the discretion to award any rate of interest as he deems reasonable, the latter talks about a default rate of interest that only kicks in where the arbitrator does not award any interest at all. The issue that arises here is that courts have read these subclauses together to mean that the arbitrator themselves cannot award an interest higher than the default rate of interest. In Star Track Fasteners (P) Ltd. v Union of India, the Bombay High Court reprimanded the arbitrator for not awarding interest on the sum awarded.¹²⁴ It held that since the underlying contract did not expressly bar the awarding of interest, the arbitrator was under the obligation to award the same and not doing so amounted to patent illegality.¹²⁵ This is erroneous, since Section 31(7)(a) gives complete discretion to the arbitrator when it comes to awarding of interest. Moreover, the presence of a default rate of interest under clause (b) also indicates that there is a remedy for when the arbitrator does not specify any interest in the award.

In Southern Railway v Santhosh Babu, the Kerala High Court committed the same error by holding that the interest awarded by the arbitrator was not in accordance with clause (b) of Section 31(7) and set aside the award on grounds of patent illegality.¹²⁶ This shows that courts have, time and again, conflated clauses (a) and (b) of Section 31(7). Both the clauses work independently of the other, and the rate of interest awarded by the arbitrator under clause (a) is unrelated to and unlimited by the default rate of interest provided under sclause (b). This distinction was reiterated clearly in the recent case of SAIL v Ores India (P) Ltd., where the Jharkhand High Court held that under Section 31(7)(a), the arbitrator has complete and unlimited discretion to award interest at any rate that he deems reasonable, irrespective of clause (b).¹²⁷

In Southern Railway especially, the Court relied upon the 246th Law Commission Report to state that the arbitrator cannot award interest at the rate of 18 per cent.¹²⁸ However, a quick perusal of the Law Commission Report reveals that they were talking in reference to Section 31(7)(b) while stating that ‘[e]xplanation 1 ensures that the default rate of interest is in line with prevailing commercial realities and not an arbitrary figure of 18%.’¹²⁹ Thus, the rationale of the Law Commission was that in the event an arbitrator does not award interest, the default interest rate should not be an arbitrary number of 18 per cent. This paragraph from the report should not, therefore, be interpreted as a limitation on the arbitrator’s authority to grant an interest rate that they deem appropriate. This is because, under clause (a), the arbitrator retains complete discretion and autonomy to award any rate of interest they deem reasonable. Thus, courts have wrongly read clauses (a) and (b) together and set aside awards on patent illegality where the rate of interest awarded by the arbitrator was not in line with the default rate.

Section 31(3)

Section 31(3) of the Act stipulates that the arbitral award shall provide reasons on which it is based.¹³⁰ Courts have relied on Section 31(3) to set aside awards on the grounds of patent illegality, arguing that the arbitrator either failed to provide reasons or did not provide sufficient reasoning. However, there is no standard for what the scope of reasons looks like or what remedy is available to a party if an award is based on insufficient reasons. Back in 2002, the 176th Law Commission Report of India noticed the lack of remedies for violations of the Act, specifically the nonassignment of reasons. It had thus recommended the insertion of a ‘Section 34A’ under which matters could be remanded back to the tribunal to rectify any errors arising from Section 31(3) of the Act.¹³¹ However, this provision was never enacted in the statute, and thus the implications of an unreasoned award remain unclear.

This was until the Supreme Court in Ssangyong held that if an arbitrator does not provide adequate reasons for an award, it would amount to a contravention of Section 31(3) of the Act and would be patently illegal.¹³² What is interesting is that Ssangyong merely cites Associate Builders for the proposition which was looking at the Section through a pre-amendment lens. The Court does not specify where this ratio is garnered from, and how such a conclusion has come about. A constitutional bench of the Apex Court, in Lalita Kumari v State of U.P., held that the first and foremost rule of construction must be literal construction.¹³³ Only when a provision raises ambiguity can the court depart from literal construction.¹³⁴ This is specifically important with respect to Ssangyong, which was decided post the 2015 Amendment, ie, when the proviso to Section 34(2A) was already in place. Since this proviso explicitly stated that a mere violation of law would not be covered under patent illegality, it is unclear as to how Ssangyong arrived at this conclusion.

Even if one assumes that the standard pronounced in Ssangyong is correct, the problem of vagueness persists. The Court in Ssangyong does not clarify what is meant by ‘adequate’ reasoning. One may reasonably assume that Ssangyong was relying on ONGC Ltd. v Saw Pipes Ltd.’s discussion on Section 31(3) for the proposition since it was ONGC that was expressly given effect in the 2015 amendment.¹³⁵ Interestingly, while ONGC did discuss Section 31(3), it only spoke of a ‘nonspeaking’ award being patently illegal.¹³⁶ There was no discussion regarding the ‘adequacy’ of reasoning. Today, 39.34 per cent of all violations of law cases are just violations of Sections 31(3) wherein one party alleges that the award was not adequately reasoned.

The Supreme Court, in Dyna Technologies v Crompton Greaves, held that Section 31(3)’s mandate is only to have a reasoning that was intelligible and adequate enough for the court to understand the same from a fair reading of the award.¹³⁷ The mandate did not include the requirement for the arbitrator to write an elaborate judgment on all its findings, having regard to the speedy disposal of the proceedings.¹³⁸ It also held that irrespective of everything, inadequacy of awards could not be a ground for setting aside the award.¹³⁹ Recourse had to be taken to Section 34(4), and the award would have to be remanded back to the arbitrator to provide adequate reasons.¹⁴⁰ This, however, has not been followed by High Courts in subsequent cases.

In Ganesh Babu v Adventuries India, the Madras High Court held that the award was patently illegal on grounds of inadequacy since the arbitrator had not given reasons as to why every single exhibit placed before it was accepted or denied.¹⁴¹ If every single exhibit or piece of evidence has to be accepted or denied by the arbitrator with reasons, it not only places an undue burden on the arbitrator but also defeats the purpose of arbitration, ie efficiency and speed. In S.N. Mukherjee v Union of India, the Court had held that an arbitrator is not supposed to provide detailed and elaborate judgment nor are they expected to write a judgment akin to that of a court.¹⁴² This was followed by Vashdev Morumal Sawlani v. Yogesh Mehta, which held that the policy behind Section 31(3) was to enable the court to discern whether or not the tribunal or the arbitrator had either exceeded their jurisdiction or acted without one.¹⁴³ Therefore, asking for a detailed award that discusses every piece of evidence is unreasonable, undue, and against the very purpose of Section 31(3).

Violation of other Provisions of the Arbitration and Conciliation Act

Section 34(2A) is increasingly being interpreted as a residuary provision to include all violations that are not covered by other sub-sections of Section 34. Even when grounds are available under Section 34(2), patent illegality continues to be invoked. For example, the Madras High Court, in Prime Store v. Sugam Vanijya Holdings (P) Ltd., held that the award was patently illegal and liable to be set aside because the arbitrator was invalidly appointed.¹⁴⁴ This is surprising since section 34(2)(a)(v) allows for setting aside the award precisely on this ground.¹⁴⁵ Similarly, in Syed Basheer Ahmed v Cholamandalam Investments and Finance Company Ltd., the award was set aside on patent illegality because no notice was given to the opposite party in accordance with Section 21 of the Act.¹⁴⁶ This was again despite the recourse under Section 34(2)(a)(iii), which allows for the award to be set aside on non-receipt of notice.¹⁴⁷

It can be argued that since the effect of both provisions is the same, it does not matter what route is used to achieve the outcome. However, analysing which section is utilized is salient because of the inherent ambiguity of the scope of Section 34(2A). If the Section is employed as a catch-all provision for all matters (including those that have specific remedies), its already unclear scope becomes further blurry.

Conclusion

The examination of patent illegality in Indian arbitration law reveals significant challenges and inconsistencies in judicial interpretation. Despite the Supreme Court’s attempts to confine the scope of patent illegality, lower courts frequently overreach into the substantive merits of cases, undermining the finality of arbitral awards. This inconsistency is exacerbated by varying standards leading to contradictory verdicts. Key rulings such as Delhi Airport Metro Express highlight the judiciary’s struggle to maintain a balance between legal oversight and arbitration finality. The findings underscore the critical need for clearer judicial guidelines to ensure a more predictable and stable arbitration environment. To address these issues, several steps need to be taken. First, the judiciary needs to establish clearer guidelines that strictly delineate the boundaries of patent illegality, ensuring it is applied only in cases of blatant legal errors and not to reassess the merits of the case. Training programs and workshops for judges could help reinforce the intended narrow scope of this doctrine. Second, the introduction of a specialized arbitration appellate tribunal could provide a more consistent and expert review of arbitral awards, reducing the burden on regular courts and ensuring a uniform application of the law. Third, legislative amendments could clarify the definitions and scope of patent illegality, aligning statutory provisions with judicial interpretations to reduce ambiguity. By implementing these measures, the Indian arbitration framework can better balance the need for legal oversight with the principle of arbitration finality, fostering a more predictable and stable arbitration environment.

Annexure A:

• 1) Year

• 2) Case Name and Citation

• 3) Forum

• 4) Bench Size

• 5) Applicability of Section 34(2A) – Whether the amendment applied to the case

• 6) Whether patent illegality was used to set the award aside

• 7) Reasoning of the court:

  • a) Perversity

  • b) Jurisdiction

  • c) Interpretation of law

  • d) Interpretation of contracts

  • e) Others

Footnotes

Author notes