Modifying Awards without Overreach: Balaswamy’s Overlooked Balance

Prem Vinod Parwani & Pranshu Gupta*

Introduction

A recurring theme in the law of arbitration is the tussle between the principles of fairness and finality of awards. Arbitration, being a creature of contract, is grounded in party autonomy. Consequently, awards are ordinarily final, with limited scope for judicial interference. Nonetheless, arbitration law allows certain carve-outs where the principle of fairness justifies limited departure from this finality. One such carveout is section 34 of the Arbitration Act 1996 (Act’), which provides for the circumstances in which such judicial interference is warranted.

Section 34 contemplates ‘setting aside’ on broadly three grounds: a) where the tribunal lacks jurisdiction, b) where the award conflicts with the ‘public policy of India’, and c) where ‘patent illegality’ is clear on the face of the award.[1] While much literature has been devoted to exploring the precise scope of these grounds, this article focuses on the form of intervention- specifically, the possibility of modification/variance of awards.

The power to modify arbitral awards has been contentious for long now, and especially after the recent Constitution Bench decision in Gayatri Balaswamy, which held that supervisory Courts have a limited power to modify arbitral awards under s.34 of the Act.[2]

Critics have argued that Courts cannot modify awards, since it would amount to excessive judicial interference in the scheme of the Act (here, here and here). Even after the judgement was released, it has been assailed for being arbitration-unfriendly, or contradicting parliamentary intent (here and here).

 This piece will deal with these criticisms in turn – it argues that the majority’s stance in favour of a limited modification power is both normatively desirable and doctrinally consistent with the scheme of the Act.

Firstly, we examine the legal limits of judicial intervention under s.34, analysing textual limitations and broader arbitration principles informing judicial intervention. Secondly, we draw comparative insights from the English Arbitration Act to demonstrate the utility of modification powers. Thirdly, we build upon this doctrinal and comparative foundation to argue that the majority’s limited guardrails for modification ensures doctrinal consistency with the Act while promoting the objects of speedy and efficient arbitration.

The Legal Framework Today: Back and Forth On The Power Of Modification

The earliest reported instance of court modification is in Gautam Constructions (2000), where the Supreme Court modified the interest rate awarded from 18% to 12% since it was “directly opposed to the specific terms in the contract.” This trend continued[3] - with the Supreme Court and High Courts taking it upon themselves to modify awards (particularly on interests and damages) as part of their jurisdiction (here, and here), albeit without addressing whether the scheme of the Act permitted such action.

Some Courts read this power into s.34, while other High Courts rejected modification powers for resembling an appellate court, which is impermissible in the scheme of s.34 (here, here and here).

In 2021, this issue was settled (seemingly) in Project Director v Hakeem. The 2-judge bench of the Supreme Court held:

46. Quite obviously if one were to include the power to modify an award in Section 34, one would be crossing the Lakshman Rekha…..Parliament very clearly intended that no power of modification of an award exists in Section 34 of the Arbitration Act, 1996.

However, the issue emerged again in Gayatri Balaswamy, where a single judge bench of the Madras High Court modified an award in the interests of speedy and efficient adjudication. Its decision was further affirmed by a division bench. This went on appeal before a Constitutional bench, which was set up to reconsider Hakeem’s restriction on modification powers. 

Hakeem’s wholesale rejection of modification was grounded in the text: under s.34(4) the Court can only allow the tribunal to a) resume proceedings, or b) take such action that would eliminate the grounds for ‘setting aside’ the award.[4] The repeated use of ‘setting aside’ in s.34 suggests legislative intent for an all-or-nothing intervention model. Further, this provision is triggered only on a s.34(1) application, which only contemplates such ‘setting aside’. Clearly, there are textual limitations in the Act restricting reading in the power of modification/variance. Justice Viswanathan also firmly roots his dissent in these textual limitations. This is because s.34 is framed as a “protective mechanism”- and judicial modification could defeat arbitration’s purpose of award finality and minimal court interference.

However, there are indubitable benefits for an arbitration framework to incorporate certain forms of modification powers. The next section explores these benefits with reference to English law.

How Modification Can Promote the Objects of Arbitration: A Comparative View

Interestingly, the single judge of the Madras HC in Gayatri Balaswamy, that originally modified the arbitral award, relied on comparable powers in the arbitration statutes of other jurisdictions (paras 41-50). In England, Singapore, Australia and the US, the statutes explicitly provide for modification despite adopting the UNCITRAL Model Law. As such, existing literature (and even the SC in Hakeem)[5] has engaged with this law in two limited ways - first, by referring to it only in passing and second, by referring to it to dismiss the existence of modification powers in s.34, since the Indian legislation lacks such express wording for modification. Here, we will focus on the powers of modification in the English context.[6]

The English Arbitration Act 1996 provides instructive guidance - allowing the court to modify or vary the award on two grounds: i) challenges to the substantive jurisdiction of the tribunal (s.67(3)(b)), and ii) on a question of law arising out of the award (s. 69(7)(b)). Both these provisions are carefully circumscribed by safeguards: s.67(1) mandates exhaustion of arbitral remedies, while s.69 requires that the point of law must substantially affect the rights of parties and be one that has been raised previously before a tribunal. While s.69 has been subjected to criticism for violating principles of arbitration, Mustill and Boyd maintain that such a limited right to appeal is consistent with the parties’ decision to arbitrate instead of litigating. They explain that where parties expressly choose governing law for an agreement, they expect its proper application by the arbitral tribunal (p.441-442). Thus, it would be unjust if an “obviously wrong” decision on an important question of law remained immune from being set right by the court. (p.554). Consequently, English Courts have pragmatically chosen to vary awards in cases where remission would be inappropriate or lead to further costs and delays.

Notably, the English High Court in Fence Gate laid down that the course of judicial intervention under s. 69(7) would be guided by general principles of arbitration, namely “fair resolution by an impartial tribunal without any further unnecessary delay or expense” (para 97). The Court opted to vary rather than remit the award due to two important considerations: display of adverse bias by the arbitrator (compromising standards of impartiality and fairness); and the likelihood of additional costs and delays due to a complete rehearing. On the other hand, the court could speedily resolve the costs issue using existing materials previously placed before the arbitrator without any fresh fact-finding (para 94-95, 102-104). While the court acknowledged that remission was ordinarily more appropriate, especially where insufficient information exists in the award for variation by courts in light of its findings, variation was preferable where the outcome of remission was self-evident and required no further fact-finding (para 92).

This approach aligns with fundamental arbitration principles of expeditious and cost-effective dispute resolution. The decision’s significance for our analysis lies in its development of precise parameters for modification powers, demonstrating that courts can exercise limited variation powers without transforming it into a general appeal mechanism. As Fence Gate shows, the key lies in restricting modification - it did so to scenarios where: (a) the necessary factual record is complete, (b) the outcome is self-evident from the existing material (such as the contract or the award), and (c) remission would either create unnecessary delay and expense or compromise principles of arbitral impartiality and fairness. We find further support in CLVC v Arab Petroleum, where the Court, finding that the arbitrator had wrongly implied terms into guarantees between parties, substituted its conclusion instead of remitting it back to the arbitrator. The decision exemplifies the balance sought to be struck in Fence Gate, especially where remission “serves no substantive purpose” or “affects speedy finality” (paras 90-93).

These benefits of modification powers bear out in Indian jurisprudence as well. Ironically, while the Supreme Court in Hakeem refused to read in modification powers in the Act, it still allowed the lower Court’s modification (albeit under Article 142). After finding that the award suffered from a perversity in law, it allowed the modification for two reasons - first, that remitting the matter to the very tribunal member that committed this perversity would be absurd (para 48). Second, that the arbitration had taken place nearly 10 years ago, making a de novo review impracticable (para 58). This instance highlights the practical limitations of not recognising modification powers in the Act. The point of this is not to say that English law can be imported without due attention to the limited scheme of the Act. Rather, it is to show that a limited power of modification is desirable as well as necessary.

These are all the more important as India’s arbitration regime is repeatedly criticised for its extreme sluggishness and routine abuse of s.34. Courts are clogged with decades old arbitration appeals (para 10),  and corruption (by securing awards through bribes) is not uncommon (para 1.1.9). These realities warrant a pragmatic approach through a limited modification power.

This is rooted in the belief that judicial intervention in arbitral awards can: potentially enhance arbitral integrity and efficiency; reduce the risk of arbitrary decisions and improper conduct of arbitrators; and enhance business trust and confidence. On the other hand, the absence of sufficient judicial scrutiny can “adversely affect the victims of manifestly flawed arbitrations” (p.2). Thus, striking an appropriate balance between competing objectives of finality and fairness requires some degree of judicial intervention. It is true that excessive judicial intervention might undermine arbitration's efficacy as a private dispute resolution mechanism. However, emerging critiques of the traditional balance towards finality emphasise the correctness of awards and advocate for expanded review options, especially in high-stakes disputes where finality may represent an “immeasurable risk” rather than a virtue. In light of this, carefully circumscribed modification powers could enhance both systemic integrity and procedural fairness without unduly compromising finality (p.11). Importantly, limited modification would be particularly effective in addressing serious procedural irregularities like arbitrator bias and corruption, where the usual options of setting-aside and remission might backfire (p.8), as Fence Gate draws attention to.

Notably, the 2024 Viswanathan Committee for reforms in arbitration law suggested introducing a limited modification power into the Act, citing its ability to “provide a quietus to the matter, so as to avoid further litigation” and to “save time, effort and resources for all parties involved. This would strike a balance between preserving arbitral finality and ensuring fairness” (paras.3.25.9 - 11). The Committee proposed permitting modification "in exceptional circumstances" where "strict parameters for setting aside" are met, without requiring fresh evidence (para 3.25.10). This aligns with the preconditions we identified in English jurisprudence. However, while there are certainly benefits to the powers of modification, the Committee’s suggestion lacked clear guidance and defined parameters - to what extent should modification be allowed, given the limited scheme of the Act? The next section addresses this.

Balaswamy’s Overlooked Balance

When seeking to balance the powers of modification alongside the need to minimize judicial interference, it is key to ask what the content of such modification powers would look like. Clearly, unbridled powers to modify awards would stretch the scheme of the Act too far. The Court would effectively substitute the tribunal. The majority in Balaswamy recognizes this tension, holding that “[this concerns] a conflict between equity and justice, on the one hand, and the fetters imposed by the court’s jurisdictional limits, on the other…..therefore, it is crucial to adopt a balanced approach...” (para 25). Following this logic, the majority in their conclusion chalked out a clear demarcation for modification: (para 85):

1. When the award is severable, by severing the “invalid” portion from the “valid” portion of the award;

2. By correcting any clerical, computational or typographical errors which appear erroneous on the face of the record, (as held in Part IV and V of our Analysis);

3. Modifying post award interest in some circumstances (as held in Part IX of our Analysis); and/or

4. Applying Article 142 of the Constitution, albeit cautiously and within the limits of the constitutional power (as outlined in Part XII of our Analysis). [7]

The majority is quite clear in recognising these powers as an exhaustive list. For instance, while modifying costs where it “follows inevitably from the tribunal’s determination of a question of law” (para 10, 46) was argued for and discussed by the Court, it was ultimately not included in its conclusion. 

However, many have rejected the idea that a power of modification (howsoever limited) could find place in the scheme of the Act, and have argued that the majority has merely “cherry-picked” equitable considerations. Others have even noted that this could open the floodgates for Courts to modify awards indiscriminately, contrary to the scheme of the UNCITRAL model law on which the Act is based. Dealing with these objections requires addressing two points – first, what the scope and content of these modification powers may be. To examine this, we refer to precedents where modification powers have been invoked. We will show that applying these powers tends to promote, rather than militate, against the objects of arbitration. Second, it must doctrinally be examined whether the legislature intended to exclude modification powers from s.34. We show that this is not necessarily the case. To begin with, the Court recognises the power to modify only in the following limited circumstances (para 85):

Only severable awards can be modified: 

It is important to note here that this is not per se a substantive limitation on the power to modify – rather, it acts as a precondition to exercising the power to modify. Notably, this principle is also found in s.15 of the old Arbitration Act, 1940. The exact limitation is of much practical significance, preventing modification of ‘composite awards’ (para 29) – where enforceable claims in the final award are a single amount and intertwined with the other claims. This limitation aims to place a stop-gap on excessive judicial interference – if such awards were modified, claims that were not affected by patent illegality (or any other ground in s.34 used to modify) would also be affected by the modification, running contrary to the limited intervention envisaged in s.34. At the very least, this guardrail allays the concern that a supervising Court could overhaul the award altogether; instead, regard must be had to its severability from other claims.

Correcting Computational, Typographical Errors:

This is akin to the tribunal’s limited power in s.33, and thus highly unlikely to result in undue hardship or unprincipled modifications. These concern manifest/inadvertent errors on part of the tribunal – a matter which would otherwise need to undergo the lengthy rigor of remission. Such power has been safely exercised before. For example, in JC Budharaja, the damages awarded were in excess of those claimed. Thus, the Supreme Court through award modification, reduced the awarded damages to the extent of the claims made (para 28). If the traditional ‘setting aside’ procedure was followed, this would lead to further delay merely for the de novo proceedings before the tribunal to alter the amount awarded. Thus, where the grounds for ‘setting aside’ are met, but a simple modification would clearly resolve the issue in the award, modification may be warranted. This is opposed to situations where the award suffers from a fundamental error in its reasoning/interpretation, in which case the Court would have to redo the award, which would be clearly impermissible by s.34.

Curiously, while Justice Viswanathan’s jurisprudential approach to modification rebuffs the power to modify interest rates, it agrees with the majority on this point. He holds that even though s.33 empowers tribunals to correct such errors, the supervisory Court under s.34 would also hold these powers despite the absence of textual or legislative basis for the same (para 141). In doing so, he resorts to the maxim “actus curiae neminem gravabit” – that the actions of the court should not harm any parties. Thus, despite his otherwise textualist, doctrinal approach, he nevertheless identifies and plugs this deficiency in the Act.

Modifying Interest Rates: 

This is ostensibly the most controversial power, and rightly so. After all, the Court’s new guardrails of modification are otherwise identical to modification powers in the old Arbitration Act, with the addition of the power to modify interest rates.[8] Critics fear that Courts will use this power to award interest on its whims and fancy. However, we suggest that this may not be the case. Take Arvind Seth - (notably, in 2021 after Hakeem) a division bench of the Bombay High Court held that the single judge’s alteration of the interest rate awarded did not constitute an impermissible modification, since the modified interest rate was following the terms of the contract. The Court said that had the interest rate been higher, this would constitute an impermissible modification (para 12). This provides us with some inkling of what the guardrails for modification powers would look like. While changing the interest rate would be a modification in the literal sense - the Bombay High Court held that it would not be an impermissible modification within the scheme of the Act since it toed the lines of the contract. Similarly, in Ssangyong, J.Nariman modified the award (albeit under Article 142) by upholding the minority award because “there [was] a minority award which awards the appellant its claim based upon the formula mentioned in the agreement between the parties” (para 49). A closer reading of this case would reveal that it is incorrect to characterise such a modification as a rewriting of the contract (as has been argued elsewhere), since the Court adopts the minority award for the limited purpose of correcting the error of law in the computation of damages. In cases such as these, the contract anchors the modification powers of the Court.

Is the power to modify interest rates vulnerable to misuse, potentially amounting to a merits review? While this concern may seem valid, the majority’s interpretation makes such misuse unlikely. First, any modification requires one of the grounds under s.34 to be met. In other words, it must be found that the awarded interest rate would fall foul of the public policy of India or be patently illegal on its face – which by itself, is a difficult threshold to meet. One has a fairly weak conscience if only interest rates can shock it! As such, one scarcely needs to be concerned about modification powers being exercised at this stage. Second, because the guardrails laid down by the majority account for situations like this, it would be incorrect to interpret the power to modify interest as a carte blanche exercise in picking the interest rate via a lottery. The Court in Balaswamy was careful to note that modification will not involve a merits review (para 106). As we have noted in the examples of the judgements above, modification does not per se involve a merits review. It is perfectly feasible for the Court to alter a patently illegal awarding of interest rate without affecting the substantive merits of the case.

A limited power of modification is sensible in the scheme of the Act - instead of following the lengthy process of a de novo s.34 review on remission, the Court adopts the available view without exercising any independent judgement of its own. However, Justice Viswanathan dissents from this, holding that the proper course of action would be to ‘record reasons and remit the matter to the arbitrator’ (para 135).  He argues that there is a ‘fallacy’ in arguing that undue hardship would be caused to parties to the arbitration. Two reasons are offered– first, that arbitration proceedings could easily begin afresh under s.43(4); a process already contemplated in the scheme of the Act (para 103). Second, that parties voluntarily assented to the existing scheme of the Act by undertaking to arbitrate instead of litigate (para 83). However, neither of these reasons account for the concerns over delay and backlog. Beginning arbitration afresh is an extremely tedious process for parties that have sunk several millions and years into hearings and arbitrator fees. It would be difficult to say that merely because parties opted for arbitration, they ought to be subject to the vicissitudes that come with a sluggish regime like India’s.

However, while modification powers of courts are desirable in practice, we must now ask whether modification can doctrinally be reconciled with the text of the Act. While it is rightly argued that it cannot be located strictly in the text of ‘setting aside’ in s.34(1), the text is broad enough to encompass the remedy of modification. This is because the right in s.34 is not necessarily limited by the prescribed remedy. In s.34(1), the right is the “recourse to a Court against an arbitral award”, while the remedy is an application of “setting aside” - and as the single judge notes in Balaswamy, the right of recourse is a comprehensive and inclusive one, and the “the manner in which a remedy is to be sought can never curtail or limit the right” (para 51). Justice Viswanathan refers to the opening words of s.34(2), which lay down when a Court can set aside awards : “an arbitral award may be set aside by the court only if the party make out the grounds..” to reject this argument. He uses the words “only if” to argue that the power to modify cannot be read in s.34, since  the text indicates that setting aside the award is the only available recourse. However, as we have noted, the word ‘only’ in s.34(2) does not qualify the form of recourse (i.e., setting aside), but the grounds (i.e., patent illegality, etc).

Even the travaux of the UNCITRAL Model law supports this reading – the discussion on s.34 being ‘exclusive’ revolved around the grounds enumerated in the provision. There was no discussion on whether the form of recourse itself could be limited. At best, one could conclude that the model law was silent on the matter. In such situations, it is settled Indian law that the Indian Arbitration Act can be moulded to suit its local conditions (para 12) – which is accomplished by the recognition of these modification powers. Moreover, it cannot be said that the Parliament ‘intended’ to do away with modification powers. Unlike nations such as the UK and Singapore, which modified their arbitration legislations, India almost entirely  replicated the 1996 UNCITRAL Model law. There was no evidence of intelligent or conscious design to exclude the powers of modification. The UNCITRAL Model law was designed for international arbitration – but unlike countries such as UK and Singapore, the Indian legislature did not tinker with the Model law to suit domestic arbitration. By and large, it remained a replica of the Model law. Indeed, this is what has led to the legislative deficiencies we scramble to plug today. Thus, to say that Parliament ‘intended’ to do away with the power is to take a rather narrow jurisprudential view of Parliamentary intent. Echoing this approach, Lord Denning in Seaford Court Estates Ltd. v. Asher (1949) held:

"when a defect appears a Judge cannot simply fold his hands and blame the draftsman. He must set to work on the constructive task of finding the intention of Parliament, and he must do this not only from the language of the statute, but also from a consideration of the social condition which gave rise to it and of the mischief which it was passed to remedy, and then he must supplement the written word so as to give 'force and life' to the intention of the Legislature.”

Recognizing modification powers is thus not an act of judicial invention or overreach, but of giving effect to a hastily enacted Act, the limited ambit of which hinders the objects of arbitration, ignoring the practical realities of the regime. This method of interpreting the Act outside its strict textual limitations is not without precedent. Although the Act lacks a provision for emergency arbitrators, Indian Courts have recast Sections 9 and 17 to give effect to their awards, keeping in mind promoting the objects of arbitration. It is true that recognizing modification powers would be in a sense, fitting a square peg in a round hole. However, the same could be said of the judicial readings of emergency arbitrators, the group of companies doctrine in s.8, or the A. Ayyaswamy position on the arbitrability of fraud in s.16.[9] Though the text is silent, Parliament has often read provisions in a manner consistent with global practices and the broader objectives of arbitration. As such, it is hasty to decry Balaswamy for its doctrinal inconsistency.

Thus, one can gather that modification powers can promote the objects of arbitration by promoting speedy resolution and limiting judicial interference – particularly given the narrow domain in which they have been engrafted. Arbitrarily exercised modification powers would be (and have been) shot down - in SV Sundaram, the Supreme Court rejected the modification of awards since the modified interest awarded by the Court was “without any legal basis” and where the damages awarded where “without justification” (paras 41- 46) - essentially, that these modifications were an exercise of the Court’s unanchored discretion.

The consequence of limiting the right to the remedy would be to interpret the recourse in s.34 in an all-or-nothing manner - either uphold or set aside the award. Even where it is severable, it would have to be remitted to the tribunal under s.34(4), reviving the same concerns of delay and bias, thereby militating against the objects of arbitration. Thus, appropriately delimited, recognising the powers of modification has the potential to promote a more arbitration-friendly regime.

Conclusion

It is no one’s case that a clarificatory Parliamentary amendment is undesirable. However, the majority in Balaswamy has taken up the mantle to see beyond the all-or-nothing approach to s.34 powers. While the arguments on textual limits are appealing, there are several reasons why limited powers of modification can be consistent with the scheme of the Act while simultaneously promoting the objects of speedy disposal of cases. While the precise scope of these powers is yet to play out in cases, this piece has examined precedent to suggest that it may not necessarily turn India into an ‘arbitration-unfriendly regime’ – a pressing concern for critics. Instead, it can address the various concerns faced by India’s arbitration regime -- its delays and corrupt practices -- with appropriate judicial intervention. While the tussle between principles of judicial interference and arbitral finality is a never-ending one, developments in arbitration law continue to refine rough edges to balance them better. In Gayatri Balaswamy, the Supreme Court offers an opportunity to not merely reaffirm first principles, but to refine them in service of a more efficient framework of arbitration.

---

*Prem Vinod Parwani & Pranshu Gupta are B.A., LL.B. (Hons.) students at the National Law School of India University (NLSIU), Bengaluru

---

[1] Notably, this applies only to domestic arbitrations and not international arbitrations.

[2] Section 37 of the Act is concerned with appeals.

[3] Here, the Supreme Court altered the date from which the interest would be calculated.

[4] The text of the legislation is as follows:

s.34 (1) Recourse to a Court against an arbitral award may be made only by an application for setting aside such award in accordance with subsection (2) and sub-section (3).....

(4) On receipt of an application under sub-section (1), the Court may, where it is appropriate and it is so requested by a party, adjourn the proceedings for a period of time determined by it in order to give the arbitral tribunal an opportunity to resume the arbitral proceedings or to take such other action as in the opinion of arbitral tribunal will eliminate the grounds for setting aside the arbitral award.

[5] Project Director National Highways v M. Hakeem AIR 2021 SC 3471 [41].

[6] Similar modification powers exist across other Model Law jurisdictions. For example, s.49(8)(b) of the Singapore Arbitration Act 2001 empowers courts to vary arbitral awards where questions of law substantially affect parties' rights, and the tribunal's determination was "obviously wrong", raises "serious doubt", or where court intervention is "just and proper". Notably, the Singapore Academy of Law's 2020 Report advocates extending similar powers to international arbitration through proposed amendments to the International Arbitration Act 1994. The Australian Commercial Arbitration Act 1986 likewise codifies judicial modification powers under sections 38(3)(a) and 38(7), further evidencing a recognized need for limited court modification in appropriate circumstances.

[7] An important qualification must be noted - modification powers can also be exercised de hors the Act by virtue of Article 142 of the Constitution. This is not the subject of our enquiry. Rather, we are concerned with locating a general power to modify under Section 34 of the Act. The Supreme Court has upheld the constitutionality of the art. 142 route to modification on multiple occasions, and this power  (while dealt with and answered), was not not strictly impugned in the Gayatri Balaswamy reference. As such, we deal only  withthe power to modify awards under the Act.

[8] Arbitration and Conciliation Act 1940, s15; “15. Power of Court to modify award:- The Court may by order modify or correct an award – (a) where it appears that a part of the award is upon a matter not referred to arbitration  and such part can be separated from the other part and does not affect the decision on the matter  referred ; or (b) where the award is imperfect in form, or contains any obvious error which can be  amended without affecting such decision ; or (c) where the award contains a clerical mistake or an error arising from an accidental slip or omission.”

[9] In both instances, the text is silent on the threshold – s. 8 simply says party claiming “through or under”, while s.16 is silent on whether (and what nature of fraud) vitiates an arbitration agreement. In A. Ayyaswmy, the Court held that only ‘serious’ allegations of fraud involving ‘public elements’ would be non-arbitrable, despite such mandate lacking statutory sanction in s.34.