Introduction
The justification for using alternative dispute resolution (ADR) in India is typically presented as a consequence of the inability of the courts to handle all of the disputes that come before them, so that certain types of disputes should be moved away from the courts and into other types of processes. While this view helps explain why ADR is administratively attractive, it fails to consider adequately the experience of those involved in the disputes. A party may settle a dispute quickly but never receive payment. Another party may agree to settle because he or she cannot afford to continue litigating. An online procedure may be available throughout the country and still be unusable by someone who cannot read the interface. In each of these examples, “disposal” and “justice” represent two very different outcomes.
Constitutional access to justice is a broader concept than the management of the docket. Articles 14 and 21 of the Constitution protect equality and life and personal liberty respectively, and Article 39A directs the State to secure that the operation of the legal system promotes justice on a basis of equal opportunity and, in particular, to provide free legal aid. Collectively, these articles provide a basis for determining whether procedural arrangements make it possible for people actually to obtain and use remedies. The question is therefore not simply whether ADR reduces the number of pending cases, but whether it creates conditions under which people can obtain and use remedies. This article asks how the various forms of ADR used in India differ, under what circumstances speed and lower cost coexist with meaningful consent and procedural fairness, and how institutions should evaluate success beyond counting the settlements achieved. The study is doctrinal, analysing legislation and leading judicial decisions together with a normative evaluation of institutional incentives; it does not estimate settlement rates or claim a causal relationship between ADR and reduced pendency. Its thesis is that ADR is best viewed as a differentiated part of the justice system whose legitimacy rests on appropriate allocation, active participation and enforceable outcomes.1
Constitutional access to justice as a benchmark
In Anita Kushwaha v. Pushap Sudan, the Supreme Court treated access to justice as a facet of Articles 21 and 14 and identified an effective adjudicatory mechanism, reasonable accessibility in terms of distance, speedy adjudication and affordability as its essential elements. Although the case concerned the power to transfer cases, the framework the Court developed is useful for evaluating ADR. A procedure cannot be evaluated solely on the basis of its length if users cannot reach it, cannot afford to use it or do not receive meaningful relief.2
Two implications arise from this. First, speed must encompass the entire journey to a usable remedy. The interval between referral and signature excludes preparation, prior negotiations, unsuccessful settlement attempts and enforcement. Second, affordability is relative. The same fee may be manageable for a corporation but prohibitive for an individual claimant. Wages lost in attending hearings, travel expenses, translation costs and legal advice may be as significant as institutional fees. These are evaluative implications proposed here from the constitutional benchmark; they are not a prescription by the Court for measuring access to justice.
Access also requires a realistic opportunity for a party to refuse an unsuitable settlement. If adjudication is indefinitely delayed, nominal consent may indicate financial exhaustion rather than true acceptance. This does not mean that every compromise is involuntary. Parties may value certainty and may trade part of a contested claim for prompt performance. The concern arises when institutions encourage settlement without ensuring that participants understand their rights, the available alternatives and the consequences of signing.
Public adjudication performs functions that private settlement may not perform: authoritative interpretation, public reasoning and the determination of claims affecting third parties.3 Accordingly, an ADR policy should preserve space for decisions that clarify rights or address recurring unlawful conduct. The constitutional approach proposed here acknowledges that settlement has value while refusing to treat an individual compromise as sufficient evidence of institutional justice.
Differentiating the principal mechanisms
A. Court referral and procedural suitability
Section 89 of the Code of Civil Procedure, 1908, and Order X, Rules 1A to 1C, provide a procedural context for referring disputes to alternative processes. In Afcons Infrastructure Ltd. v. Cherian Varkey Construction Co. (P) Ltd., the Supreme Court clarified how this framework operates and distinguished the consent requirements of the different processes. Specifically, a court cannot use Section 89 to compel arbitration where there is no agreement to arbitrate. In practice, referral requires diagnosis. A dispute arising out of a commercial relationship may benefit from facilitated negotiation; a dispute requiring a binding determination may be appropriate for arbitration if validly agreed upon and legally arbitrable. A case involving immediate protection or broader rights may require judicial attention. Institutions should record why a process is appropriate rather than refer a dispute merely because an ADR centre exists.4
B. Arbitration and the price of private adjudication
Arbitration involves a binding decision rendered by a tribunal, as opposed to a settlement based on continuing agreement. The Arbitration and Conciliation Act, 1996 provides for equal treatment of the parties and a full opportunity for each to present its case; it also regulates disclosure, challenges and enforcement. These protections explain why arbitration belongs in an analysis of access to justice even though it is commonly associated with commercial autonomy. Arbitration’s primary advantage lies in the ability to design a procedure proportionate to the disagreement. A relatively minor documentary issue may not require extensive oral testimony, multiple scheduling conferences or extensive written submissions. On the other hand, technical complexity may justify expert testimony and additional time. The relevant comparison is between procedures capable of fairly resolving the same dispute, not between an idealised arbitration and the slowest possible lawsuit.5
Private adjudication can impose tribunal fees and administrative costs in addition to lawyers’ fees. A small claim may become economically unfeasible if it is processed using procedures developed for high-value commercial disputes. Unequal resources and repeat appearances may also undermine confidence in impartiality. Formal safeguards do not answer whether users can afford to invoke them. Institutions should therefore disclose the anticipated stages, estimated costs and simplified alternatives before significant expense is incurred. Sections 29A and 29B regulate aspects of arbitral timing and procedure, but post-award proceedings may remain relevant. A fair evaluation should include the time to award and the time to satisfaction, as well as the cost and outcome of any court proceedings relating to the arbitration.6
C. Mediation and conciliation
Mediation allows parties to reach a mutually acceptable resolution with the assistance of a neutral third party. Its distinctive advantage is its flexibility in addressing remedial needs. Payment schedules, revised delivery arrangements and agreed communication protocols may satisfy needs that a binary decision cannot. Flexibility is beneficial when parties have sufficient information and are free to reject a proposed resolution. It becomes problematic when the desire to reach a settlement overrides consideration of unequal bargaining power.
Part III of the Arbitration and Conciliation Act (sections 61 to 81) establishes a framework for conciliation, under which sections 73 and 74 make a settlement agreement signed by the parties final and binding and give it the same status and effect as an arbitral award on agreed terms. This illustrates why the route taken to reach a settlement matters. A settlement reached informally should not automatically be considered equivalent to an arbitral award, a decree or an agreement enforceable under a different statute. The appropriate formalities and governing provisions must therefore be determined in each case.7
The Mediation Act, 2023 establishes a separate statutory framework. The enacted text addresses pre-litigation mediation, exclusions, time limits, the enforcement of and challenges to mediated settlement agreements, and online mediation. The Act is brought into force provision by provision. The notification issued on 9 October 2023 brought certain provisions into effect, and the enactment of a statute does not by itself establish that all of its substantive provisions are in effect. Section 61, for instance, read with the Sixth Schedule, would substitute new provisions for sections 61 to 81 of the Arbitration and Conciliation Act, but it was not among the provisions notified.8 In a parliamentary answer of December 2023, the Government stated that some provisions had been notified, that the rules recommended for the operation of the Act were under consultation and that, the Act being at a nascent stage of implementation, no time frame could yet be indicated for the institutions and infrastructure needed to implement it.9 Statements regarding the enforceability of mediated settlements, or the replacement of an earlier regime, must therefore be verified against the relevant commencement notifications and operational frameworks, and the current legal status of a provision confirmed before it is relied on in practice. This article evaluates the statutory design without assuming that all of its provisions are currently in effect.
D. Mandatory initiation and voluntary settlement
Section 12A of the Commercial Courts Act, 2015 requires pre-institution mediation before a suit that does not contemplate any urgent interim relief is instituted. In Patil Automation Pvt. Ltd. v. Rakheja Engineers Pvt. Ltd., the Supreme Court held the requirement mandatory and held that a suit instituted in violation of it must be visited with rejection of the plaint under Order VII Rule 11, a declaration it made effective from 20 August 2022. The requirement concerns entry into the prescribed process; it does not authorise a mediator to dictate the settlement.10
A mandatory first step can create an opportunity for negotiation before litigation positions harden. It can also increase costs if the defendant refuses to participate or uses the process strategically. The statutory framework addresses time and limitation, but institutional service quality remains critical. Timely appointment, clear notice and prompt documentation of failed processes are necessary if the requirement is to facilitate rather than impede access.11
The safeguard suggested here is straightforward: assess compliance with legitimate procedural responsibilities without treating the refusal of an offer as institutional failure. An individual who receives advice and reasonably rejects an inadequate settlement has used mediation successfully. The process can identify the real issues and allow for a more focused lawsuit even when no settlement is reached. Performance systems should recognise the value that mediation provides without creating a settlement where none exists.
E. Ordinary Lok Adalats and Permanent Lok Adalats
Ordinary Lok Adalats facilitate compromise under the Legal Services Authorities Act, 1987. Their awards are deemed to be decrees of a civil court and are final and binding, but their settlement function must not be equated with ordinary adjudication. In State of Punjab v. Jalour Singh, the Supreme Court emphasised that Lok Adalats have no adjudicatory or judicial functions, so that an ordinary Lok Adalat cannot decide the merits of a case where settlement is absent. This maintains the distinction between an agreed result and an imposed decision.12
NALSA states that no court fee is payable when a matter is filed in a Lok Adalat and that the court fee originally paid is refunded if a referred court matter is settled. This reduces one identifiable element of user cost. It does not establish that every settlement is substantively adequate, that no other expense is incurred or that agreed payments are ultimately made.13
Permanent Lok Adalats for public utility services have a different statutory role. Under Section 22C, they first attempt conciliation and, where the parties fail to reach agreement, they must decide the dispute if it does not relate to an offence, within their jurisdiction and subject to the statutory conditions. Section 22D requires them to be guided by the principles of natural justice, objectivity, fair play and equity. Describing both institutions simply as consensual settlement forums would obscure this fundamental difference.14
Preparation for ordinary Lok Adalats should include verifying the validity of the claim, explaining the offer and confirming that the individual signing has authority and understands the implications. Preparation for Permanent Lok Adalats should also include informing users that the proceeding may move from conciliation to determination. An accessible forum must inform users of its powers; proximity alone does not make its procedures comprehensible.
F. Online dispute resolution (ODR)
ODR can reduce the need for travel and enable parties to exchange documents and proposals asynchronously. NITI Aayog’s ODR policy plan presents ODR as relying on asynchronous communication and removing the need for the parties’ physical presence, and it builds its recommendations around increasing access to digital infrastructure, building capacity and building trust in ODR, within a soft-touch regulatory model and a phased implementation. Technology should therefore be evaluated as an arrangement for delivering a process, not as proof that the process is accessible.15
A usable service should offer assisted access, understandable language and a viable alternative when the connection is lost. Shared phones can hinder private communication. Automated reminders can assist attendance, while interfaces that hide rejection options can distort consent. These are design risks requiring testing, not empirical findings about every Indian platform. Before a service is expanded, actual user journeys should be examined. Automated tools that summarise documents or propose options should remain accountable. Users should know when automation is in use, have a means of correcting errors and have access to a human decision-maker or mediator as needed. Efficiency gains should be measured alongside errors, abandonment and exclusion. A digital procedure that saves staff time by transferring difficult tasks to unsupported users may reduce access even as it reduces administrative costs.
Why disposal statistics cannot establish justice outcomes
A. Defining the correct unit of measurement
Counting settlements shows how many matters reached a specific procedural endpoint. It does not show how many disputes were resolved permanently or how many pending cases would otherwise have required adjudication. A pre-litigation settlement may prevent a future suit, but that counterfactual cannot be assumed for every matter. Similarly, a pending case referred late in its life has already consumed judicial resources. These categories should be reported separately. A credible evaluation begins by defining a cohort: all eligible matters entering a programme during a specified period. It should report referrals, attendance, withdrawals, settlements, failures and cases awaiting an outcome. Reporting only successful agreements presents a favourable image by excluding individuals who could not participate. Follow-up should then assess compliance, further proceedings and unresolved obligations, with appropriate privacy protections.
B. Time and cost across the entire dispute
Time should be measured from first institutional contact to effective relief, while separately identifying the stages controlled by the ADR provider. Both the median and the slower end of the distribution matter. A programme may perform well for simple claims but leave complex or vulnerable users waiting significantly longer, and averages alone can mask this disparity. Comparisons between institutions handling different types of disputes should be explained. Cost assessment should include official charges and reasonable estimates of travel, representation, interpretation and lost working time.16 These components should remain visible rather than being collapsed into a single unexplained figure. Compliance over time matters for instalment settlements: recording the full amount promised as recovered would overstate its practical value. Failed processes should also be included, since their costs form part of the cost of accessing the system.
C. Voluntariness and distributive consequences
A signed agreement demonstrates assent, but it does not by itself establish informed understanding or the absence of coercion.17 Evaluation should therefore ask whether users understood their options, had access to advice and felt able to decline. Surveys should be administered independently wherever possible. Questions posed by the same entity that encouraged settlement may discourage honest feedback, particularly when participants expect to need its assistance in future. Comparative outcomes require careful interpretation. A lower recovery may reflect weak evidence, negotiated instalments or a legitimate trade-off for certainty; it does not in itself indicate exploitation. Conversely, high satisfaction immediately after settlement may reflect relief from stress rather than an understanding of a waiver. Quantitative indicators should be read together with anonymous reviews of process quality and follow-up on performance. No single indicator should replace judgment.
A practical reform agenda
A. Early screening and supported choice
Intake should identify urgency, representation, language, disability-related needs and circumstances affecting safe participation. The intake officer should explain the available processes and their legal consequences in plain terms. Screening is not permission to decide the merits prematurely. Its purpose is to identify procedural needs and to prevent an inappropriate referral from consuming time or putting unnecessary pressure on a participant. Legal services institutions already operate within a statutory eligibility framework, and ADR support should link eligible users to that framework. Other bodies can provide limited advice at fixed prices and understandable information, while maintaining the distinction between neutral facilitation and representation.18
B. Proportional procedures and capable institutions
Procedures should be tailored to the amount, evidence and significance of the dispute. Shorter documentary procedures may be appropriate, predictable fee schedules may be established and transparent mechanisms for selecting neutrals may be implemented. Quality assurance should examine training, conflicts of interest, complaints and the treatment of unrepresented parties. A credential alone cannot demonstrate that a practitioner manages pressure, misunderstanding or unequal participation effectively. Government entities need particular administrative attention. A representative without authority to negotiate cannot make a settlement process effective. Departments should designate officials authorised to consider proposals, provide the necessary records and record reasons for accepting or rejecting an agreement. Such arrangements should allow accountable decision-making without making settlement itself a mandatory objective. This recommendation is administrative and does not presume that every public dispute is legally amenable to compromise.
C. Designing for performance and enforcement
Settlement agreements should specify the parties, obligations, dates, payment methods and consequences of default with sufficient clarity. The applicable legal route should be stated accurately. An agreement that expresses general goodwill but leaves performance unclear can generate another dispute. Where obligations extend over time, users should receive a clear timeline and instructions on the remedies available if performance is deficient.19,20
Institutions can provide follow-up without reopening confidential negotiations. Recording whether an obligation was fulfilled and whether enforcement was necessary is different from publishing the parties’ bargaining positions. Aggregate reporting can protect confidentiality while revealing whether a programme produces tangible benefits. Data collection should have a defined purpose, restricted access and a retention policy, rather than serving as an indiscriminate repository of sensitive disputes.
D. Independent evaluation and balanced incentives
A pilot programme should disclose its evaluation criteria in advance and compare similar categories of disputes. Where feasible, phased implementation or carefully matched comparison groups can improve understanding of whether changes in outcomes are attributable to the intervention. Any causal claim must disclose its assumptions and limitations. This article proposes such a research design; it does not claim to have carried it out.
Institutional incentives should reward fair handling, timely communication and sustained compliance alongside appropriate settlement. A neutral who identifies an inappropriate matter or facilitates an informed decision not to settle should not be penalised merely because a disposal target was not met. Funding and appraisal arrangements should support this broader definition of professional competence. Otherwise, even well-designed procedural safeguards may be undermined by pressure to maximise signatures.
Conclusion
ADR can contribute to speedy, affordable and accessible justice in India, but its contribution depends on the specific mechanism used and the conditions under which individuals participate. Arbitration provides adjudication with procedural flexibility; mediation and conciliation support negotiated resolutions; Lok Adalats provide statutory settlement options. Permanent Lok Adalats have a distinct, conditional adjudicatory role, and ODR changes how processes are accessed and delivered. Treating them as interchangeable obscures their differing costs, powers and protections. Answering the questions this article poses therefore requires a change in the standard of evaluation. An institution should ask whether users received understandable, voluntary (where applicable), affordable and enduring relief, including an effective method of enforcing any agreement reached. Time to signature and the number of disposals are relevant but insufficient. Screening, independent assistance, proportional procedures, assisted access and transparent evaluation of outcomes can connect institutional efficiency with constitutional access to justice. Reducing pendency remains desirable, but it is strongest as the result of a justice system that resolves disputes well.
*****
Footnotes
1. India Const. arts. 14, 21, 39A.
2. Anita Kushwaha v. Pushap Sudan, (2016) 8 SCC 509, ¶¶ 31–33 (India).
3. See Owen M. Fiss, Against Settlement, 93 Yale L.J. 1073, 1085 (1984) (the judicial task is “to explicate and give force to the values embodied in authoritative texts such as the Constitution and statutes”, a duty that “is not discharged when the parties settle”).
4. Code of Civil Procedure, No. 5 of 1908, § 89, Order X, rr. 1A–1C (India); Afcons Infrastructure Ltd. v. Cherian Varkey Construction Co. (P) Ltd., (2010) 8 SCC 24 (India).
5. Arbitration and Conciliation Act, No. 26 of 1996, §§ 12, 18, 34, 36 (India).
6. Arbitration and Conciliation Act, No. 26 of 1996, §§ 29A–29B, 34, 36–37 (India).
7. Arbitration and Conciliation Act, No. 26 of 1996, §§ 61–81, especially §§ 73–74 (India).
8. Mediation Act, No. 32 of 2023, §§ 1(3), 5–6, 18, 27–28, 30, 61, sch. 6 (India); Ministry of Law & Justice, Notification S.O. 4384(E), Gazette of India, Extraordinary, pt. II, sec. 3(ii) (Oct. 9, 2023), https://egazette.gov.in/WriteReadData/2023/249277.pdf. The notified provisions were §§ 1, 3, 26, 31–38, 45–47, 50–54, 56–57.
9. Ministry of Law & Justice, Department of Legal Affairs, Implementation of Mediation Act, 2023, Rajya Sabha Unstarred Question No. 2192 (answered Dec. 21, 2023), https://sansad.in/getFile/annex/262/AU2192.pdf?source=pqars.
10. Commercial Courts Act, No. 4 of 2016, § 12A (India); Patil Automation Pvt. Ltd. v. Rakheja Engineers Pvt. Ltd., (2022) 10 SCC 1, ¶¶ 99, 113 (India).
11. Commercial Courts Act, No. 4 of 2016, § 12A(3)–(5) (India); Commercial Courts (Pre-Institution Mediation and Settlement) Rules, 2018, rr. 3–5 (India).
12. Legal Services Authorities Act, No. 39 of 1987, §§ 19–21 (India); State of Punjab v. Jalour Singh, (2008) 2 SCC 660, ¶¶ 8–12 (India).
13. National Legal Services Authority, Lok Adalats, https://nalsa.gov.in/lok-adalats/ (last visited Oct. 15, 2026); see also Legal Services Authorities Act, No. 39 of 1987, § 21(1) (India).
14. Legal Services Authorities Act, No. 39 of 1987, §§ 22A–22E, especially §§ 22C(8), 22D (India).
15. NITI Aayog, Designing the Future of Dispute Resolution: The ODR Policy Plan for India iii, 65 (2021), https://www.niti.gov.in/sites/default/files/2021-11/odr-report-29-11-2021.pdf.
16. See generally Martin Gramatikov, Maurits Barendrecht & Jin Ho Verdonschot, Measuring the Costs and Quality of Paths to Justice: Contours of a Methodology, 3 Hague J. on Rule L. 349 (2011), https://doi.org/10.1017/S1876404511200101 (setting out a methodology for measuring the costs, procedural quality and outcome quality of paths to justice from the perspective of the user).
17. See Fiss, supra note 3, at 1075–76 (settlement as “the civil analogue of plea bargaining”, in which “[c]onsent is often coerced”, and the poorer party may be induced to accept less in order to accelerate payment).
18. Legal Services Authorities Act, No. 39 of 1987, §§ 12–13 (India).
19. Arbitration and Conciliation Act, No. 26 of 1996, §§ 30, 31, 36 (India).
20. Mediation Act, No. 32 of 2023, §§ 19, 27 (India).