Although the term ‘Constitutional Morality’ featured in Dr. Ambedkar’s speech introducing the Draft Constitution, it was not expressly incorporated into the text of Article 13, which recognises power of judicial review. In 1951, the newly established Supreme Court of India declined to accept that the ‘spirit of the Constitution’, a term often used interchangeably with constitutional morality, constituted a legitimate ground for judicial review, holding that “the court should construe the language of Article 13 according to the established rules of interpretation and arrive at its true meaning uninfluenced by any assumed spirit of the Constitution.” This position was reiterated in State of Bihar v. Kameshwar Singh, despite Dr. Ambedkar appearing as a counsel and urging the Court to accept the spirit of the Constitution as a valid test of statutory legitimacy.

The tide turned with the landmark decision in Kesavananda Bharati v. State of Kerala, wherein the Court, while adjudicating upon the scope of the Parliament’s power under Article 368, held that the “basic structure” of the Constitution could not be amended. In its subsequent decisions, the Court expanded this basic structure, which now encompasses various fundamental rights and constitutional norms. The recognition and expansion of the basic structure doctrine created a clear anomaly: while a constitutional amendment passed by the Parliament could be struck down for breaching this ever-expanding doctrine, an ordinary legislation could be invalidated only on the limited ground of infringing Fundamental Rights in the Constitution. It is perhaps to remedy this very anomaly that constitutional morality emerged as a test for legislative validity, effectively elevating the undefined moral code within the Constitution over its express text.

The doctrine of constitutional morality played a crucial role in ending the prohibition on the entry of menstruating women into the Sabrimala temple. Article 25 and 26 of the Constitution guarantee religious freedoms to individuals and groups, respectively, subject to public order, health, and morality. Previously, the Court, while striking down the exclusionary practice, held that the term ‘morality’ in the aforesaid provisions must be understood as constitutional morality and not morality as perceived by individuals or religious sects. Thus, by reading constitutional morality into Article 25 and 26, the Court was able to uphold women’s right under Article 25 to enter the temple, despite the same being at loggerheads with the existing religious beliefs, while simultaneously limiting the ambit of denominational rights under Article 26 by denying constitutional protection to religious practices that are discriminatory, oppressive and exclusionary.

While advancing arguments in the ongoing review proceedings, the Solicitor General of India urged the Supreme Court to reinterpret the term ‘morality’ in Article 25 and 26 as “public morality and not some vague concept of constitutional morality”. If accepted, such an interpretation would not only result in denying menstruating women entry into the Sabrimala temple, but would also subject the legitimate exercise of religious freedoms, particularly by individuals belonging to marginalised groups, to popular yet regressive religious sentiments. Importantly, the Court was further urged to strike down the doctrine of constitutional morality altogether, with the Solicitor suggesting that “it better dies sooner”. However, such a declaration would dilute the Supreme Court’s recent judgments that have emboldened India’s human rights framework, and would crucially impact the fate of numerous constitutionally significant cases pending before the Court.

There are projects in some jurisdictions – including China, Brazil, and Argentina – as to how these technologies could be employed in the judicial decision-making process to reduce case backlogs and enhance the delivery of justice. Contrarily, some countries – such as the UK, Ireland, New Zealand, and Canada – have taken an opposing stance to prevent the adoption of LLMs in core judicial functions, including determination and adjudication. Instead, they permit LLMs as support tools for judges and court staff to perform routine work like summarising text, speech writing, searching for information, sources, and precedents, enhancing clarity and linguistic flourish, translating, and transcribing.

However, such ‘innocent’ use of LLM tools in routine or secondary court functions has the potential to influence judicial cognition and thereby affect the core judicial decision-making process. Institutional permission and repetitive use of these tools may lead to several cognitive effects, such as automation bias, cognitive offloading, anchoring, and the framing effect. Automation bias leads to over-reliance on automated recommendations, while cognitive offloading reduces mental effort through external tools. Anchoring causes over-dependence on initial information, and the framing effect shows how decision-making shifts based on information presentation.

Judging involves reading the case materials, identifying the issues, and developing the reasons by themselves, whereas a preliminary recommendation produced by an LLM system has the potential to serve as a starting point for such cognitive involvement of judges, anchoring the point of departure. On the other hand, when a judge outsources the meaning of legal lexicon, a summary of long texts, or the identification of precedents to an LLM, they ultimately engage in cognitive offloading. This cognitive offloading can lead to a superficial review process where judges make only ‘surface-level’ scrutiny or modifications to LLM-generated text, bypassing the deep cognitive work required to fully understand the legal reasoning and verdict.

Therefore, simply relying on a human judge to use LLM systems in preparatory stages, or to audit the veracity of the outcomes of such systems subsequently in the judicial decision-making process, is potentially problematic. Though judges are trained to be impartial, cognitive and behavioural science demonstrates that no decision-maker is immune to systematic cognitive biases. Decades of behavioural research demonstrate that professional expertise does not eliminate susceptibility to anchoring, framing, or automation bias. Judges, like other professionals, work within cognitive constraints.

These cognitive defects are significant because human judges are regarded as the principal safeguard to check and control any type of bias and are also regarded as neutral decision-makers to ensure justice. In the circumstances of heavy judicial workloads, exhaustion, and in some cases institutional pressure to deliver timely decisions with limited resources, cognitive issues can arise. As a result, it can undermine the fundamental principles of fair trial, including judicial independence, the judicial duty to provide well-reasoned decisions, and ultimately the rule of law.

The inherent persuasive features of LLMs that randomly generate text with human-like reasoning and confidence, in spite of factual or legal accuracy, may make human judges vulnerable to this uncritical adoption of LLMs even as a support tool for judicial decision-making. This creates a deep tension with the fundamental and uncompressible judicial duty of independent reasoning, signifying that without precise safeguards and a deep understanding of this cognitive susceptibility, the integration of LLMs in adjudication, even as secondary tools, could unintentionally undermine the deliberative process it seeks to augment. Hence, in an increasingly technology-driven judiciary, it is critical to examine whether judges’ cognitive limitations prevent them from effectively guarding against aggravating this tension.

Using two recent cases (Dr Jaya Thakur v. Union of India and Suhas Chakma v. Union of India), delivered by two different judicial benches of the Supreme Court (‘Court’), it appears that the contours of PILs seem to be changing. Courts now seem to be conscious of overstepping their limits as dictated by the ‘separation of powers’, while simultaneously attempting to solve the public interest matter at hand.

Firstly, rather than sweeping directions being provided which transcend separation of powers and lead to non-compliance, the Court is now adopting a ‘delegated feasibility model’. This model ensures that rather than the Court imposing its directions on the States and Union Territories, governments are required to first assess their feasibility using strict criteria; and then modify the protocol and instructions to suit their realities. In Suhas Chakma, in order to ensure provision of open prisons, the Court instructed feasibility studies to take place with indicators of the minimum common standards. In Jaya Thakur, the extent of menstrual hygiene products that were to be provided were left to the periodic inspections undertaken by District Education Officers.

This model ensures better compliance and prevents governmental nonchalance, which was extremely common and prevalent erstwhile. This is done by providing tailored context-sensitive solutions incorporated after considering implementational flaws, acknowledging prevalent infrastructure, and adopting perspectives of diverse stakeholders. This also leads to the prevalence of a structured channel of dialogue between the Court and political actors.

Secondly, rather than the Court supervising the implementation of the directions through ‘continuing mandamus’ orders, there has been a shift towards a ‘delegated monitoring model’. ‘Continuing mandamus’ are orders which allow the Court to regularly monitor the enforcement of their directions. Herein, the Court continues to supervise the case until and unless substantial compliance has been reached. The delegated monitoring model on the other hand, entrusts the responsibility of monitoring the implementation of the judicial directions with not the Court but with quasi-judicial bodies such as committees and commissions or subordinate courts. Additionally, the Court empowers these bodies to prevent non-compliance. In Jaya Thakur, the Court mandated the National Commission for Protection of Child Rights and the State Commission for Protection of Child Rights to oversee the implementation of its directions (Para 176). In case of non-compliance, these bodies were instructed to take necessary steps as provided under the legislation. In Suhas Chakma, the Court formed a ‘Monitoring Committee’ and a ‘High-Powered Committee’ to ensure appropriate supervision. Moreover, High Courts were instructed to register a suo moto case to facilitate effective oversight.

This model allows the Court to delegate its monitoring work to stakeholders that are responsible first-hand to ensure compliance. Moreover, such delegation ensures that the cases and beneficiaries get the adequate attention they deserve at an appropriate enforcement level by a competent authority.

Thirdly, rather than emphasising on a complete satisfaction of the directions provided, the Court adopts a ‘substantial compliance model’. Herein, the government is required to not follow each and every direction, but largely comply with the overall spirit of the directions.

Fourthly, rather than subsuming the existing legislation by its directions, the Courts observes a ‘substantial accentuation model’ which means that its directions do not replace the legislation but merely accentuate it by filling the gaps. Thus, by not changing legislation under the garb of judicial observations, the Court is setting-up the PIL jurisdiction for success.

Ultimately, the effectiveness of these models and oversight mechanisms lie in their faithful implementation by the political executive. However, the Court has ensured that they substantially respond to the criticisms levied and accommodate ‘separation of powers’ concerns vis-a-vis PILs. This leads us to hope that PIL will not merely be a ‘romance gone wrong’ but a ‘romance worth fighting for’.

The Wholesale Ban Without Individualized Assessment

The Court disclaims intent to stifle critique, yet bans an entire chapter without identifying which passages warrant suppression. This violates the principles set out in the United Nations’ (UN) UN Human Rights Committee’s General Comment 34, the authoritative interpretation of Article 19 of the International Covenant on Civil and Political Rights (ICCPR), which protects freedom of expression. The Committee held that expression restrictions cannot include blanket publication bans unless authorities demonstrate that specific, non-severable content can be legitimately prohibited. The wholesale ban reveals the Court treats teaching students about judicial corruption as presumptively illegitimate.

The blacklisting without a hearing compounds this. The European Court of Human Rights (ECtHR), in Kyprianou v Cyprus, identified a structural problem also present in the Indian Supreme Court’s orders: when the judge who is the subject of criticism also adjudicates the contempt charge, fair-trial guarantees are violated. Here, the entire institution judges its own criticism.

Institutional Capture Through Circular Logic

The orders reveal progression from suppression to control. Having banned the chapter, the Court extended its reach to monitor who discusses orders and determine who may teach about the judiciary. The March 11 order directs the identification of social media accounts that acted irresponsibly, but provides no standard defining irresponsibility, no safeguards, and no temporal boundary.

The UN Office of the High Commissioner for Human Rights has documented that even the possibility of surveillance creates a chilling effect on Article 19 ICCPR expression rights. Citizens must now calculate whether discussing a Supreme Court order might expose them to contempt liability. This transforms critique from democratic practice into risk calculation.

But surveillance alone does not complete institutional capture. The curriculum mandate does. The Court requires any rewritten chapter to be approved by a committee, including a former judge collaborating with the NJA. The judiciary determines whether its criticism is legitimate, blacklists academics whose work fails this test, and then mandates that judges approve future curriculum.

The ECtHR in Mustafa Erdoğan v Turkey, applying Article 10 of the European Convention on Human Rights (protecting freedom of expression), held that academic freedom extends to the examination and criticism of public institutions, as democracies require independent assessment. India’s Supreme Court requires such examination to be vetted by the institution examined. When courts determine what may be taught about courts, they eliminate external scrutiny that legitimacy requires and invert accountability fundamental to democratic governance.

Why Nemo Judex Matters

When courts control boundaries of permissible criticism, they claim institutional privilege incompatible with democratic accountability. The global retreat from scandalizing contempt reflects this. The United Kingdom abolished the offense of scandalizing the court in 2013 through Section 33 of the Crime and Courts Act 2013, emphasizing that respect for the judiciary is undermined by criminal offenses that provide special protection from criticism. Canada’s Ontario Court of Appeal in R v Kopyto held that the same offense violated freedom of expression under Section 2 of the Canadian Charter of Rights and Freedoms, reasoning that courts earn public regard through merit, not through suppressing questioning.

This principle (that courts earn respect through merit, not by suppressing criticism) is embodied in international standards. The Commonwealth Latimer House Principles, which India endorses, state that contempt proceedings should not restrict legitimate criticism of judicial performance. The 2002 Joint Declaration by the UN, the Organization for Security and Cooperation in Europe (OSCE), and the Organization of American States (OAS) Special Rapporteurs on freedom of expression declares that special restrictions on commenting on courts cannot be justified. These instruments reflect a principle about what courts protect when invoking contempt power.

South Africa’s Constitutional Court, in S v Mamabolo, captured that principle: contempt law protects not judicial officers’ self-esteem or dignity, but the moral authority of the judicial process, which depends on informed public scrutiny as a democratic check. When courts police the boundaries of criticism, they undermine the legitimacy they claim to protect. India’s Supreme Court exemplifies this: requiring judicial pre-approval for curriculum on judicial corruption exercises institutional veto over accountability itself.

Proving the Thesis Through Suppression

The Court proved what it sought to suppress. Banning curriculum on judicial corruption, blacklisting academics without hearings, and mandating judicial approval to treat critique as a threat rather than accountability. If courts control what is taught, what prevents other institutions from claiming privilege? Where does self-protection end and accountability begin? When global consensus rejects contempt to shield courts, why India’s opposite trajectory? The institution that claims to protect independence demonstrates the insecurity that undermines it.

What does it mean to die with dignity? In India, this question has a constitutional history. The right to life under Article 21 of the Indian Constitution has, over time, been read to encompass the quality and character of that existence, and has been interpreted to include the right to die with dignity.

The journey began in Gian Kaur v. State of Punjab, wherein the Court, while rejecting any general right to die under Article 21, recognised that for patients who were terminally ill or in PVS, the right to die with dignity would be a part of the right to live with dignity. In Aruna Shanbaug v. Union of India, this right was given substance by recognising the legality of withholding or withdrawing life-sustaining treatment, when the decision to withdraw was in the patient’s best interests. However, the approval for the same had to be granted by a High Court. In Common Cause, a five-judge constitution bench grounded the right firmly in Article 21, displaced the courts from the centre of decision-making and laid down a framework for withholding or withdrawing treatment.

Under this framework, where there is no Advance Medical Directive, a treating physician, after determining that a patient is in an irreversible condition, may inform the hospital, which will constitute a primary medical board (PMB). The PMB, after discussing with the patient’s next of kin, gives an opinion on whether withdrawal of medical treatment is in the best interests of the patient. If it certifies withdrawal, a secondary medical board is constituted. Where both boards are unanimous, the hospital can proceed with withdrawal after informing the judicial magistrate. When there is a disagreement at any stage, the matter may be taken to the High Court, which must decide, keeping in mind the best interests of the patient. In Harish Rana, the Court, while applying this framework, also clarifies what a best interest assessment involves, resolves whether CANH constitutes medical treatment, and strengthens the institutional framework through which the right operates.

Best Interests Standards

While Common Cause established that any decision to withdraw treatment from incompetent patients was to be taken only when it was in their best interests, it did not expand on what this principle entailed. Harish Rana fills that gap by conducting a cross-jurisdictional analysis across seven legal systems and drawing from them a set of considerations that should be taken into account while determining the best interests of the patient.

The Court clarifies that a best interests analysis cannot merely include a clinical assessment of medical futility, which is necessary but not sufficient. The inquiry must also examine the patient’s own wishes and values, reconstructed through the evidence of those who knew them. Neither clinical judgment nor family preference alone is determinative; the constitutional obligation requires the assessment of the patient as an individual (para 233).

Resolving the CANH Classification 

Under Common Cause, only an intervention that qualifies as ‘medical treatment’ could be considered for withdrawal. Earlier, there was a lack of clarity on whether CANH fell within the definition of ‘medical treatment’ or whether it was basic primary care. The Court, in Harish Rana, clarifies this by recognising CANH as ‘medical treatment,’ thus making it subject to the procedure laid down in Common Cause. Without this classification, the withdrawal framework could not be triggered for a significant category of patients, including Harish Rana himself.

Strengthening the Framework

Beyond clarifying doctrine, the judgment also strengthens the infrastructure through which the right must operate. The Court directed states to ensure that a panel of registered medical practitioners is established without delay, who will be available to serve on medical boards. It further directed High Courts to issue directions to judicial magistrates to receive intimation from hospitals where medical boards are unanimous in their decision to withdraw or withhold treatment. These directions serve to ensure that the right can be realised not only in principle, but also in practice.

The Legislative Gap

While the decision in Harish Rana is a significant advance, the right to die with dignity still rests on judge-made guidelines with no statutory backing. The path to realising this right, as demonstrated by the case itself, has been spearheaded by the judiciary. This is not a sustainable model for accessing a right. The right to die with dignity cannot only be a right for those who can litigate it. A constitutional guarantee of this significance must be embedded in statute, with clear procedures and trained institutional actors.

The Court notes that, despite explicit judicial direction since 2018 and Law Commission recommendations dating back to 2006, the Parliament has still not acted. The Harish Rana case has illustrated what operationalising a right can look like; what remains is to see what accessing such a right will look like.

Domestic criminal law in countries like Bangladesh, India, and Pakistan permit this process to ensure trial of principal accused through the evidence of their accomplice, so that the trial is not hindered due to lack of evidence in complex and crucial cases. National criminal law usually sets conditions for such pardon including i) full, ii) true and iii) whole disclosure about the circumstances. If the approver does not fulfil the conditions, then the pardon will be revoked. The approver can be tried for the original offence for which they were pardoned, and their own confession can be used against them in that trial. Thus, the pardon initially offered to the approver does not provide blanket immunity.

Article 65 of the Rome Statute of the International Criminal Court (ICC) outlines a framework comparable to the provision of an approver. It governs proceedings based on an admission of guilt. Under this framework, the Chamber is not bound by the admission and must instead satisfy itself of several conditions. First, the accused must understand the nature and consequences of the admission. Second, the admission must be made voluntarily following sufficient consultation with defence counsel. Finally, the admission must also be supported by the facts of the case. This assessment is based on the charges, evidence presented by the prosecutor, and any other materials provided by the accused.

Nevertheless, it is clear in the Rome Statute that the accused remains accused and does not turn into a prosecution witness like in domestic law, and the Chamber may convict the accused even if the accused satisfies the above requirements. Hence, there is no scope for a predetermined or lenient sentence for an admission under Article 78 of the Rome Statute. This is because of the jus cogens character of international crimes. Consequently, pardon for such crimes should not merely be a domestic policy choice.

Additionally, the Guidelines for Agreements Regarding Admission of Guilt adopted by the ICC heavily emphasize the Court’s independent duty to establish the truth. Judges must examine not just the agreement between the parties, but also other evidence presented by the prosecutor and any other evidence, which can include victim representations and other sources, to ensure the facts are complete. Besides, the Guidelines explicitly recognize the need for victims to be informed of the proceedings and to be able to present their views and concerns, ensuring the process does not overlook their interests.

In contrast, the statutes and rules of the International Criminal Tribunals for the former Yugoslavia (ICTY) and Rwanda (ICTR) contained provisions for formal plea agreements. An accused could plead guilty to specific charges, often after negotiations with the prosecutor, who might agree to drop other counts or recommend a sentence. Although the Trial Chamber was not bound by the agreement. Judges had to verify that the plea was voluntary, informed, and unequivocal and that a sufficient factual basis for the crimes existed. In these cases, a convicted person received a sentence determined by the judges, with a guilty plea being a significant mitigating factor, often leading to a substantially reduced term.

Nevertheless, the primary goals of international criminal law and tribunals are not just to punish, but to establish a historical record of atrocities, promote reconciliation, deter future crimes, whereas a unilateral pardon could undermine these goals by appearing to offer impunity. Sentences are meant to reflect the individual’s guilt and the gravity of the crime, while a pardon would circumvent this carefully calibrated judicial process

In November 2025, the Pakistani government passed the 27^th Constitutional Amendment. Among other changes to Pakistan’s distribution of power, such as expanding presidential immunities and centralising military authority under the Chief of the Army Staff, the amendment created a new Federal Constitutional Court (FCC).

The FCC assumes the Supreme Court’s prior jurisdiction over constitutional interpretation, federal-provincial disputes, and the enforcement of fundamental rights. The consequence is a bifurcation of supreme judicial authority in Pakistan: the Supreme Court retains appellate jurisdiction over non-constitutional matters, while the FCC is the authoritative voice on constitutional adjudication. These latter matters are typically the most crucial in protecting democratic rights and norms against the state’s authoritarian intervention.

Justification of Specialised Constitutional Courts

Removing constitutional adjudication from the Supreme Court is argued to reduce that court’s significant case backlog, which stood at 56,169 cases in October 2025.

In addition to increased judicial efficiency, the Bill’s objects and reasons refer to improved consistency of constitutional adjudication by employing judges with “demonstrated expertise in constitutional law.” Constitutional decision-making is a specialised discipline involving complex, value-laden, systemic reasoning. Arguably, establishing a court with these specialised skills, rather than the more varied legal reasoning required in mixed courts, improves constitutional decisions.

Specialised constitutional courts also have a normatively important function in establishing constitutional respect. The argument is that a specialised court, in protecting its institutional interests, reinforces the salience of the constitution for both political actors and civil society. This constitutionalism-promoting function is especially important in countries where politicians have historically disregarded the constitution. This is true of Pakistan, where once-military and now increasingly authoritarian leadership has consistently ignored judicial pushback.

Institutional Capture

However, specialised constitutional courts are not necessary to protect democracy and human rights. Ordinary courts in successful constitutional democracies such as Australia, India and Canada exercise both constitutional jurisdiction and supreme appellate authority. Conversely, countries such as Turkey and Hungary maintain specialised constitutional courts that are criticised as ineffectual due to executive control. Judicial independence, rather than judicial specialisation, is the primary determinant of a court’s effectiveness in protecting democratic rights and norms.

Pakistan’s judicial history underscores this. Following the military leadership’s firm control over judicial elections and tenure from the 1950s, the judiciary took control of its own appointment process in the tenuous civilian rule of the 1990s. Consequently, under the military leadership of General Musharraf (1999–2008), the Supreme Court became progressively more assertive, challenging the regime’s violations of civil rights and the unconstitutional accrual of executive authority.

This independence has recently eroded. In 2024, the 26^th Constitutional Amendment reintroduced political control over judicial appointments by reconstituting the Judicial Commission such that only five of its thirteen members are now from the judiciary. Nonetheless, the Supreme Court in Sunni Ittehad Council v Election Commission of Pakistan (2024) challenged the consolidation of reserved seats among ruling parties. This indicates the court’s ongoing independence from, and sporadic resistance to, the regime.

The 27^th Amendment can plausibly be read as a response to such resistance. The first FCC judges are to be appointed by the President on the Prime Minister’s advice. No objective criteria or accountability procedure exists for this appointment. While subsequent appointments will be made by the Judicial Commission, as mentioned, this body is increasingly subject to executive capture. In reconstituting the bench of the country’s highest constitutional court, the regime reduces the likelihood of judicial intervention against its authoritarian interests.

Ultimately, the effectiveness of judicial institutions in protecting constitutional, democratic, and human rights norms depends primarily on their institutional independence. Eroding this independence in Pakistan appears to be the real reason behind the FCC’s creation. As these reforms form part of a broader global trend of judicial capture and associated democratic backsliding, the warning is clear: we should not allow institutional labels to mask democratic erosion.

The report has three main sections: first, how the Court discusses caste and varna; second, how it portrays the Scheduled Castes (Dalits); and third, what remedies it considers appropriate to address caste-based injustices. Put simply, the report highlights the contradictory nature of judicial discourse on caste over the years. While certain strands of judicial discourse downplay the pernicious nature of the caste system and reinforce negative stereotypes against Dalits, others challenge these narratives. These discursive choices also show that adjudication, particularly in caste-related matters, is not necessarily a neutral act, but an ‘ideological’ one shaped by judges’ political predilections.

For instance, consider the way in which judges discuss the Chaturvarna model of caste. Chaturvarna posits a fourfold division of society into Brahmins (priests), Kshatriyas (warriors), Vaishyas (traders), and Shudras (service castes). The fifth varna, “untouchables,” who are regarded as polluted and forced to perform menial manual labour, are part of this model by being placed outside it. This Chaturvarna model does not fully capture the intricacies of the caste system, as it reduces caste to a Hindu religious phenomenon and overlooks its political and economic dimensions. Nevertheless, it provides the caste system with the ‘ideological rationale’ of endogamy, hierarchy, and purity and pollution.

The report captures the significant differences in how the judgments discuss Chaturvarna. Some judgments recognise it for what it is. However, others portray it merely as a harmless division of society into distinct functional groups. In this view, the varnas were mutually interdependent and were later corrupted into the caste system that we see today. This view overlooks several key factors, including occupational, environmental, and technological influences, which better explain the origins of the caste system. It ignores how caste perpetuates itself through violence against women and Dalits. Moreover, it implicitly depicts the Chaturvarna—which the prominent civil rights leader Ambedkar denounced as ‘vicious’ and ‘harmful’—as an ideal worth recovering, by portraying it as a benign institution compared to the oppressive caste system.

Similarly, consider the Court’s description of Dalits as the beneficiaries of affirmative action. Even when the judgments recognise the necessity of reservation quotas for Dalits to create a level playing field, they use metaphors, such as “crutches” and “handicap,” to justify such a conclusion. The report shows how these metaphors that equate affirmative action beneficiaries with people with disabilities violate the dignity of both groups and overlook the historical and structural causes responsible for the pitiable condition of marginalised castes.

These examples illuminate the problematic judicial language surrounding caste. However, the report stops short of examining how this language influences case outcomes or government policies. Rather, it rests on the reasonable assumption that legal vocabulary significantly influences jurisprudence. In doing so, the report mainly aims to persuade the judiciary and policymakers to use language that affirms the constitutional mandate of equality and reflects the social realities of the caste system.

While significant, the report’s call for a change in language has its limitations. Put plainly, language change cannot, by itself, lead to structural change, just as speaking sensitively about the poor does not eradicate poverty. For instance, some judgments understand merit as stemming from accumulated caste privileges and therefore not undermined by reservations. However, this does not necessarily change the humiliation faced by the beneficiaries of reservation when they are mocked as quotawallas—a derogatory term implying that they are non-meritorious and would not have received the educational or employment opportunity without affirmative action.

Even though social reality matters, so does the language describing it. As caste is also a ‘construct’, its conceptual framing can influence the reality it describes. Further, how the state understands the social world significantly matters, as it has the capacity to impose such an understanding to transform that world. However, the report does not prescribe “correct” ways to discuss caste. It highlights judicial language aligned with constitutional values, without suggesting this is the only valid approach. A prescriptive exercise might be counter-productive due to the difficulties and controversies surrounding “correct” language. Thus, the report’s significance lies not in its potential of being a manifesto of “right” words but in its call to the state authorities to reflect on their language and its consequences.

Special measures

The Youth Justice and Criminal Evidence Act 1999 (YJCEA) measures (‘special measures’) which facilitate vulnerable and intimidated witnesses’ testimonies are considered ‘the most radical rewriting of the orthodox rules for treatment of witnesses in the adversarial system in the common law world’. YJCEA measures (Sections 23-29) include: witness screening vis-à-vis the accused, live link evidence, evidence in private, wig/gown removal, evidence-in-chief video-recording, cross-examination/re-examination video-recording, witness examination through an intermediary, and aid provision for communication. While this Act concerns juvenile offenders, these special measures protect witnesses regardless of their age. Moreover, under international human rights law and international & comparative criminal law standards, the YJCEA special measures are, mutatis mutandis, similar to those measures granted to protect vulnerable witnesses, especially when they have suffered violent crimes or sexual and gender violence.

Concerning the accused’s fair trial right, first, under common law (see, e.g., here, and here): the right to examine witnesses is not free-standing, but only a fair trial element; generally, there is no common law right to confrontation; and the right to witness examination does not involve physical confrontation. Second, while the European Convention on Human Rights (ECHR) (Article 6(3)) and European Court of Human Rights (ECtHR)-jurisprudence (e.g., Van Mechelen v. Netherlands) recognizes the right to witness examination, confrontation is not part of a fair trial: special measures to shield vulnerable witnesses from the accused do not contravene the ECHR if they are strictly necessary. Third, while common law requires evidence given in public, with press access, exceptions are allowed if strictly necessary. Special measures arguably fall into such exceptions. Additionally, ECHR (Article 6 (1)) and related jurisprudence are more permissive and broader. The House of Lords recognized common law exceptions to the orality principle and that special measures respect the ECHR. Fourth, in cases against the United Kingdom, the ECtHR found that witness safety/privacy may limit open and public criminal proceedings when this is strictly necessary and permissible under the ECHR. E.g., it determined that this concerns limitations to public proceedings to protect minors.

Sexual crimes

Regarding sexual crimes, under the YJCEA (Section 41), no evidence may be adduced, and questions cannot be asked by/on behalf of the accused about the complainant’s previous sexual behaviour with a third party/the accused, without judicial permission.

In R v. A, the House of Lords (majority) considered that although the YJCEA pursued desirable goals, it made an ‘excessive inroad’ into the fair trial right: according to Lord Steyn, ‘a prior relationship between a complainant and an accused may sometimes be relevant to what decision was made’. Scholars found YJCEA compatible with the ECHR by invoking the Human Rights Act, as the YJCEA should be read subject to an implicit provision: evidence or questioning needed for a fair trial should be admissible. Occasionally, complainant-accused relevant sexual experiences would be admissible; however, judges could exclude evidence of an isolated incident distant in time and circumstances.

Witness anonymity

An exceptional measure is witness anonymity vis-à-vis the accused and criminal procedure participants. Under the Coroners and Justice Act 2009 (CJA), Section 88(3), witness anonymity orders require: the need for protecting witness’s/another person’s safety or for avoiding serious property damage; consistency with the accused’s fair trial right; and that the witness’s testimony is so important for justice interests that their testimony is needed and the witness would not testify non-anonymously or that public interest would be harmed without anonymity.

Overall, the CJA arguably dilutes the accused’s fair trial right under the ECHR: grounds for allowing anonymity are expanded beyond threats to the safety of the witness or of another person. Moreover, it does not require that convictions cannot be based on anonymous testimony only/decisively. This is inconsistent with ECtHR-jurisprudence. E.g., under Kostovski v. The Netherlands, anonymous testimony should not be the sole/decisive basis for conviction. Al-Khawaja and Tahery v. United Kingdom concluded that ‘Even when “counterbalancing” procedures are found to compensate sufficiently the handicaps under which the defence labours, a conviction should not be based either solely or to a decisive extent on anonymous statements’. The above-mentioned CJA anonymity conditions seemingly introduce a lower threshold than the ECtHR’s ‘strictly necessary’ standard for measures restricting the accused’s fair trial right. By allowing serious property damage or real harm to the public interest as grounds for anonymity, the CJA diverges from ECtHR-jurisprudence: the latter requires life/integrity threats. Nevertheless, the other above-mentioned CJA-provisions on anonymous witnesses arguably meet ECtHR-jurisprudence, particularly those provisions concerning protections for the defendants, including that, when deciding on witness anonymity, the judge should consider: i) consistency ‘with the defendant receiving a fair trial’ (CJA-Section 88(4)); and ii) the defendant’s general right ‘to know the identity of a witness in the proceedings’ (CJA-Section 89(2)(a)).

The England/Wales Court of Appeals (Lord Chief Justice) in R. v. Mayers and Others soundly emphasized that the judge’s role involves a ‘delicate balance’ between the accused’s fair trial right (ECHR-Article 6) and witnesses’ rights (ECHR-Articles 2 (life), 3 (physical security) and 8 (private life)).

Conclusion

Witness safety/privacy protection drives the above-examined measures, adding a needed dimension that enforces human rights of witnesses, usually victims too, in English criminal law. Nevertheless, balancing this protection against the accused’s fair trial right is expected. Thus, the ECHR/ECtHR-jurisprudence are standards to follow.

The Legal Framework: A Contentious Push for Control

The legal lynchpin for the ban is the Directive on Regulating the Use of Social Media, 2080 (2023), the objective of which was to curb misinformation, hate speech, and other online harms, making social media platforms more accountable under Nepali law. This directive requires social media companies to register with the Department of Information and Broadcasting, establish a liaison office in Nepal, and appoint a grievance officer. Notably, this directive is a direct consequence of a Supreme Court order by Justices Dhungana and Thapa mandating the regulation of unregistered social media and OTT platforms.

However, enforcing a complete shutdown is where the legal argument falls apart. This ban is disproportionate and unconstitutional. It has a chilling effect on citizens’ fundamental rights, particularly the right to freedom of speech and expression enshrined under Article 17 of the Constitution of Nepal. Even though the right is not absolute, in Madhav Basnet v Ministry of Information and Communication (2002), the Supreme Court held that restrictions on freedom of expression must be reasonable and proportionate, and cannot extinguish the right altogether. This approach was reaffirmed in 2006 when provisions of the 1992 Broadcasting Act and the 1991 Publications Act were struck down for permitting arbitrary censorship. By simultaneously blocking 26 platforms, including popular ones such as Facebook, WhatsApp, Instagram, YouTube, LinkedIn, and Reddit, the Nepali government effectively crumbled the primary spaces for communication, business, association, and organization in the digital age. Hence, the ban cannot withstand either of the tests of reasonableness and proportionality entrenched in Nepal’s constitutional jurisprudence.

Nepal’s Breach of International Legal Commitments 

Moreover, the government’s actions place Nepal in direct breach of its international legal commitments as a party to the International Covenant on Civil and Political Rights (ICCPR), in particular Article 19. International human rights jurisprudence clarifies that any restriction on these rights must pursue a legitimate aim and, crucially, satisfy the test of “necessity” and “proportionality.” The UN Human Rights Committee (UNHRC) has consistently held that restrictions must pursue a legitimate aim and meet the strict tests of necessity and proportionality. Additionally, it has condemned internet shutdowns as being contrary to international human rights law. Furthermore, General Comment No. 34 (para. 34) emphasizes that restrictions must be the least intrusive means available. A ban that silences millions of users, regardless of context or conduct, does not meet this threshold, nor is it narrowly tailored. Instead, it transforms a regulatory measure into a censorship tool, collapsing the space for free speech, peaceful assembly, and democratic dissent.

Determining the Contours of Internet Access

Although the government did not cut internet connectivity outright, blocking major platforms had a functionally similar effect, depriving citizens of their primary spaces for speech, association, and civic participation. The right to internet access, increasingly recognised in international practice, goes beyond mere connectivity to encompass meaningful access to the platforms that make the internet a vehicle for rights. Constitutional guarantees under Article 17 of Nepal’s Constitution become hollow without them. As the UNHRC has affirmed, “the same rights that people have offline must also be protected online.”

Internet and social media crackdowns have taken place in the last five years throughout the Indian Subcontinent in India, Myanmar, Pakistan, Bangladesh, and Sri Lanka. Although there have been resolutions highlighting the importance of such access, there remains a lack of further analysis on what constitutes “substantive” internet access. The UN does not provide a comprehensive definition of the right to internet access. This ambiguity in the definition of rights of access has created a legal vacuum where it is convenient to curb the freedoms on the internet rather than curtail the Right to the internet. As per the UN Special Rapporteur Frank La Rue, “access to the internet is not a right in itself but a means to exercise a range of human rights.”

The resuscitated social media, after the ban was lifted, elected Nepal’s first female Prime Minister through the popular gaming app ‘Discord’, which reflects the force of social media as a modern public square.