The Supreme Court-appointed committee on gender-sensitive judgment writing places an extraordinary expression in its glossary of problematic judicial language: “the body of women as playground”. It recommends replacing this with a statement that the complainant’s bodily autonomy was violated.The substitution does more than improve tone. It changes the legal subject of the sentence. The first turns sexual violence into a florid metaphor about women collectively. The second identifies the right infringed and directs attention to the alleged act.But the difference also exposes the limits of vocabulary reform. A judge may stop calling a complainant a “helpless woman” or someone who “lost her chastity”. The same judge may still distrust her for reporting late, having a prior relationship, lacking visible injuries or behaving calmly afterwards. The central problem is therefore a language-justice gap. Can courts abandon patriarchal words without discarding the assumptions that produced them?The question is no longer academic. The report, titled Judgments and Gender: Sensitivity and Compassion in Writing Judgments, now carries the force of a judicial direction. On July 14, a bench of Chief Justice Surya Kant and Justices Joymalya Bagchi and V. Mohana approved it and directed all courts to strictly follow its expressions. The order requires publication on the websites of the Supreme Court, the high courts and district courts. State police chiefs and directors of prosecution must instruct police stations to apply it while recording FIRs and filing chargesheets. The full text became public on August 3.The committee was constituted under the court’s judgment of February 10, 2026. Chaired by former Supreme Court judge Justice Aniruddha Bose, director of the National Judicial Academy, it included former Gujarat high court Chief Justice Sonia Gokani, former Madhya Pradesh DGP Anuradha Shankar, Supreme Court advocate Surat Singh and anthropologist Lucy T.V. Zehol. It analysed 125 trial court judgments from across India, consulted State Judicial Academies and gathered stakeholder responses through District Legal Services Authorities.When language becomes a findingThe case that produced the report shows why judicial language cannot be separated from judicial reasoning. According to the allegations recorded by the Allahabad high court, a 14-year-old girl was taken on a motorcycle on the assurance of being dropped home. The accused allegedly grabbed her breasts, dragged her towards a culvert and broke the string of her lower garment. They fled when two passersby arrived on hearing her cries.The high court held on March 17, 2025 that these acts did not cross from preparation into an attempt to rape. It reduced the offences for which the accused had been summoned. The Supreme Court stayed those observations on March 26, 2025, describing them as reflecting a totally “insensitive and inhuman approach”. On February 10, it set the judgment aside as contrary to settled criminal jurisprudence and restored the original summons.Also read: Housekeeping Secret: How the Supreme Court Turned Down Its Own VolumeThis was more than a dispute over offensive phrasing. The high court’s description of what happened shaped its legal classification of the conduct. Language organised the facts, minimised some acts and fixed the point at which the law was thought to intervene.The committee addresses this connection in its strongest recommendation. Judicial reasoning must rest on evidence, not assumptions or moral perceptions about the complainant. It cautions against equating absence of injuries, delayed reporting or lack of resistance with consent. It asks courts to replace the language of “modesty”, “honour” and “chastity” with dignity, bodily autonomy and personal integrity. The term that gives this reform its shorthand is “prosecutrix”. A relic of colonial procedural vocabulary, it defines a woman by the prosecution rather than by what she endured. The report replaces it with “victim”, “survivor” or “complainant”, and the July 14 order makes that vocabulary binding on every court. Sensitive judicial language, the report says, is “not merely a matter of politeness”.Such changes matter because “modesty” locates the injury in a woman’s supposed sexual virtue rather than in the violation of her choice. Yet a modern vocabulary can coexist with the “ideal victim” stereotype. That stereotype imagines a complainant who resists, cries out, reports immediately and never wavers in her account. Any departure is then read as fabrication rather than as a possible consequence of trauma, fear or social pressure. Mrinal Satish’s empirical work on rape sentencing shows that rape myths survive their removal from statutory language. They migrate into judicial discretion and shape what courts treat as relevant at conviction and sentencing. The glossary is therefore valuable only as a guide to reasoning, not as a judicial thesaurus.The courtroom produces its own languageThe report moves beyond judgment writing when it examines compassionate court practices. It asks judges to control humiliating cross-examination, ensure legal aid from the FIR stage, hold in-camera trials and provide screens or video links for vulnerable witnesses. Confidentiality of victims’ identities, mandated in Nipun Saxena v. Union of India (2018), remains absolute.This is the report’s most consequential section. A survivor experiences the law through the police station, hospital, prosecutor’s office and waiting area long before she reads a judgment. Pratiksha Baxi’s ethnographic study of rape trials shows how testimony can be disqualified at each of these sites, even where written law formally protects it.The report’s own survey annexure records repeated adjournments, poor communication, pressure from police or opposing counsel and the absence of separate waiting areas. Most strikingly, 80.7% of respondents said they were unaware of the Witness Protection Scheme. A right unknown to its intended beneficiary exists mainly in institutional paperwork.Gender sensitivity also requires evidentiary discipline. It does not ask courts to presume every allegation true or to curtail legitimate cross-examination. It asks judges to separate questions that test reliability from questions that merely reproduce social myths. Fairness to the accused and dignity for the complainant are compatible when both rest on relevance and reasoned findings.What the July 14 order fixed, and what it did notThe report presents itself partly as an empirical exercise. But it does not identify the 125 judgments, explain their selection, specify the period covered or give a state-wise distribution. The survey annexure supplies percentages without disclosing respondent numbers or profiles. These omissions do not invalidate the experiences recorded. They limit the conclusions that can safely be drawn. They also prevent researchers from testing whether training changes judgment writing over time.The report must also be read alongside the Supreme Court’s 2023 Handbook on Combating Gender Stereotypes. That handbook already supplied alternative terminology and warned that stereotypes may distort the application of law even where the outcome appears correct. The new report’s distinct contribution lies in binding judgment writing to courtroom management and to the survivor’s experience of process.The July 14 order answers part of the implementation question. Compliance machinery now exists: mandatory publication, police circulars, circulation to judicial academies and law universities. What the order does not create is an audit. Nothing requires appellate courts to expressly flag stereotypical reasoning when correcting it. No mechanism documents recurring errors or measures compliance in subordinate courts. At the July 14 hearing itself, Justice Mohana asked whether the court’s February judgment had even been cited below. The record justifies the doubt. Aparna Bhat v. State of Madhya Pradesh issued binding directions on judicial language in 2021. The Allahabad order came four years later.The court had also asked the committee to compile offensive expressions from different Indian languages and consider translated public guidelines. The report records some Hindi remarks encountered by survivors but offers no systematic multilingual compilation. More troublingly, one passage groups hijra, thirunangai (transgender women), thirunambi (transgender men) and jogappa (a traditional transgender community) together, calls them implicitly pejorative and proposes “intersex” as the acceptable term. That conflates distinct concepts. NALSA v. Union of India (2014) recognised that Indian gender-diverse communities carry specific social and cultural identities. “Intersex” concerns variations in sex characteristics and cannot substitute for them. A report devoted to precise terminology should correct this before it becomes training material.From sensitisation to accountabilityThe committee is right that words carry institutional power. A judgment is a statement issued in the name of the state. Its descriptions guide subordinate courts, enter legal databases and shape future ideas about credible behaviour. But lasting reform requires three levels of work. Lexical reform removes degrading expressions. Reasoning reform stops stereotypes from deciding consent, credibility and sentence. Institutional reform changes how evidence is recorded, witnesses are protected and judicial performance is evaluated.Compassion here should not mean paternal sympathy for the “helpless female” the report rightly denotes it. It should mean a method that takes trauma seriously without replacing evidence and protects dignity without prejudging guilt. The report has begun changing how courts speak. Its success depends on whether the judiciary will change how courts listen, reason and account for their failures.V. Venkatesan is a veteran legal journalist.