<p>The foundational maxim that justice must not only be done but also be seen to be done is no mere rhetorical flourish. It is the vital artery of judicial legitimacy. It is a structural restraint on judicial power, one that recognises a truth: courts derive legitimacy not only from the correctness of their decisions but from the credibility of the process by which those decisions are reached.</p>.<p>For decades, this principle served as a minimum standard of discipline. Even amid political upheaval, the judiciary fought to preserve the appearance of fairness, understanding that this “appearance” was not superficial but constitutional.</p>.<p>Today, that promise is being quietly recalibrated. As the judiciary grapples with high-stakes constitutional questions, a disturbing shift emerges. The recusal plea by Arvind Kejriwal, arising out of proceedings in the Delhi excise policy matter and argued by him in person, compels a more difficult inquiry. The issue is not whether bias exists; it is whether the circumstances give rise to a reasonable apprehension of bias, and whether they sustain a litigant’s confidence in receiving a fair hearing. The law understands that once a litigant loses faith in the umpire, the game is already lost.</p>.<p>Justice Swarana Kanta Sharma addressed each of the grounds urged by Kejriwal through a structured rebuttal in her 115-page order. This episode marks a distinguishable departure from the established approach to judicial recusal in India.</p>.<p>Legal history offers a stark contrast. The origins of the concept are found in R v Sussex Justices, ex parte McCarthy, decided in England in 1924. There, a court quashed a conviction not because the decision was wrong, but because a clerk with a potential conflict had been present during deliberations. The judges accepted that he had not influenced the outcome. It did not matter. The mere possibility that justice might not appear to have been done was enough to invalidate the proceedings.</p>.<p>Indian jurisprudence was forged in this same fire. In Manak Lal v Dr Prem Chand Singhvi, the Supreme Court held that the likelihood of bias, judged by a reasonable person, is enough to vitiate any proceeding. This was echoed in the Kraipak (1969) and Ranjit Thakur (1987) rulings, which established that the judiciary does not demand proof of a “crooked heart.” It demands a process that is beyond reproach. The test is not what the judge feels, but what the public fears.</p>.<p>When we view the Kejriwal order through this lens, the institutional danger becomes clear. When recusal is resisted with such categorical intensity, the message to the public is that the judge’s internal sense of integrity outweighs the public’s external need for confidence. It reframes recusal as a matter of personal pride rather than a matter of public office. This is how the baseline of justice is redefined: not by rewriting the law, but by slowly raising the bar for what counts as a “reasonable” fear until only the most blatant corruption is recognised.</p>.<p>Justice Sharma has indicated that she may appoint three senior advocates as amici curiae following Kejriwal’s decision to boycott the proceedings after the refusal of recusal.</p>.<p>Reframing bias</p>.<p>The concern is not confined to a single case. Questions about roster allocation, the concentration of administrative power within the court, recurring debates over judicial appointments, and the increasing strategic attention paid by the Bar to forum and bench – all point to a wider unease. These are not isolated anxieties. They shift how institutional fairness is understood, debated, and, at times, doubted.</p>.<p>Justice Sharma observes that “recusal has to stem from law, and not from narrative.” Ironically, this ignores the fact that every judicial order is itself a narrative.</p>.<p>While Kejriwal has responded by invoking a form of satyagraha – an appeal, in effect, beyond the courtroom, the proceedings will continue. So, inevitably, will the theatre of justice. The stage remains, but the performance of justice as we know it may not.</p>.<p>The court has effectively ruled that while justice must be seen to be done, the public is currently denied standing to watch it. I submit that we have traded stare decisis for ‘don’t stare at us,’ where a reasonable apprehension of bias is treated like a frivolous <br>motion. The ‘appearance of fairness’ hasn’t been overruled; it has just been summarily dismissed for failing to fit the court’s new narrative. How, then, is the public expected to understand such reasoning? Res ipsa loquitur – the facts speak for themselves, but the court is clearly no longer on speaking terms with the facts.</p>.<p><em>(The writer is a lawyer and former advisor to the AAP government)</em></p>
<p>The foundational maxim that justice must not only be done but also be seen to be done is no mere rhetorical flourish. It is the vital artery of judicial legitimacy. It is a structural restraint on judicial power, one that recognises a truth: courts derive legitimacy not only from the correctness of their decisions but from the credibility of the process by which those decisions are reached.</p>.<p>For decades, this principle served as a minimum standard of discipline. Even amid political upheaval, the judiciary fought to preserve the appearance of fairness, understanding that this “appearance” was not superficial but constitutional.</p>.<p>Today, that promise is being quietly recalibrated. As the judiciary grapples with high-stakes constitutional questions, a disturbing shift emerges. The recusal plea by Arvind Kejriwal, arising out of proceedings in the Delhi excise policy matter and argued by him in person, compels a more difficult inquiry. The issue is not whether bias exists; it is whether the circumstances give rise to a reasonable apprehension of bias, and whether they sustain a litigant’s confidence in receiving a fair hearing. The law understands that once a litigant loses faith in the umpire, the game is already lost.</p>.<p>Justice Swarana Kanta Sharma addressed each of the grounds urged by Kejriwal through a structured rebuttal in her 115-page order. This episode marks a distinguishable departure from the established approach to judicial recusal in India.</p>.<p>Legal history offers a stark contrast. The origins of the concept are found in R v Sussex Justices, ex parte McCarthy, decided in England in 1924. There, a court quashed a conviction not because the decision was wrong, but because a clerk with a potential conflict had been present during deliberations. The judges accepted that he had not influenced the outcome. It did not matter. The mere possibility that justice might not appear to have been done was enough to invalidate the proceedings.</p>.<p>Indian jurisprudence was forged in this same fire. In Manak Lal v Dr Prem Chand Singhvi, the Supreme Court held that the likelihood of bias, judged by a reasonable person, is enough to vitiate any proceeding. This was echoed in the Kraipak (1969) and Ranjit Thakur (1987) rulings, which established that the judiciary does not demand proof of a “crooked heart.” It demands a process that is beyond reproach. The test is not what the judge feels, but what the public fears.</p>.<p>When we view the Kejriwal order through this lens, the institutional danger becomes clear. When recusal is resisted with such categorical intensity, the message to the public is that the judge’s internal sense of integrity outweighs the public’s external need for confidence. It reframes recusal as a matter of personal pride rather than a matter of public office. This is how the baseline of justice is redefined: not by rewriting the law, but by slowly raising the bar for what counts as a “reasonable” fear until only the most blatant corruption is recognised.</p>.<p>Justice Sharma has indicated that she may appoint three senior advocates as amici curiae following Kejriwal’s decision to boycott the proceedings after the refusal of recusal.</p>.<p>Reframing bias</p>.<p>The concern is not confined to a single case. Questions about roster allocation, the concentration of administrative power within the court, recurring debates over judicial appointments, and the increasing strategic attention paid by the Bar to forum and bench – all point to a wider unease. These are not isolated anxieties. They shift how institutional fairness is understood, debated, and, at times, doubted.</p>.<p>Justice Sharma observes that “recusal has to stem from law, and not from narrative.” Ironically, this ignores the fact that every judicial order is itself a narrative.</p>.<p>While Kejriwal has responded by invoking a form of satyagraha – an appeal, in effect, beyond the courtroom, the proceedings will continue. So, inevitably, will the theatre of justice. The stage remains, but the performance of justice as we know it may not.</p>.<p>The court has effectively ruled that while justice must be seen to be done, the public is currently denied standing to watch it. I submit that we have traded stare decisis for ‘don’t stare at us,’ where a reasonable apprehension of bias is treated like a frivolous <br>motion. The ‘appearance of fairness’ hasn’t been overruled; it has just been summarily dismissed for failing to fit the court’s new narrative. How, then, is the public expected to understand such reasoning? Res ipsa loquitur – the facts speak for themselves, but the court is clearly no longer on speaking terms with the facts.</p>.<p><em>(The writer is a lawyer and former advisor to the AAP government)</em></p>
