Discuss the doctrine of "reasonable apprehension of bias" and how it applies to judicial recusal in India. Do we need a codified recusal framework?

The maxim nemo judex in causa sua (no one should judge their own cause) underpins natural justice. Its modern test, reasonable apprehension of bias, asks whether a fair-minded observer would reasonably suspect bias, not whether the judge was biased. A curative petition seeks recall of the September 2026 split verdict on the CEC appointment law, alleging that a Bench judge's son is ECI standing counsel [1]. It shows why the doctrine's uncodified enforcement matters.

The Doctrine

• Appearance, not proof: justice must be seen to be done, so actual bias need not be proved.
• Forms: pecuniary, personal (relatives, friends) and official or subject-matter bias. A direct pecuniary interest disqualifies automatically.
• Indian test: Ranjit Thakur v. Union of India (1987) asked whether the party could reasonably apprehend that bias might operate against him.

Application to Judicial Recusal in India

• Self-regulated: recusal rests on the individual judge's conscience. No statute or court rule governs it.
• Soft law: the Restatement of Values of Judicial Life (1997) bars a judge from hearing matters involving family or close relations [2]. The Bangalore Principles (2002) say the same.
• No reasons recorded: in the NJAC case (2015), Justice Kurian Joseph urged judges to give reasons for recusal. Practice remains inconsistent.
• Disclosure gap: the petitioner learnt of the link from the media only after the verdict [1]. Even perceived partiality is costly when the case concerns the ECI, which the Supreme Court places on a "higher pedestal" [3].

Do We Need a Codified Framework? Yes, because it would bring:

• Mandatory disclosure at the first hearing of any relative appearing for a party, so objections come before the verdict, not after.
• Clear criteria, with contested pleas decided by the CJI or another bench rather than the judge concerned.
• Recorded reasons that build consistent precedent.

But with safeguards against:

• Bench-hunting: litigants may manufacture conflicts after adverse rulings.
• Threats to independence: a code drafted by the executive could chill judges, and small benches could be paralysed.

A calibrated middle path is best. The Supreme Court's full court, not Parliament, should turn the 1997 Restatement into a binding disclosure-and-recusal code, with time-bound objections and costs for frivolous pleas. This protects judges and litigants alike and honours the fair-procedure guarantee of Articles 14 and 21. It ensures justice is not only done but seen to be done.