To strengthen the Supreme Court, divide it

The Hindu, November 14, 2026  

By Pradeep Mehta & Vijay Kelkar

In 2016, three judges of the Supreme Court of India paused to ask a question about their own institution: had the apex court, established by the Constitution’s framers to safeguard the Republic’s foundational document, lost its way, drowning in ordinary appeals while its constitutional work withered? They framed 11 questions and referred it to a larger Bench. A decade later, the reference remains unheard. No emblem of the problem is sharper, nor any case for reform stronger: split the Supreme Court in two.

As 2025 drew to a close, the court carried 92,318 pending cases, its heaviest backlog ever and nearly 10,000 more than a year earlier. The caseload is growing at a pace that no reasonable enlargement of the Bench can outrun.

The remedy is simple and not new: a lean Constitutional Court of about 11 judges in Delhi, hearing only constitutional cases and substantial questions of law, with four or five regional Courts of Appeal to handle the appeals that now flood the apex court. The idea is not to shrink the judiciary’s reach but to sort its work, so that questions shaping the Republic are not decided in the same queue as those shaping a single case.

Consider how far the court has drifted. In the first half of the 1960s, its Constitution Benches decided an average of 134 cases a year; over the last decade, that has collapsed to fewer than 10. The docket, instead, has been swallowed by Special Leave Petitions, discretionary appeals under Article 136 that were intended to serve as a rare safety valve. By one count, they accounted for 84.6% of the court’s admission docket in 2011.

A tribunal designed to be the final word on the Constitution has turned itself into a general court of fourth appeal. It has admitted as much: in the Bihar Legal Support Society (1986), a Constitution Bench under Chief Justice P.N. Bhagwati said it was never meant to be an ordinary court of appeal.

Beyond the backlog

The case for splitting the court is not merely that it is overcrowded. A two-tier structure would allow specialisation to achieve what an increase in judicial capacity alone cannot. A dedicated Constitutional Court, hearing a curated stream of cases, could sit in full Benches and deliver consistent, predictable rulings that settle questions once.

Regional courts of appeal could specialise in the ordinary caseload and clear it far faster than an apex court. They would also bring the final tier of ordinary justice closer to the citizens who need it. Every large legal system has converged on this pyramid-shaped structure: a final court works best when it decides a few cases carefully, while the tiers below it handle many more quickly.

The 229th Law Commission Report (2009) laid out a full blueprint for a Constitution Bench in Delhi with four regional Cassation Benches. In 2023, a Parliamentary Standing Committee recommended regional Benches, which the Union Law Ministry has since accepted. The model is well tested in the very democracies India is compared to.

The U.S. Supreme Court sits atop 13 federal Courts of Appeals, deciding only a sliver of cases. South Africa runs a two-court system: a Constitutional Court in Johannesburg, and a separate Supreme Court of Appeal in Bloemfontein for everything else. Germany’s Federal Constitutional Court in Karlsruhe hears constitutional matters alone, while specialised federal courts handle other areas of law.

And this is not only a matter of systems built that way from the start: countries have made the very change India now contemplates. Brazil is the closest recent parallel. Its Supreme Federal Tribunal once carried the same double burden as ours, guarding the Constitution while drowning in ordinary federal appeals. The 1988 Constitution created a separate Superior Tribunal of Justice to take over that appellate load.

Brazil, however, also offers a warning: the split alone did not ease the backlog, which declined only after the apex court was given the power to filter out cases it did not need to hear. The tool is familiar from the very courts India admires. The U.S. Supreme Court grants review to roughly one percent of the 7,000 petitions it receives each year, a discretion held since 1925; Brazil’s apex court, after a 2004 reform, hears a constitutional appeal only where the question carries a “general repercussion” beyond the parties.

Without such a filter, a Constitutional Court would inherit, in miniature, the very congestion it was built to escape. And the route need not be treacherous: the 229th Report argued Article 130 may permit regional Benches without a constitutional amendment, meaning the first steps could be taken by the court itself, without waiting for Parliamentary approval.

The strongest objection must be stated at full strength, because it is why this reform has stalled for so long, and because it is held by those who know the institution best.

The concern, voiced by successive Chief Justices, is that a divided Supreme Court may no longer be a single supreme one: separate the constitutional work from the appellate work, and the regional courts could come to be seen as lesser venues, with litigants pressing to reach the apex court in Delhi, until the authority that flows from having one final court is diffused. There is real force in this. An apex court’s power lies partly in its indivisibility: it is the last word, with no higher forum to appeal to and no rival court to be played against it.

Restoring the court

But the concern mistakes the disease for the cure. A single court is only a single voice when it can speak with one, and ours cannot: it already sits in a dozen small Benches that often contradict each other, its constitutional work reduced to a handful of Constitution Benches a year.

The indivisibility being defended is a fiction; the court is already divided, not by design but by sheer volume, into scattered panels whose rulings no one can fully reconcile. Separation would rescue the constitutional core of one that has fractured on its own.

When the Bench was expanded to 34 in 2019, pendency stood near 60,000; today it is close to 93,000. In 2025, the court disposed of 65,615 cases and still fell behind, as 75,410 fresh matters arrived. In 2026, Parliament raised the sanctioned strength again, to 38. Adding judges without sorting the work risks enlarging the same overburdened court. The task is not to move the same work faster, but to sort it.

The Supreme Court that gave us the basic structure doctrine, and once stood as the boldest constitutional court in the democratic world, now rarely sits in constitutional judgment at all. No outside hand has diminished it; it has been buried under its own docket.

To leave it as it is, out of reverence for a unity that no longer exists, is to mistake sentiment for strength. Splitting the court would let it return to its pristine structure, while a new tier of regional appeals courts would finally give ordinary litigants the swift justice they are owed. That would not be its diminishment, but its restoration.

Pradeep S. Mehta is the Secretary-General of CUTS International (Consumer Unity and Trust Society); Vijay L. Kelkar is the Vice President of the Pune International Centre, with contributions from Anushka Kewlani of CUTS.

This article can also be viewed at:
https://www.thehindu.com/