A Rule & Its Quiet Reversal
Apeksha Kachhawaha* & Kshitij Saruparia**
On 11 June 2026, the Supreme Court of India decided that a tree branch falling on a stationary autorickshaw is not, legally speaking, an accident caused by the use of a motor vehicle. A municipal corporation, an insurer, and a state horticulture department had spent over a decade litigating exactly this point. In paragraph 13, the Court built a clear rule: where a vehicle plays no role beyond providing the location of an injury, no claim lies under India’s principal motor accident compensation statute. Two paragraphs later, the Court ignored that rule’s consequences entirely and ordered the same parties, under the same statutory apportionment, to pay an enhanced sum to the same claimant. It did this using Article 142 of the Constitution, the Supreme Court’s power to do “complete justice,” without citing, distinguishing, or even mentioning the one precedent that defines the limits of that power.
The case is The Commissioner, Bruhat Bangalore Mahanagara Palike v. K.K. Umesh Kumar, and the claimant’s injuries explain why it took fifteen years and four rounds of litigation to reach paragraph 15. A branch fell on the autorickshaw sheltering him from rain near a Bangalore stadium in 2007, leaving him permanently paraplegic and doubly incontinent. He sought compensation under Section 166 of the Motor Vehicles Act, 1988, which permits claims for injuries “arising out of the use of motor vehicles.” A tribunal dismissed the claim as a natural calamity. The High Court dismissed it again on delay. The Supreme Court remanded. A second High Court bench finally awarded compensation, apportioning liability across the corporation, the insurer, and the horticulture department. The corporation's appeal against that apportionment put before the Supreme Court the question every earlier court had avoided: whether this was an accident arising from vehicle use at all.
The Court’s answer is doctrinally careful. Citing its own precedent in Shivaji Dayanu Patil v. Vatsala Uttam More, which had given the phrase “arising out of the use of a motor vehicle” a wide reading covering even stationary, broken-down vehicles, the Court nonetheless held that the wide reading has a limit. A vehicle that merely provides the location of an otherwise unrelated event, it reasoned, is not part of the proximate cause of the injury. To illustrate the point, the Court offered a simple hypothetical: a pedestrian sheltering under the same tree would have suffered the identical injury. In that situation, the presence of the autorickshaw made no causal difference. That yields a real legal rule, applicable to future cases with similar facts, and narrows liability under Section 166 in a predictable and citable way.
Then the Court’s reasoning changes the register entirely. “A question however refuses to leave us,” the judgment states, before pivoting from statutory interpretation to conscience: a person left paraplegic and incontinent “does not appeal to the conscience of justice” if denied compensation. The Court found the High Court's quantum “insufficient” for taking “a somewhat technical approach.” It enhanced compensation from ₹17.10 lakh to ₹25 lakh, left the underlying apportionment among the corporation, the insurer, and the horticulture department undisturbed, and attributed the whole exercise to Article 142.
This is not, on its own, an unusual move. Indian courts have leaned on Article 142 to award compensation beyond a strict statutory reading since the Bhopal gas litigation, and the doctrine survives because hard cases generate sympathy that hard rules cannot always answer. The informed objection is that this is one more routine instance of that practice. The Court confirmed in Shilpa Sailesh v. Varun Sreenivasan that Article 142 may depart from substantive law where “fundamental general and specific public policy” justifies it, and elsewhere has refused such relief outright when facts do not warrant it. On this view, Umesh Kumar is unremarkable: a sympathetic claimant, a humane outcome, nothing to detain a constitutional lawyer.
The objection holds for every Article 142 compensation case except this one. Shilpa Sailesh and the broader public policy line end litigation on terms that do not increase a losing party’s exposure beyond what the merits already supported. The Court’s own controlling authority on Article 142, Supreme Court Bar Association v. Union of India, reaffirmed at the paragraph level by a Constitution Bench as recently as 2024, defines the boundary precisely: the power exists to “balance the equities between the conflicting claims of the litigating parties” by “ironing out the creases” in a dispute already before it, not to enlarge what a party must pay once the statutory basis for that payment has been narrowed. Umesh Kumar does the opposite. Paragraph 13 narrows the statutory basis for the respondents’ liability. Paragraph 15 enhances what those same parties must pay, on an apportionment scheme whose statutory foundation the Court had weakened two paragraphs earlier, without asking whether that combination fits a power said to exist only for settling disputes between litigants already before it.
This gap matters beyond Umesh Kumar’s own facts because the Court had a citable test directly on point and never applied it. Commentators have warned for years that unreasoned Article 142 departures from statute breed uncertainty, since neither courts nor litigants can predict when “complete justice” will override a rule just laid down. Umesh Kumar illustrates that failure mode in a context where the Court could have applied its own test and did not.
The closest comparison shows what a more disciplined use of that power looks like, not a claim the two disputes are otherwise alike. In State of Tamil Nadu v. Governor of Tamil Nadu, the Court invoked Article 142 to deem gubernatorial assent to ten bills that had been withheld for years. The bench built an explicit justification: unconstitutional executive delay, no competing statutory text barring the remedy, and a genuine constitutional vacuum needing to be filled. That reasoning was subsequently tested. A Constitution Bench in In re: Assent, Withholding or Reservation of Bills held that “deemed assent” has no constitutional basis and that Article 142 cannot supplant the constitutional role of the Governor or President. Even a reasoned invocation of the power did not survive closer scrutiny. Umesh Kumar offers no reasoning at all. The Court does not address Supreme Court Bar Association or explain why a power held to exist only for settling disputes between parties before it can be used to enlarge a respondent's liability after narrowing its statutory basis. The next bench facing a less sympathetic claimant has nothing here to work from.
None of this means the claimant should have received less. The criticism is not that a man with permanent paraplegia and double incontinence received an additional ₹7.90 lakh. It is that the Court reached that result through a power bound to stated limits, and never tested its decision against those limits, not that it necessarily breached them. Paragraph 13’s rule survives this judgment and remains available to a future bench. What does not survive is any indication of how that bench should treat the next request to enhance liability after narrowing the basis for it. A court that builds a careful rule and then grants equitable relief running the opposite direction, in the same judgment, without asking whether the two fit together, has not shown that complete justice and its own precedent agree. It has shown that the question never came up.
*Apeksha Kachhawaha is a graduate of Maharashtra National Law University, Nagpur.
**Kshitij Saruparia is a graduate of NALSAR University of Law, Hyderabad.
The views expressed above are the author's alone and do not represent the beliefs of Pith & Substance: The CCAL Blog.

.png)



Comments