Delhi High Court Flags Confusion on IP Jurisdiction, Refers Issue to Larger Bench (HUL v. Kwick Living)

In HUL v. Kwick Living, the Delhi High Court declined to decide territorial jurisdiction and referred three questions to a Larger Bench. The underlying dispute is a familiar disparagement fight — HUL's grievance against a campaign styled "War on What's Hidden" that it says denigrates Vim and Surf Excel. Confronted at the threshold with the defendant's objection that Delhi has no territorial jurisdiction, Justice A.J. Bhambhani surveys the wreckage of the case law, and refers three questions to a Larger Bench.

The factual matrix is almost engineered to expose the doctrinal fault lines:

• Both parties have their registered offices in Mumbai.
• HUL additionally has a corporate office in Delhi — a subordinate office, in the language of the statute.
• The only impugned hoarding the plaintiff could actually confirm at the time of filing was in Mumbai (para 75 of the plaint).
• The sole hook for Delhi is digital: the campaign is accessible on YouTube, Instagram, and the defendant's website, which is said to offer the defendant's competing products for sale within Delhi.

The textbook problem: A corporate plaintiff, headquartered where a part of the cause of action demonstrably arose, wants to sue at a distant forum where it happens to keep a branch office and where — it says — the internet reaches. That is precisely the fact pattern the Supreme Court had in mind when it decided Indian Performing Rights Society Ltd. v. Sanjay Dalia in 2015.

Three statutory tracks, one recurring collision

The court correctly frames the question as an interplay of three provisions:

1. Section 20 CPC — the default rule: sue where the defendant resides or carries on business, or where the cause of action wholly or in part arises (s.20(c)).
2. Section 134(2), Trade Marks Act, 1999 and Section 62(2), Copyright Act, 1957 — the "long-arm" provisions that additionally let a plaintiff sue where it resides or carries on business, notwithstanding the CPC.
3. The relationship between the two.

The friction is structural. Sections 134(2) and 62(2) were enacted to help rights-holders by adding a plaintiff-friendly forum. Sanjay Dalia then read a restriction back into that generosity: if the plaintiff's principal office is also a place where the cause of action arose, it must sue there and cannot drag the defendant to a far-flung branch-office forum under the guise of "carrying on business" everywhere. The stated purpose is anti-abuse — to stop large corporations with offices across the country from cherry-picking a convenient bench.

So far, so principled. The problems begin the moment one asks two questions the Sanjay Dalia bench never squarely answered.

Fault line one: Does the Sanjay Dalia rider even touch Section 20?

Sanjay Dalia was, on its own terms, an interpretation of Sections 134(2) and 62(2) — the special provisions. It restricted the additional forum those sections create. It did not purport to narrow the ordinary cause-of-action jurisdiction under Section 20 CPC. The Delhi High Court (single bench) in Travellers Exchange Corporation v. Celebrities Management (2019), held that Sanjay Dalia applies only when jurisdiction is invoked via Section 62 or 134, and not if the territorial jurisdiction is invoked under Section 20 of CPC because the cause of action arose in the forum.

If that reading is right, HUL's case is far stronger than the defendant's framing suggests. HUL has a branch office in Delhi, so it can say it carries on business in Delhi. HUL is also saying that part of its cause of action arose in Delhi — through disparaging content accessible here and products sold here. That is a Section 20(c) argument, and Sanjay Dalia arguably has nothing to say about it.

The defendant's answer — and the deeper worry — is that this distinction can be gamed. If "accessible on the internet" reliably converts into "cause of action arose here," then the Sanjay Dalia rider is dead on arrival: every plaintiff simply pleads Section 20(c) and points at the web. Which is why the two fault lines in this case are not parallel. They are interlocking. Whether Delhi has jurisdiction at all turns on whether digital accessibility manufactures a Section 20(c) cause of action in Delhi — and that is the second unresolved war.

Fault line two: The internet, and the two irreconcilable tests

Here the judgment is at its analytical best, because it resists the lazy move of treating all the internet-jurisdiction cases as saying the same thing. They do not.

The Banyan Tree line (2009, followed in FedEx Securities) sets a demanding, cause-of-action-focused test. Mere accessibility of a website in the forum — even an interactive one — is not enough. The plaintiff must show purposeful availment, specific targeting of the forum, and, where the "effects" test is invoked, an injury felt within the forum. Under Banyan Tree, the internet does not create a cause of action in every place a browser can reach.

The World Wrestling Entertainment / Kohinoor Seed Fields line attacks a different limb — "carries on business" — and reaches a far more expansive result. If a plaintiff (or, per Kohinoor, a defendant) runs an interactive website over which a transaction could be concluded, it is treated as carrying on business — as having a "brick-and-mortar store" — in every place the site is accessible. Actual completion of a transaction is "no longer indispensable."

Justice Bhambhani's sharpest observation is that these two lines operate on different statutory concepts — Banyan Tree on "cause of action" (Section 20(c)), WWE/Kohinoor on "carries on business" (Sections 20(a) and 134(2)). On paper, they need not conflict.

But the judgment then makes the crucial practical point: the distinction collapses in application. Once "carries on business" attaches wherever an interactive site is reachable, a nationwide e-commerce presence gives a plaintiff a nationwide menu of forums under Section 134(2) — the very forum-shopping Sanjay Dalia set out to kill, now reintroduced through the back door of "carrying on business." Further, Nilesh Girkar Vs. Zee Entertainment Enterprise Limited & Ors. (Divisional Bench, DHC 2025), held that where the film, that was subject to the proceedings, was being communicated to the public on the defendants’ OTT platform throughout the territories of India and was also available within the territorial limits of the jurisdiction of this court, in view of section 20(c) of the CPC, since at least a part of the cause of action had arisen within the jurisdiction of this court. The two tracks converge on the same alarming destination: jurisdiction almost everywhere. Para 52 captures the anxiety precisely — that territorial jurisdiction cannot be so diluted or be made so vague that a corporation "could sue in just about any location within the country," which would "throw the very concept of territorial jurisdiction of courts to the winds."

Fault line three: A Division-Bench-versus-Division-Bench standoff

What genuinely justifies a reference — as opposed to a decision — is not that the law is hard. It is that a Single Judge is bound by conflicting authority he cannot reconcile.

That is what has happened here. Ultra Home Construction (2016) faithfully systematised Sanjay Dalia into a four-scenario table, holding that where the cause of action arises at the principal office, the plaintiff cannot sue at the subordinate-office forum. But Kohinoor Seed Fields (2025), a co-equal Division Bench, expressly doubted that very proposition, suggesting that forcing a plaintiff to sue at its subordinate office because the cause of action arose there would amount to "re-writing Section 134 of the Trade Marks Act." Layer on Astral Ltd. Vs Ajay Enterprises (Single Judge DHC, 2025), which preserves concurrent jurisdiction at both the principal and subordinate office, and Rukhmani Keshwani vs. Raju Agarbatti Works & Anr. and ITC v. Adyar Gate Hotels (Divisional Bench, DHC 2026), which confine Sanjay Dalia to cases where the chosen forum "bears no real nexus" to the cause of action — and you have not a gap in the law but a genuine schism between benches of identical strength.

A Single Judge cannot pick a winner between two Division Benches. On this narrow but decisive point, the reference is not an evasion. It is the constitutionally correct route.

The defamation analogy: clever, but imperfect

The defendant's most creative argument borrows from Escorts Ltd. v. Tejpal Singh Sisodia (Single Judge, DHC 2019), a corporate-defamation case decided under Section 19 CPC, to warn against "court shopping" and "libel tourism." The parallel drawn is that disparagement of a trade mark is, in substance, defamation of a business.

The instinct is sound; the equivalence is not airtight. Defamation injures reputation at large, which is diffuse and hard to locate, making the "sue anywhere it was read" logic especially dangerous. Trademark disparagement injures goodwill and sales — commercial harm that is, at least in principle, more territorially traceable to where the goods are sold and the market is contested. A Larger Bench should resist importing Section 19's reasoning wholesale; the better analogy treats disparagement as a commercial tort with a locatable market effect, which points back toward Banyan Tree's "injury within the forum" requirement rather than a blanket "accessible-equals-actionable-everywhere" rule.

Was a decision possible? A respectful critique of the reference

Here is where a critical reading must push back gently. A reference is the right tool for a DB-versus-DB conflict — but did this case need to go up in its entirety, or could the suit have been decided on its facts with only the pure question of law referred?

There is a real argument that the case was decidable. HUL is one of the country's largest FMCG companies and undisputedly carries on business in Delhi (para 18 records its Delhi corporate office; nobody seriously contends HUL does not trade in Delhi). Under the WWE/Kohinoor line the defendant relies on elsewhere, that alone would let HUL invoke Section 134(2). The Sanjay Dalia bar engages only if the plaintiff is fleeing a principal-office forum where the cause of action also arose — and HUL is not obviously the "far-flung branch office" opportunist Dalia targeted; it is a national trader suing where it both operates and claims injury.

Against that, the honest counter is the one the judgment implicitly makes: HUL's Delhi jurisdiction ultimately rests on the digital limb, and the rule governing that limb is exactly what is broken. If Banyan Tree governs and the plaint pleads no purposeful availment or targeting by the defendant (the defendant's central complaint), the Delhi cause of action evaporates and Sanjay Dalia funnels the suit to Mumbai. If Kohinoor governs, the defendant's transactional website plants jurisdiction in Delhi. The outcome genuinely cannot be determined without first resolving the conflict. That is what rescues the reference from the charge of abdication — though a tighter judgment might have decided the "carries on business" point in HUL's favour and referred only the internet-cause-of-action question, sparing the parties a full detour to a Larger Bench on an interim application.

What the Larger Bench should actually fix

If the reference is to be worth the delay, three things need clean answers:

1. Decouple the limbs and say so expressly. "Cause of action" (Section 20(c)) and "carries on business" (Sections 20(a)/134(2)) are different inquiries with different tests. Most of the confusion is a category error in which courts apply Sanjay Dalia's carries-on-business rider to a plaintiff who is actually pleading cause of action, or import Banyan Tree's cause-of-action test into a carries-on-business analysis.
2. Confine Sanjay Dalia to its holding. It restricts the special forum under Sections 134(2)/62(2). It should not be read to amputate ordinary Section 20(c) jurisdiction where the cause of action genuinely arises in the forum — the Travellers Exchange reading is the more faithful one.
3. Pick a principled internet test, and let it be Banyan Tree. For a cause of action to arise online, mere accessibility should never be enough; there must be purposeful targeting of the forum plus a demonstrable effect there. "Carries on business via interactive website" should not be allowed to metastasise into a de facto pan-India forum that swallows the Dalia rider whole. The direction of travel in WWE, Kohinoor and Nilesh Girkar is convenient for plaintiffs but corrosive to the concept of territorial jurisdiction — a point para 52 makes with unusual candour for a reference order.

Practical takeaways for litigants and drafters

Until the Larger Bench speaks, prudent practice is unchanged and, if anything, sharpened:

• Plaintiffs: Do not rest a jurisdiction plea on bare website accessibility. Plead specific targeting of the forum, local sales with invoices or trap purchases, and injury felt in the forum. If part of the cause of action also arose where your principal office sits, weigh filing there — a Sanjay Dalia objection is cheap for the defendant to raise and expensive for you to lose.
• Defendants: The threshold objection remains live and, post-reference, temporarily unsettled — which cuts both ways. Preserve it, but expect courts to keep entertaining suits pending the Larger Bench, given that the issue arises "on a day-to-day basis."
• Everyone: Watch the docket. A Single Judge has openly asked the institution to resolve this "at the earliest convenience." The answer, when it comes, will reset internet jurisdiction for IP disputes across the country.

Conclusion

The most striking feature of HUL v. Kwick Living is its restraint. It would have been easy — and, for the parties, faster — to grab one of the available lines of authority, grant or refuse the injunction, and move on. Justice Bhambhani instead names the incoherence for what it is: a Supreme Court rider whose reach into Section 20(c) is unsettled, an internet-jurisdiction doctrine pulling in opposite directions, and Division Benches of equal strength openly disagreeing. On that terrain, a Single Judge who decided would only have deepened the split. The reference is the honest move.

Whether it is also the optimal move — versus deciding the case and referring only the abstract question — is a fair debate. But the larger point stands. Indian IP litigation has quietly drifted toward a world in which, thanks to the internet, everywhere is somewhere, and therefore anywhere will do. This reference is the judiciary catching itself at the edge of that cliff. The Larger Bench now has to decide whether territorial jurisdiction still means anything in the age of the accessible-everywhere website — or whether, as para 52 fears, the concept has already been thrown to the winds.