The Jurisprudence of Arbitrability (PART 3)

Kompetenz-Kompetenz and Arbitrability: Who Really Gets to Say “Stop”?

Picture this: Two companies are locked in a bitter dispute. One rushes to court. The other waves the arbitration clause and says, “Not so fast – we go to arbitration.”

The court leans back and asks the million‑dollar question:

Who decides first whether this fight can even be arbitrated – the court or the arbitral tribunal?

That is the drama at the heart of Kompetenz-Kompetenz and arbitrability.

Below is a simple, case‑rich, India‑focused, yet globally grounded walkthrough of this entire Chapter.

1. Meaning and evolution of the doctrine of Kompetenz-Kompetenz

What does it mean?

Kompetenz-Kompetenz (or competence-competence) is the principle that an arbitral tribunal has the power to rule on its own jurisdiction, including any objection to:

• existence of the arbitration agreement

• validity of the arbitration agreement

• scope/applicability of the clause

This is now a pillar of modern arbitration, recognised in:

• UNCITRAL Model Law, Article 16(1) – tribunal “may rule on its own jurisdiction, including any objections with respect to the existence or validity of the arbitration agreement” - English Arbitration Act 1996, section 30(1)– tribunal may rule on substantive jurisdiction - Major institutional rules – UNCITRAL Rules, ICC Rules, LCIA Rules, ICDR Rules (Art. 19). In India, section 16(1) of the Arbitration and Conciliation Act, 1996 expressly incorporates Kompetenz-Kompetenz, allowing the tribunal to rule on its own jurisdiction, including objections to existence or validity of the arbitration agreement.

  
  

How did it evolve?

• Under the Arbitration Act 1940, Indian law had no equivalent to section 16 – jurisdictional issues were heavily court‑centric.

• The 1996 Act, modeled on the UNCITRAL Model Law, flipped the script: tribunals were empowered to decide jurisdiction “in the first instance”, with minimal judicial intervention as the legislative policy.

• Internationally, Model Law and pro‑arbitration statutes spread the idea that arbitrators should decide jurisdiction first, with courts reviewing only later (set‑aside or enforcement).

But this neat design ran into a problem: courts do not easily let go.

Indian cases like SBP & Co. v. Patel Engineering Ltd. (2005) expanded judicial review at the section 11 appointment stage and “chipped away” at the effect of section 16 and the negative side of Kompetenz-Kompetenz. That sets the stage for the amendments.

2. Positive and negative effects of Kompetenz-Kompetenz

Scholars and courts now talk about two faces of Kompetenz-Kompetenz:

Positive effect

This is the tribunal’s authority:

• The tribunal may rule on its own jurisdiction, including challenges to the arbitration agreement.

• Recognised in Model Law Article 16(1), section 16 of the 1996 Act, and rules like UNCITRAL, ICC, LCIA, ICDR.

Upside:

• Quick handling of objections – saving time and costs- Avoids tactical court suits meant to delay arbitration - Aligns with party autonomy – parties chose arbitrators; let them decide.

  
  

Negative effect

This is the “hands off, for now” rule for courts:

• Courts should not fully decide jurisdiction/arbitrability at the outset; they generally defer to the tribunal first, reviewing its decision only later (set‑aside, enforcement, or limited interlocutory recourse).

• It gives the tribunal chronological priority on jurisdiction.

But the strength of this negative effect varies:

• Some systems (e.g., Nigeria’s new law, French‑influenced practice) push strongly toward tribunal‑first and limited review.

• Others (e.g., India for many years, some common‑law courts) keep broad space for courts to step in early, weakening the negative effect and encouraging jurisdictional skirmishes in court.

  
  

3. Can tribunals rule on their own jurisdiction when arbitrability is challenged?

Arbitrability asks: Is this kind of dispute even allowed to be settled by arbitration, or is it reserved for courts? Two main dimensions:

• Subjective arbitrability – can this party arbitrate (e.g., states, public entities needing authorisation) - Objective arbitrability – can this type of dispute be arbitrated (e.g., criminal matters, certain insolvency, family status, some public‑law issues). Under Kompetenz-Kompetenz, most modern regimes allow the tribunal, at least in the first instance, to rule on arbitrability as part of its jurisdiction:

• Model Law Article 16 and national copies (including section 16, 1996 Act) cover “jurisdiction” broadly, and scholarship treats arbitrability as part of that jurisdictional analysis.

• Tribunals regularly decide whether the dispute is within the scope of the arbitration clause and whether any non‑arbitrability rule is triggered.

However, there is a safety valve:

• Courts almost always retain the power, at the annulment or enforcement stage, to refuse recognition if the dispute was non‑arbitrable or the award violates public policy.

• So the tribunal can rule on arbitrability first, but not necessarily last.

  
  

4. Limits of arbitral jurisdiction in non‑arbitrable disputes

Kompetenz-Kompetenz is not a blank cheque. Arbitrators cannot decide everything.

Key limits (often expressed as non‑arbitrability and public policy) include:

• Subjective limits – some entities (e.g., certain state bodies) may lack capacity to enter arbitration agreements without special approval.

• Objective limits – whole subject areas may be carved out:

  • criminal liability

  • some family‑law/status issues

  • some insolvency or winding‑up proceedings

  • core public‑law and regulatory measures, depending on local law

If a dispute is non‑arbitrable under the applicable law:

• The arbitration agreement may be invalid, inoperative, or incapable of being performed.

• The tribunal lacks jurisdiction, and any award risks being set aside or denied enforcement on public policy or non‑arbitrability grounds.

Arbitrability rules are highly jurisdiction‑specific and depend on how far a State is willing to entrust sensitive matters to private adjudication.

5. Indian judicial approach post‑2015 and 2019 amendments

Now to the Indian story – full of twists.

Statutory framework

Key provisions of the Arbitration and Conciliation Act, 1996 on this topic:

• Section 5 – minimal judicial intervention.

• Section 7 – arbitration agreement.

• Section 8 – reference to arbitration when matter is subject to arbitration agreement.

• Section 11 – appointment of arbitrators (site of the big Kompetenz-Kompetenz battle).

• Section 16 – Kompetenz-Kompetenz: tribunal may rule on its own jurisdiction.

• Section 34(2)(b) – award may be set aside if subject matter is not capable of settlement by arbitration, or award conflicts with public policy (non‑arbitrability gateway).

The early blow: SBP & Co. v. Patel Engineering Ltd. (2005)

• The Supreme Court held that Section 11 powers are judicial, not administrative.

• Courts could examine issues like validity of the arbitration agreement, arbitrability, limitation, accord and satisfaction at the appointment stage.

• This undercut Section 16 and the negative effect of Kompetenz-Kompetenz, leading to heavy front‑loaded court review.

The corrective: 2015 Amendment – Section 11(6A)

The Arbitration and Conciliation (Amendment) Act, 2015 inserted:

• Section 11(6A) – court’s examination at appointment was confined only to the existence of an arbitration agreement.

Purpose:

• To reverse SBP’s broad approach, strengthen Section 16, and align more closely with the negative effect of Kompetenz-Kompetenz – tribunal first, court later.

In parallel, Supreme Court decisions such as Duro Felguera S.A. v. Gangavaram Port Ltd. (2017) and Mayavati Trading v. Pradyuat Deb Burman (2019) endorsed the narrow 11(6A) view, stressing that courts should not wade into deeper jurisdictional or arbitrability issues at the Section 11 stage (this specific case law is not in the papers but is core to the doctrinal narrative).

The twist: 2019 Amendment and omission of Section 11(6A)

The 2019 Amendment Act then omitted Section 11(6A).

Yet scholarship notes that, even after this deletion, Indian jurisprudence has been slowly assimilating the full dual (positive and negative) effect of Kompetenz-Kompetenz, though often with a residual “inherent distrust” leading to recurring judicial intervention.

Combined with case law on non‑arbitrability and reference under Section 8 and Section 11 (e.g., Booz Allen, A. Ayyasamy, Vidya Drolia – not in the dataset but central in practice), India now shows a mixed character:

• On paper – pro‑tribunal, pro‑arbitration, section 16 intact.

• In practice – courts sometimes still take a robust role on arbitrability and jurisdiction at the threshold.

  
  

6. Comparative perspective: UNCITRAL Model Law jurisdictions

How do other Model Law jurisdictions handle Kompetenz-Kompetenz and arbitrability?

Across Model Law states, one sees a shared positive effect (tribunal may rule on jurisdiction), but differing negative effect (how far courts stand back initially).

Comparative snapshot

Jurisdiction / model	Basic stance on tribunal’s jurisdiction	Who decides first?
UNCITRAL Model Law archetype (Art. 16)	Tribunal may rule on its own jurisdiction; courts may later review via set‑aside / enforcement.	Typically tribunal first during arbitration; courts step in later or on limited recourse.
France‑influenced “strong negative effect”	Courts do only prima facie review at referral stage, leaving full jurisdiction and arbitrability to arbitrators first.	Tribunal clearly first; courts mainly at annulment/enforcement.
Nigeria (ACA 2004 & AMA 2023)	Strong judicial non‑interference; AMA s.14(5)–(6) gives near‑final effect to tribunal’s jurisdictional ruling, ousting most court review.	Tribunal first and often last; court review largely ousted, except narrow interlocutory channel.
England (s. 30, 1996 Act)	Tribunal can rule on jurisdiction; court retains power to decide jurisdiction de novo on challenge, but often respects tribunal’s first look.	Mixed – either forum may be seized; tribunal usually first but court can step in early.
US (post‑Henry Schein)	Strong respect for agreements delegating “arbitrability” to arbitrators – courts must honour clear delegation and avoid “wholly groundless” review.	If delegation clause is clear, tribunal first on arbitrability; courts step back.

Figure 1: Different national implementations of Kompetenz-Kompetenz and arbitrability.

Most Model Law systems accept that:

• Tribunals can and should rule on their jurisdiction (including arbitrability) during the proceedings.

• Courts retain a final control function, often grounded in non‑arbitrability and public policy, at the enforcement or set‑aside stage.

Where they differ is exactly the question you care about most.

7. Core issue: Who decides arbitrability first – the court or the tribunal?

Strip away the Latin and statutes, and the question is stark:

When one party says, “This dispute is not arbitrable,”who gets the first serious shot at answering – the arbitral tribunal or the court?

Three broad models emerge

1. Tribunal‑first / strong negative effect model

   • Tribunal must decide jurisdiction (including arbitrability) first.

   • Courts will only perform limited or prima facie review at the outset, saving full scrutiny for set‑aside or enforcement.

   • Seen in France and some robust pro‑arbitration systems, and strongly in Nigeria’s 2023 AMA where tribunal jurisdictional rulings can be practically final.

2. Concurrent / flexible Model Law approach

   • Both courts and tribunals have competence to decide jurisdiction; which speaks first depends on who is seized first and national judicial attitudes.

   • Courts in many Model Law states aim to respect Kompetenz-Kompetenz but may still entertain early challenges in particular circumstances.

3. Hybrid “court‑helpful” common‑law models (including India in practice)

   • Statutes adopt Kompetenz-Kompetenz (e.g., section 16 in India; Article 16 Model Law).

   • Yet courts keep significant discretion to:

     • decide jurisdiction/arbitrability at referral or appointment stages, or

     • allow de novo review of tribunal jurisdictional rulings.

   • This can dilute the negative effect and encourage parties to front‑load arbitrability battles in court, undermining efficiency.

     
     

So who “should” decide first?

From the perspective of arbitration policy and efficiency:

• The doctrine of Kompetenz-Kompetenz aims squarely at tribunal‑first:

  
  

  let arbitrators decide jurisdiction (including arbitrability) promptly, with later, focused judicial control to protect non‑arbitrability and public policy.

From the perspective of public law and control:

• Courts insist that they remain the final guardians of:

  • non‑arbitrable subject matters,

  • validity of arbitration agreements, and

  • fundamental procedural safeguards.

The best‑designed systems, as comparative scholarship stresses, look for functional complementarity:

• arbitrators have room to act first,

• courts retain the last word, but only where necessary to protect core limits like non‑arbitrability and public policy.

  
  

Final takeaway

Kompetenz-Kompetenz is not just about who has the power; it is about who goes first, who goes last, and how much they trust each other.

• Tribunal‑first models bet on speed, autonomy, and confidence in arbitral expertise.

• Court‑heavy or hybrid models bet on caution, review, and public‑law control.

Where a country falls on that spectrum is, in reality, its answer to a deeper constitutional question:

How much of our justice are we willing to privatise – and on whose word?