Privilege for The Privileged?

The Supreme Court of India’s (“SC”) judgment, In Re: Summoning Advocates who give legal opinion or represent parties during investigation of cases and related issues (“In Re: Summoning Advocates”), arrives at a moment when investigative agencies in India are actively undertaking enforcement actions. High-profile instances in which summons were first issued and then withdrawn, exposed, as some would argue, both the expanding overreach of enforcement agencies and the need to set clear boundaries when such actions are undertaken.

The SC was therefore confronted with questions at the heart of what the attorney-client privilege means in Indian jurisprudence and what it should mean. When can an investigating officer summon an advocate; what protections apply when documents or devices are sought, and should these be treated differently from a summons; and how should the law respond to digital practice, where privileged communications and evidentiary material often coexist on the same device, sometimes across multiple clients?

Did the SC need to step in and codify permissible actions around communications and documents, and do these directions provide adequate constitutional safeguards? This article examines the scope of those protections and what privilege survives through them.

While courts have occasionally restrained attempts to summon practising advocates, this is among the first decisions where the SC directly addresses the legality of such summons during an investigation. The SC declined to frame guidelines under Article 142 of the Constitution, instead grounding its reasoning in statutory interpretation and rejecting proposals for external or peer-review mechanisms.

On privilege, the SC held that Section 132 Bhartiya Sakshya Adhiniyam, 2023 (“BSA”) protects confidential professional communications across all contexts, and that advocates may be summoned only where statutory exceptions are expressly invoked with the written approval of a Superintendent of Police. On document production, it held that documents may be summoned under Section 94 of the Bharatiya Nagarik Suraksha Sanhita, 2023 (“BNSS”), or its civil equivalents under Section 165 BSA and Order XVI Rule 7 Code of Civil Procedure, 1908 (“CPC”), but must be produced before the court adjudicates privilege objections.

For “digital devices” to be produced by an IO, they shall be produced only before the relevant court. Some safeguards around this process are suggested, including hearing the party and the advocate, and, if the objection is overruled, procedural safeguards such as opening the device in front of the party, the advocate, and another person with expertise in digital technology (for example, a forensic advisor). The SC also cautioned that the confidentiality of other client data must be preserved and that discovery must be confined to material found to be permissible and admissible.

Crucially, relying on older precedents such as Gangaram v. Habib-Ullah and The Public Prosecutor, Madras v. M.S. Menoki of Calicut, the SC affirmed that documents must first be produced and inspected before objections are decided, and that a lawyer cannot claim privilege over documents where the client itself could not.

On in-house counsel, the SC held that salaried employment excludes them from the definition of “advocate” under the Advocates Act, 1961, relying on Satish Kumar Sharma v. Bar Council of Himachal Pradesh and Akzo Nobel Limited v. European Commission. It emphasised that privilege requires independence, and therefore does not extend to in-house counsel under Section 132 BSA.

The SC’s ruling on document production, in particular, results in an entirely unsatisfactory outcome. First, when documents are summoned from an advocate, they must be produced, subject to any objections. The court will peruse the document and then determine whether the objection is sustained or otherwise. If the document is shown to contain legal advice or facts which the lawyer has come to know in the course of the engagement, presumably this would not be permissible to produce.

But the SC does not go so far as to say the latter. In particular, when it comes to facts or strategic advice based on information given by a client during the course of the engagement, in confidence, it will become especially difficult for a court to (1) parse through the data and (2) determine the nature of the data, including whether it includes any strategic legal advice.

Legal advice is not always explicit in a document; it may appear through draft exchanges or materials without any written advice. Requiring such documents to be first produced, examined by the court, and debated between the lawyer, client, and IO before privilege is granted might create serious practical risks. This exercise, and even the procedural steps set up, will not withstand scrutiny in practice, and there is a danger that the process will eventually lead to the privileged data being revealed or to the privilege being given up as a result of the hearing itself. The SC’s reliance on limited precedent to justify this approach appears to be inadequate, especially in criminal proceedings where privilege is closely tied to the protection against self-incrimination under Article 20(3).

Secondly, while the SC declined to formally “issue guidelines,” it did not leave the matter to unfettered investigative discretion. Instead, it introduced procedural safeguards such as written superior approval for summonses, recorded justification for invoking Section 132 BSA exceptions, production before a court rather than private inspection, and access to judicial review. Though framed as an interpretation, these safeguards function as de facto guidelines and achieve substantially what the petitioners sought. If this were indeed the case, some argue that the SC could well have gone the whole hog and issued detailed, robust guidelines, particularly in the sphere of digital device and document production. Instead, this decision sits uncomfortably in the middle, stopping short of providing detailed procedural steps and watertight constitutional protections against the overreach of investigating agencies.

This is evident in the treatment of Section 132 BSA exceptions. IOs may proceed on the basis of a recorded belief, subject to supervisory approval, with judicial review largely ex post. This framework may be tactically weaponised by pressuring advocates, chilling clients, and triggering costly interlocutory litigation, where even eventual judicial intervention may come too late to undo the harm.

From a policy perspective, these safeguards do not address the structural asymmetry inherent in investigative demands for privileged material. Indian criminal procedure continues to treat the IO as the primary fact-finder, with supervisory approval as the principal internal check, an approach ill-suited when the target is an advocate performing a representative constitutional role. This mismatch is compounded by the fact that the initial determination of whether a Section 132 BSA exception applies rests with the investigating agency itself. One wonders, then, whether this presents an adequate framework to prevent IOs from accessing the defence strategy. Without an independent, prior adjudicatory check to prevent disclosure until privilege is resolved, the safeguards risk becoming perfunctory.

Lastly, the decision implicitly distinguishes between privileged “communications” and “documents or things” with independent evidentiary value. Relying on Gangaram and M.S. Menoki, the SC treats confidential professional communications as protected under Section 132 BSA, but allows documents to be summoned if statutory conditions are met; accordingly, an advocate cannot resist production at the threshold under Section 94 BNSS or Order XVI Rule 7 CPC read with Section 165 BSA, and privilege is decided only after judicial inspection.

This distinction is difficult to sustain in the digital context. Modern devices store emails, drafts, chats, attachments, and metadata in integrated threads, making it artificial to separate legal advice from underlying material. Granting access to an advocate’s device, therefore, risks exposing privileged material before any privilege determination is made, and the SC leaves unresolved key safeguards such as who images the device, how data is verified, what searches are run, and how unrelated or personal data is protected. Without clear forensic standards, privilege becomes contingent on resources and procedure, and lawyer-client communication is likely to be chilled.

The SC’s decision undoubtedly stabilises the doctrinal core of the attorney-client privilege and clarifies the position of privilege from the standpoint of in-house counsel. From those standpoints, the decision is helpful. But it does not do enough to meet the challenges of contemporary practice.

As this article has shown, the judgment relies on procedural steps that are often subject to the vagaries of the parties concerned rather than on prior structural protections. What emerges, some would argue, is a landscape of uncertainty, in which privilege could depend not only on doctrine but also on the discretion of individual officers, the technical capacity of trial courts, and the resources of the parties who can bring in tech-savvy forensic experts.

The judgment is therefore best understood as a transitional moment. It acknowledges the pressures of modern investigative practice but falls short of providing equally protective safeguards. Until that gap is addressed, the promise of attorney-client privilege in India should be dealt with caution and care.

About the authors: Shalaka Patil is a Partner at Trilegal.

Arnav Mathur is a student at NALSAR University of Law.

Disclaimer: The opinions expressed in this article are those of the author(s). The opinions presented do not necessarily reflect the views of Bar & Bench.

If you would like your Deals, Columns, Press Releases to be published on Bar & Bench, please fill in the form available here.