Police search of your home — what the law actually requires
A search of a private residence by the police is one of the most invasive acts the State performs against a citizen. The Code distinguishes a Court-issued warrant search under Sections 96 to 98 of the BNSS [Sections 93 to 98 CrPC] from a warrantless investigatory search under Section 185 BNSS [Section 165 CrPC]; Puttaswamy has overlaid the entire framework with a privacy lens that the doctrine is still working through.
No procedural protection in Indian criminal law has had a more uneven life than the law of search. The Code lays down two parallel regimes — a warrant search ordered by a Magistrate, and a warrantless search conducted by an officer in charge of a police station during investigation — each with its own preconditions, each surrounded by Supreme Court decisions that have, in the main, preferred admissibility of evidence to procedural purity. The Constitution Bench in Pooran Mal v. Director of Inspection, (1974) 1 SCC 345 held that even an illegal search does not exclude what was seized from being used as evidence. Forty-three years later, Justice K.S. Puttaswamy v. Union of India, (2017) 10 SCC 1 located the right to privacy in Articles 14, 19 and 21, and put the entire architecture of search and seizure — including the searches of digital devices that did not exist when the Code was drafted — into a new and unsettled frame. The doctrine is being worked out in the High Courts even as this is written.
The statutory framework
The provisions that govern a police search of a home are clustered in Chapter 7 of the BNSS [Chapter VII CrPC, "Processes to compel the production of things"]. Section 96 BNSS [Section 93 CrPC] empowers a Court to issue a search warrant in three situations: where a person to whom a summons under Section 94 [old Section 91] has been or might be addressed will not produce the document or thing; where the document or thing is not known to be in the possession of any person; and where the Court considers that the purposes of an inquiry, trial or other proceeding will be served by a general search. Section 97 BNSS [old Section 94] permits a District Magistrate, Sub-divisional Magistrate or a Magistrate of the first class to issue a warrant where any place is suspected of being used for deposit or sale of stolen property, or for the deposit, sale or production of certain objectionable articles. Section 100 BNSS [old Section 97] permits a search for a person confined under circumstances that amount to an offence. Section 103 BNSS [old Section 100] prescribes how the search is to be conducted — in the presence of two or more independent and respectable inhabitants of the locality, with a seizure list signed by the witnesses, a copy delivered to the occupant.
The warrantless investigatory search lives separately, in Section 185 BNSS [Section 165 CrPC]. Where an officer in charge of a police station has reasonable grounds for believing that anything necessary for the purposes of an investigation into any offence may be found in any place within the limits of his station, and that the thing cannot in his opinion be otherwise obtained without undue delay, he may, after recording the grounds in writing and specifying so far as possible the thing to be searched for, search or cause search to be made for such thing. Section 185 incorporates by reference the procedure of Section 103 BNSS — independent witnesses, seizure list, copy to the occupant.
Two further provisions matter for a complete picture. Section 105 BNSS [old Section 102] empowers any police officer to seize property which may be alleged or suspected to have been stolen, or which may be found under circumstances which create suspicion of the commission of any offence. Section 104 BNSS [old Section 101] governs the disposal of a thing seized in execution of a warrant beyond the local jurisdiction of the issuing Court.
The two regimes — warrant search vs warrantless search
The doctrinal distinction between the two regimes is the source of most of the litigation. A warrant search under Section 96 BNSS is a judicial act. The Court must apply its mind to the existence of one of the three conditions in sub-section (1) and record reasons. The Privy Council decision in Clarke v. Brojendra Kishore Roy Chowdhury, (1912) 39 Cal 953 established that pendency of an inquiry, trial or other proceeding is not a precondition for issuing a search warrant — the warrant may be issued in view of an inquiry about to be made. The Supreme Court in I.T.O. v. Seth Brothers, (1969) 2 SCC 324 cautioned that a search is a serious invasion of the rights and privacy of the individual, and a search warrant cannot be issued for the mere asking. State of Rajasthan v. Rehman, (1960) 1 SCR 991 is the early authority cited by Basu and Sarkar for the proposition that an order under what is now Section 96 must be specific to a definite document or thing; a general direction to produce "all papers relevant to the case" would be illegal.
Section 185 BNSS, by contrast, dispenses with the warrant altogether for the limited situation of an officer in charge of a police station already conducting investigation into a cognizable offence. The conditions are stricter than is sometimes recognised: the officer must have reasonable grounds for believing that the thing is in a place within the limits of his station, the thing must be necessary for the purposes of investigation, the officer must in his opinion conclude that it cannot be obtained without undue delay, and he must record the grounds and specify the thing in writing before the search. State of Maharashtra v. Natwarlal Damodardas Soni, (1980) 4 SCC 669 — though primarily a Customs Act case — is part of the Supreme Court line that has confirmed that even where a search itself is illegal, the seizure of an incriminating article is not vitiated and the article remains admissible in evidence. Radhakishan v. State of U.P., AIR 1963 SC 822 had laid down the rule earlier in narrower terms.
What separates the two regimes in practice is which body has applied its mind to the necessity of the search. In a Section 96 warrant search the Magistrate has. In a Section 185 warrantless search the police officer has, and the only contemporaneous safeguard is the written record of grounds. The absence of judicial pre-authorisation is the structural weakness of Section 185, and it is the part of the doctrine on which the post-Puttaswamy developments are most likely to bite.
What the Supreme Court has held
The doctrinal centre of gravity is Pooran Mal v. Director of Inspection, (1974) 1 SCC 345. The Constitution Bench was hearing a challenge to a search under Section 132 of the Income-tax Act, 1961, but the principle it laid down has been read across into the Code of Criminal Procedure. Where the search is illegal — because the officer lacked reason to believe, or did not record reasons, or seized articles beyond the scope of the authority — the materials seized nevertheless remain admissible in evidence. The Court relied on the English rule in Kuruma v. The Queen, [1955] AC 197 and rejected the American exclusionary rule as inapplicable to the Indian setting. The Indian position is that the test of admissibility is relevance; if evidence is otherwise admissible, the means by which it was obtained does not affect its admissibility, though it may affect the weight to be attached to it.
The consequence is that procedural non-compliance with Sections 96 or 185 BNSS, while it may expose the officer to action and may permit the Court to view the evidence of search with care, will rarely defeat the prosecution. State of Maharashtra v. Natwarlal (1980) put it in terms that have been quoted in every subsequent search case: "even where the search itself is illegal, the seizure of an incriminating article is not vitiated for the purposes of the trial." Shyam Lal v. State of M.P., AIR 1972 SC 886 added the gloss that absence of independent witnesses and failure to record reasons before entering the premises are instances of irregularity that do not, of themselves, go to jurisdiction.
The Supreme Court has, however, been prepared to set aside a warrant where it is patently without jurisdiction. Kuttan v. Ramakrishnan, AIR 1980 SC 185 — Basu cites this as Kuttan Pillai v. Ramakrishnan — held that no search warrant can be issued under Section 93(1)(a) CrPC [Section 96(1)(a) BNSS] to search for a document known to be in the possession of an accused, because that would breach the protection in Article 20(3) of the Constitution against testimonial compulsion. State of Gujarat v. Shyamlal, AIR 1965 SC 1251 had earlier excluded the accused from the ambit of Section 91 CrPC [Section 94 BNSS] on the same Article 20(3) ground.
The protection of Article 20(3) does not stop a general search of premises occupied by the accused under Section 96(1)(b) or (c) BNSS — the accused may remain a passive spectator, may even be absent, and what is found may be seized. State of Bombay v. Kathi Kalu Oghad, AIR 1961 SC 1808 had separated the seizure of objects from testimonial compulsion: a search and seizure of documents, as such, is not testimonial compulsion within Article 20(3).
Witnesses, seizure memo, and the chain-of-custody rule
Section 103 BNSS [Section 100 CrPC] is the procedural backbone of every search, warrant or warrantless. Sub-section (4) requires the searching officer to call upon two or more independent and respectable inhabitants of the locality in which the place to be searched is situated, or of any other locality if no such inhabitants of the locality are available or willing, to attend and witness the search. A list of all things seized in the course of such search and of the places in which they are respectively found is then prepared by the officer and signed by the witnesses. The occupant of the place searched, or some person on his behalf, is permitted to attend during the search. A copy of the list, signed by the witnesses, is delivered to the occupant.
The Supreme Court has treated the requirement of independent witnesses as procedural rather than mandatory in the sense that failure to comply vitiates the seizure. Sunder Singh v. State of U.P., AIR 1956 SC 411 and Modan Singh v. State of Rajasthan, AIR 1978 SC 1511 between them establish that the testimony of police witnesses to the search is not, in law, of weaker probative value than civilian witnesses; but the absence of independent witnesses is a circumstance that the Court may take into account in evaluating the seizure. Where independent witnesses are available and not joined, the prosecution must explain why. Where the search yields evidence that is the substantial part of the prosecution case, courts have set aside convictions on the ground that the seizure was not properly proved.
The seizure memo — sometimes called the seizure list or panchnama — is the contemporaneous document that ties the seized article to the place from which it was seized, to the witnesses who saw it being seized, and to the chain of custody that follows. A properly drawn seizure memo records the time of entry, the time of seizure, a description of each article (with serial numbers and identifying marks), the place within the premises where it was found, the names and addresses of the witnesses, and the signature of the occupant. The Supreme Court in State of Rajasthan v. Daud Khan, (2016) 2 SCC 414 and a long line of NDPS cases has held that absence of contemporaneous recording and discrepancies in the seizure memo are matters that can break the prosecution at trial.
The chain-of-custody rule is the natural consequence. From the moment of seizure, the article passes from the seizing officer to the malkhana in-charge of the police station, from there (where required) to the Forensic Science Laboratory, and from there back to the Court. Every transfer should be entered in the malkhana register, with date, time, signature of the receiving officer, and a description that ties to the seizure memo. Hira Singh v. Union of India, (2020) 20 SCC 272 is the recent NDPS Constitution Bench that, while not strictly a chain-of-custody case, has been cited for the proposition that the prosecution must prove the integrity of the chain from seizure to forensic analysis to production in court.
Where the courts are divided
The cleanest split is over the consequence of non-compliance with Section 185 BNSS [Section 165 CrPC] — the warrantless investigatory search. The dominant line, anchored in Pooran Mal, treats the failure to record reasons before the search, or the recording of perfunctory reasons that do not reflect application of mind, as an irregularity that does not vitiate the seizure. A minority line in the High Courts has, particularly in cases involving sensitive evidence — narcotics, currency, electronic devices — treated material non-compliance as fatal, on the ground that Section 165 was drafted as an exception to the warrant requirement and the exception must be strictly construed. The Bombay and Kerala High Courts have, in NDPS cases, set aside convictions where the recording of grounds under Section 42 of the NDPS Act (which is the cognate provision in that statute) was found to be defective; the Supreme Court in State of Punjab v. Balbir Singh, (1994) 3 SCC 299 had earlier read Section 42 strictly, though subsequent decisions have softened the position.
A second split, and a more contemporary one, concerns the search of digital devices found at a residence. Section 185 BNSS, drafted with a paper-and-cupboard search in mind, says nothing about whether the seizure of a phone or a laptop authorises the officer to access its contents. The Karnataka High Court in Virendra Khanna v. State of Karnataka, 2021 SCC OnLine Kar 5032 attempted to lay down guidelines for the search of electronic devices, including a requirement of judicial supervision and the use of hash values to preserve evidentiary integrity. The Delhi High Court in Amazon Seller Services v. Amway India Enterprises and other decisions has, separately, considered the limits on the seizure of digital materials. The Supreme Court in Ritesh Sinha v. State of U.P., (2019) 8 SCC 1, dealing with the compulsion to give voice samples, gestured at the privacy implications under Puttaswamy but did not lay down a comprehensive framework for digital searches.
A third unresolved area is the relationship between Sections 94 BNSS [Section 91 CrPC] (summons to produce) and Sections 96 and 185 BNSS. The Privy Council in Clarke and the Supreme Court in Kuttan have laid down that a summons under Section 94 is not a precondition for a warrant under Section 96 or for a warrantless search under Section 185; but some High Courts, in cases where a search has been conducted without prior summons in circumstances where the document was clearly available on demand, have suggested that the failure to attempt the less invasive route may be a factor going to mala fides. The Supreme Court has not definitively settled this.
Where the law is heading
The privacy lens supplied by Justice K.S. Puttaswamy v. Union of India, (2017) 10 SCC 1 is the development that will, in the next decade, redraw the search-and-seizure doctrine. Puttaswamy held that the right to privacy is a fundamental right that flows from Articles 14, 19 and 21, and that any State interference with that right must satisfy a four-part test: legality (the action must be authorised by a law), necessity (in pursuit of a legitimate State aim), proportionality (the means must be proportionate to the object), and procedural safeguards (against arbitrary State action). The test was reaffirmed in Puttaswamy (Aadhaar), (2019) 1 SCC 1.
Each of the four limbs has implications for police searches. The legality limb requires that the empowering statute itself be sufficiently precise — a vague authorisation will fail. The necessity limb invites a question about whether a warrantless search under Section 185 BNSS, in the era of telecommunication that permits a warrant to be obtained in hours rather than days, can still be said to be necessary as a routine matter, or whether it must be confined to genuine emergencies. The proportionality limb is most consequential for digital searches: a phone today contains the entirety of a person's correspondence, medical records, financial life, intimate photographs, location history; a search authorised in language drafted to cover a tin box in a cupboard is, on its face, disproportionate to that volume of intrusion. The procedural-safeguards limb is what the Karnataka High Court was groping for in Virendra Khanna.
The Supreme Court has not yet had a head-on occasion to apply the Puttaswamy four-part test to Sections 96, 100, 103 and 185 BNSS. When it does — and the petitions are being filed — the most likely outcomes are these. First, judicial recognition that the warrantless search power under Section 185 must be confined to circumstances of genuine urgency, with contemporaneous recording of why a warrant could not have been obtained. Second, a separate framework for searches of digital devices, requiring judicial pre-authorisation of the scope of the search, hash-value preservation of the original, and a limitation on what data may be examined. Third, a stricter approach to the requirement of independent witnesses under Section 103, with greater reliance on body-worn camera footage where it exists. Fourth, perhaps, a re-examination of Pooran Mal — though the Court has been unwilling, in Mathew v. State of Kerala and other recent matters, to revisit the exclusionary rule in its entirety, the doctrine may yet be modulated for searches that breach a fundamental right rather than a mere statutory procedure.
The practitioner working in this area today should treat the law as settled at one level and unsettled at another. At the statutory level, Sections 96 to 105 and 185 of the BNSS are the framework. At the constitutional level, the framework is in motion. An objection to an illegal search at trial will not, on present authority, lead to exclusion of the evidence; but it should be raised and recorded, because the High Court hearing the appeal — or the Supreme Court hearing the special leave petition — may yet treat the breach of procedure as a breach of Puttaswamy's proportionality test, and may yet, in the right case, hold that an illegal search of a home is not a mere irregularity but a constitutional wrong.