Investigators have relied on scars and tattoos to identify bodies and suspects for as long as investigation itself has existed. What has changed is scale, not substance. A tattoo once described in words to a magistrate can now be photographed, digitised, stored in a national database, and, in some jurisdictions, run through recognition software that claims to infer affiliation rather than mere identity.1 Indian legal doctrine has not kept pace. That gap between what technology can now do and what settled legal principle actually permits is where this paper begins.
Put a tattoo before a court, and it can answer three quite different questions. Whose body is this. Who does this person associate with. What was done to this person. Section 7 of the Bharatiya Sakshya Adhiniyam 2023 simply continues the old Section 9 formula from the Indian Evidence Act 1872, treating identity-establishing facts as relevant without ever distinguishing a chosen, meaningful mark from an accidental scar.2 American courts have gone further, building a substantial body of case law under Federal Rule of Evidence 403 on gang-tattoo evidence, and reports from the field have documented how traffickers brand their victims.3 These three bodies of knowledge, forensic identification science, comparative association-evidence jurisprudence, and anti-trafficking scholarship, rarely speak to one another, though they concern the same physical object. Indian doctrine has drawn on none but the first. This tripartite ambiguity matters because each question carries a different risk profile: an error in identity evidence can convict an innocent person outright, an error in association evidence can brand a lawful expressive choice as proof of criminal disposition, and an unrecognised branding mark can leave a trafficking survivor’s testimony uncorroborated when corroboration was available all along.
This paper sets four objectives: to map Indian doctrine on tattoos as identification evidence; to draw usable lessons from comparative association-evidence jurisprudence in the United States, United Kingdom, Canada, and Australia; to examine the forensic literature on branding tattoos for its evidentiary implications; and to test how Articles 20(3) and 21 of the Constitution apply to each of these three uses, proposing concrete reform wherever the existing framework comes up short.
This paper follows a doctrinal method, supplemented by comparative and forensic literature wherever Indian sources fall silent. Primary sources include the Bharatiya Sakshya Adhiniyam 2023, the Bharatiya Nagarik Suraksha Sanhita 2023, the Bharatiya Nyaya Sanhita 2023, the Criminal Procedure (Identification) Act 2022, and the Immoral Traffic (Prevention) Act 1956, read against the relevant case law. The comparative material draws on American, British, Canadian, and Australian decisions, chosen because courts in those jurisdictions have confronted gang-tattoo evidence far more often than Indian courts have. Secondary sources include forensic medicine literature, clinical scholarship on trafficking-related branding, and reporting on automated tattoo recognition by the US National Institute of Standards and Technology.4
This paper confines itself to tattoos already present on a person’s body. It does not reach forcible removal, tattoo-based sentencing enhancements, or civil disputes over tattoo copyright. Nor does it offer an empirical survey of Indian trial-court practice, a limit the poor digitisation of trial-court records imposes rather than one this paper chooses. Its proposed framework for association evidence is necessarily prescriptive, since no reported Indian appellate decision squarely addresses the question; the framework offers a proposal for how courts might approach the issue once it arises, not a description of settled law. One further limit deserves mention: the paper relies on English-language reporting of Indian, American, British, Canadian, and Australian case law, and makes no claim to canvass vernacular-language commentary or unreported trial-court orders sitting outside the digitised record. Where comparative material is drawn upon, this paper relies on reported appellate decisions and does not purport to capture unreported first-instance rulings, plea colloquies, or sentencing memoranda in which tattoo evidence may also feature.
Forensic medicine literature treats the tattoo as a biometric feature, useful chiefly for identifying decomposed or skeletal remains, and its concerns run to descriptive accuracy and ink durability rather than to the legal weight a match ought to carry once it reaches trial. Indian criminal procedure scholarship, for its part, addresses identification evidence broadly through the test identification parade, without ever isolating tattoos as a distinct sub-category or asking what actually differs between identifying a person by a mark and identifying them by a face.
American, British, Canadian, and Australian scholarship has grappled directly with the symbolic dimension of tattoos, because courts in each of these jurisdictions have repeatedly had to decide whether a tattoo may be shown to a jury as proof of gang membership, motive, or character.5 This literature is preoccupied, rightly, with weighing probative value against unfair prejudice, and with the expressive-conduct implications of treating body art as an admission of belief. Indian scholarship has almost nothing to say on the subject, despite the organised-crime provision now sitting in the Bharatiya Nyaya Sanhita 2023, which would make such evidence just as relevant in an Indian prosecution as in an American one.
Forensic dermatology and advocacy literature documents the practice of branding trafficking victims with tattoos denoting ownership. It is written mainly for clinicians, training doctors and first responders to spot a branding mark, rather than for lawyers building a case, though prosecutors and international bodies have treated such marks as corroboration where a survivor’s testimony alone is difficult to obtain or to weigh.6
No existing work, so far as this paper has found, treats identification, association, and exploitation tattoos as points on a single evidentiary spectrum. Each is addressed by a different disciplinary community, evaluated against a different standard of reliability, and this paper’s contribution is to show why that separation matters. Indian law, lacking developed doctrine for two of the three uses, risks importing assumptions suited to ordinary identification evidence into contexts, gang prosecutions and trafficking cases, where those assumptions simply do not hold. The gap is not academic. As Sections 111 and 143 of the Bharatiya Nyaya Sanhita 2023 begin generating their first wave of prosecutions, Indian trial courts will need a workable answer sooner rather than later, and they will need it with fewer scholarly resources than their American, British, Canadian, or Australian counterparts had when they first confronted the same question.
Section 9 of the Indian Evidence Act 1872 made facts establishing identity relevant, and a tattoo fell comfortably inside it. Courts relied on that provision routinely, admitting identification marks to corroborate an eyewitness or to establish that a mutilated body belonged to a particular person. The Bharatiya Sakshya Adhiniyam 2023 carries the scheme forward unchanged in Section 7, without ever pausing to ask whether a chosen mark belongs in the same category as an accidental scar.7 The Criminal Procedure (Identification) Act 2022 now authorises the photographic capture of tattoos at arrest and their retention by the National Crime Records Bureau for up to seventy-five years, a considerably wider power than its colonial-era predecessor, the Identification of Prisoners Act 1920, which reached only convicts, persons ordered to furnish security for good behaviour, and persons arrested in connection with an offence punishable with rigorous imprisonment for a year or more, and which allowed photographs, as distinct from measurements, to be taken of an arrested person only on a Magistrate’s order.8 Where a tattoo is used to help identify an accused, the governing procedure is the test identification parade under Section 54 of the Bharatiya Nagarik Suraksha Sanhita 2023. The Supreme Court has held, repeatedly, that such a parade is corroborative rather than substantive evidence, and courts have long recognised that a distinctive mark visible on only one parade member defeats the whole point of the exercise.9 In Joginder alias Danny v State (NCT of Delhi) a headless body was identified from height, build and a tattoo on the hand; in Manjunatha v State of Karnataka a headless body was identified from a Shivalinga tattoo on the hand. Both uses are consistent with the settled principle that identification evidence corroborates rather than substitutes for substantive testimony.10 What the framework fails to do is assess a tattoo’s rarity within the relevant population before treating a match as significant. A fingerprint is generally accepted as reliable identification evidence because it rests on an established body of forensic science. No comparable science exists for tattoos. People who have never met each other can easily share a common design, a popular name, or a widely used religious symbol.
A tattoo offered to prove gang membership asks the fact-finder to travel a much longer road: from an image to a belief. American courts, applying Federal Rule of Evidence 403, insist on an evidentiary bridge between the specific tattoo and the specific offence charged. In United States v Thomas, the Seventh Circuit held that a photograph of the accused’s tattoo of two crossed revolvers should never have gone before the jury on a charge of possessing a firearm as a convicted felon: the court could see no purpose in the photograph other than to establish a propensity to possess guns, and held it merely propensity evidence of no other probative value. The Sixth Circuit reached the same conclusion on the same charge in United States v Newsom, and in United States v Ford it confirmed the converse, holding gang-affiliation evidence admissible where the relationship between alleged co-conspirators was itself in issue while reaffirming that such evidence has no place in a trial where nothing connects it to the offence charged. In People v Bynum, the Michigan Supreme Court permitted expert testimony on gang symbolism only where the offence itself supplied an independent link to gang activity.11 British courts reach a broadly similar result through the bad character provisions of the Criminal Justice Act 2003 and the fairness discretion under Section 78 of the Police and Criminal Evidence Act 1984, admitting tattoo evidence only through a statutory gateway and only after a probative-value assessment.12 Canadian courts apply the expert-evidence test from R v Mohan, demanding a demonstrated interpretive methodology before a tattoo can be characterised as a gang symbol. Australian courts exclude prejudicial tattoo evidence under Sections 135 and 137 of the Evidence Act 1995 (Cth) absent a demonstrated link to the offence charged.13 Section 111 of the Bharatiya Nyaya Sanhita 2023 now supplies India’s own organised-crime provision, against which such evidence could be tested, but no Indian appellate decision has yet built a comparable balancing framework. That is a matter of the provision’s youth, not of any considered rejection of the comparative approach.14
Traffickers frequently mark victims with tattoos meant to signal ownership: a trafficker’s name or nickname, initials, a currency symbol, or a barcode-style design denoting a debt the victim is told they must work off. These marks turn up both in visible locations, meant to assert control publicly, and in concealed ones, including the breast, eyelid, or oral mucosa, where only the trafficker is likely to notice them. The United Nations Office on Drugs and Crime has pointed to branding tattoos, among them a Chinese character denoting slave tattooed on the ankle of a woman in the United States, as physical evidence capable of corroborating a prosecution that would otherwise rest entirely on a survivor’s own difficult testimony, and the EU Anti-Trafficking Directive requires member states to treat an offence committed by the use of serious violence, or one that has caused particularly serious harm to the victim, as an aggravating circumstance carrying a higher maximum penalty, a category into which coercive marking may fall although the Directive does not name it.15 Section 143 of the Bharatiya Nyaya Sanhita 2023 criminalises trafficking broadly and renders the victim’s consent irrelevant. But neither it nor the Bharatiya Sakshya Adhiniyam so much as mentions branding tattoos as corroborative evidence, despite forensic dermatology literature offering a structured, comparatively cheap protocol for recognising them, one attentive to location, inking quality, and evidence of concealment.16
The US National Institute of Standards and Technology ran a Tattoo Recognition programme funded by the Federal Bureau of Investigation. Its first round, the Tattoo Recognition Technology Challenge of 2014 to 2015, worked on a set of more than fifteen thousand images operationally collected by law enforcement from prisoners and arrestees; its second, the Tattoo Recognition Technology Evaluation reported in 2018, tested twelve algorithms against law-enforcement databases holding up to one hundred thousand images. The programme’s own materials described tattoos as suggesting affiliation to gangs, sub-cultures, religious or ritualistic beliefs, or political ideology, and it drew sustained criticism for that ambition; on the law-enforcement data the best-performing algorithm matched a tattoo to a separate image of the same tattoo at the first attempt in under seventy per cent of cases.17 India has no publicly documented equivalent, but the underlying infrastructure already exists. The Criminal Procedure (Identification) Act 2022 supplies the data-collection machinery, and any future extension toward algorithmic inference of belief would squarely engage the proportionality standard the Supreme Court set out in K.S. Puttaswamy v Union of India.18 Compelled display of a tattoo for identification does not offend Article 20(3), since it involves exhibiting an existing physical fact rather than communicating personal knowledge. That reasoning transfers poorly, though, to a tattoo used to prove affiliation, where the evidentiary use made of the mark looks much closer to compelled testimonial communication.19 For branding marks, the constitutional concern runs in the opposite direction entirely: failing to recognise their significance under-serves the Article 21 dignity interest the Supreme Court affirmed for marginalised complainants, including sex workers, in Budhadev Karmaskar v State of West Bengal.20 Put these three threads together and one conclusion follows: no single provision can sensibly govern all three uses of tattoo evidence. Identification calls for minimal constitutional scrutiny. Association calls for close scrutiny of compelled meaning. Exploitation calls not for restraint but for a more attentive state response altogether. A regulatory gap of this kind is not unique to India; it mirrors the broader difficulty every jurisdiction has faced in fitting emerging biometric and inference technologies into evidentiary frameworks written for an earlier generation of forensic tools, though the absence of any public Indian pilot programme means the constitutional question remains, for now, prospective rather than live.
The material canvassed above already draws on the United States, United Kingdom, Canada, and Australia. This part pulls that comparison together. Every jurisdiction surveyed has built some mechanism, a bad character gateway, a Mohan-style expert admissibility test, a codified probative-value provision, to stop a tattoo from standing as freestanding proof of criminal disposition. Branding marks are beginning to be treated as corroborative evidence in trafficking cases rather than as incidental detail. The Metropolitan Police’s Gangs Violence Matrix in the United Kingdom illustrates precisely the reliability and prejudice risk this paper raises: it drew sustained criticism from civil liberties organisations and from the Mayor of London’s own oversight bodies for disproportionately recording young Black men on the basis of vague and ill-defined markers, social media activity and music videos among them, with no clear evidentiary threshold before a name was added to the list.21 At the international investigative level, INTERPOL records tattoos and other personal identifiers in the analysis file it maintains on senior figures in organised criminal groups, intelligence contributed by participating member countries through their National Central Bureaus.22 Neither the composition of that image base nor any independently verified accuracy figure for it appears in INTERPOL’s public documentation, so the reliability question this paper raises about the American programme cannot even be tested against it. India’s absence from this comparative practice reflects nothing more principled than the youth of Sections 111 and 143 of the Bharatiya Nyaya Sanhita, both introduced only in 2023. The reform this paper proposes would bring Indian doctrine into line with, rather than ahead of, the practice already developing in the jurisdictions surveyed.
Courts should treat a tattoo match as corroborative, never conclusive. They should require an assessment of the design’s rarity within the relevant population, and demand comparison against a contemporaneous photograph rather than a verbal description alone, extending safeguards already recognised as necessary for test identification parades generally.
Indian courts should adopt a structured balancing inquiry for prosecutions under Section 111 of the Bharatiya Nyaya Sanhita. A tattoo should count as evidence of syndicate membership only where the prosecution establishes an evidentiary link to the specific offence charged, and supports it, where necessary, with qualified expert testimony on the tattoo’s meaning, modelled on the American, British, Canadian, and Australian approaches discussed above. Courts adopting this framework should also require the prosecution to disclose, before trial, the specific interpretive source, whether a law-enforcement gang database, a qualified expert, or documented ethnographic literature, on which any claimed meaning of the tattoo rests, so that the defence has a fair opportunity to test it.
The Ministry of Home Affairs should expand medical examination protocols under the Bharatiya Nagarik Suraksha Sanhita to incorporate forensic guidance on recognising branding tattoos. Investigators and medical examiners need training to distinguish a branding mark from an ordinary tattoo, and to record its features as corroborative evidence of trafficking rather than as incidental detail.
Any statute authorising automated tattoo-matching technology should require Indian investigative agencies to disclose the algorithm’s tested accuracy rate whenever they offer a match as evidence, so a fact-finder can weigh it with the scepticism it deserves. Disclosure of this kind mirrors the approach recommended for other emerging forensic technologies, such as probabilistic DNA-mixture software, where it has been urged that an algorithm’s error rate accompany any evidentiary use of its output.
Judicial academies and police training curricula should build in modules on the symbolic and cultural variability of tattoos and the clinical indicators of branding, so decision-makers can ask the reliability and prejudice questions this paper raises even without a binding appellate precedent forcing them to. A short, practical checklist would go a long way here, covering rarity assessment for identification matches, the evidentiary-link requirement for association evidence, and the clinical markers of a branding tattoo. Authorities could circulate it to trial judges and investigating officers at little cost, closing much of the gap this paper has identified without waiting on a legislative amendment or an appellate ruling.
A tattoo can identify a body, imply an association, or record an act of violence against the person who carries it. Indian evidence law currently treats all three as though they were the same kind of fact. They are not. Identification evidence asks only whether a mark and a body match. Association evidence asks a fact-finder to move from an image to a belief. Exploitation evidence asks the law to recognise coercion inscribed on skin by someone else. Treating these as one undifferentiated category under Section 7 of the Bharatiya Sakshya Adhiniyam risks two failures at once: courts too ready to accept a tattoo as proof of criminal character where the inference has not been earned, and courts too ready to overlook a tattoo’s significance as proof of victimisation where recognising it would take little more than attentive investigation. Indian law has already built careful safeguards around eyewitness identification. It has the doctrinal resources, and now a clear comparative template drawn from four other common law jurisdictions, to build equally careful safeguards around the evidence written on skin. What is missing is not legal capacity. It is recognition: that a tattoo is not one kind of evidence but three, each capable, if the law keeps treating them alike, of either convicting the wrong person or failing the person the law exists to protect, and each therefore deserving its own standard of proof.
*****
1. Electronic Frontier Foundation, Tattoo Recognition Research Threatens Free Speech and Privacy (June 2, 2016), https://www.eff.org/deeplinks/2016/06/tattoo-recognition-research-threatens-free-speech-and-privacy.
2. Bharatiya Sakshya Adhiniyam, 2023, § 7 (India); Indian Evidence Act, 1872, § 9 (India) (repealed).
3. Fed. R. Evid. 403; Securing Evidence: The Key to Convicting Human Traffickers, UN News (Mar. 5, 2022), https://news.un.org/en/story/2022/03/1112712.
4. National Institute of Standards and Technology, Tattoo Recognition Technology - Evaluation (Tatt-E): Performance of Tattoo Identification Algorithms, NISTIR 8232 (2018), https://nvlpubs.nist.gov/nistpubs/ir/2018/NIST.IR.8232.pdf; Electronic Frontier Foundation, Federal Researchers Complete Second Round of Problematic Tattoo Recognition Experiments (Nov. 13, 2018), https://www.eff.org/deeplinks/2018/11/federal-researchers-complete-second-round-problematic-tattoo-recognition.
5. United States v. Thomas, 321 F.3d 627 (7th Cir. 2003); United States v. Newsom, 452 F.3d 593 (6th Cir. 2006); United States v. Ford, 761 F.3d 641 (6th Cir. 2014); People v. Bynum, 496 Mich. 610, 852 N.W.2d 570 (2014); R. v. Mohan, [1994] 2 S.C.R. 9 (Can.).
6. Securing Evidence: The Key to Convicting Human Traffickers, UN News, supra note 3.
7. Indian Evidence Act, 1872, § 9 (India) (repealed); Bharatiya Sakshya Adhiniyam, 2023, § 7 (India).
8. Criminal Procedure (Identification) Act, 2022, §§ 2(1)(b), 4(2) (India) (record of measurements to be kept by the National Crime Records Bureau for seventy-five years); Criminal Procedure (Identification) Rules, 2022 (India); Identification of Prisoners Act, 1920, §§ 3-5 (India) (repealed).
9. Bharatiya Nagarik Suraksha Sanhita, 2023, § 54 (India); Ramkishan Mithanlal Sharma v. State of Bombay, AIR 1955 SC 104; Ramanathan v. State of Tamil Nadu, AIR 1978 SC 1204; Sheikh Hasib alias Tabarak v. State of Bihar, AIR 1972 SC 283; State of Andhra Pradesh v. K. Venkata Reddy, AIR 1976 SC 2207, (1976) 3 SCC 454.
10. Joginder alias Danny v. State (NCT of Delhi), 2018 Indlaw DEL 4818 (Delhi HC); Manjunatha v. State of Karnataka, 2019 Indlaw KAR 11544 (Karnataka HC).
11. United States v. Thomas, 321 F.3d 627, 630-33 (7th Cir. 2003); United States v. Newsom, 452 F.3d 593, 602-04 (6th Cir. 2006); United States v. Ford, 761 F.3d 641, 649-50 (6th Cir. 2014); People v. Bynum, 496 Mich. 610, 626-30 (2014).
12. Criminal Justice Act, 2003, c. 44, §§ 98-113 (UK); Police and Criminal Evidence Act, 1984, c. 60, § 78 (UK).
13. R. v. Mohan, [1994] 2 S.C.R. 9 (Can.); Evidence Act 1995 (Cth) §§ 135, 137 (Austl.).
14. Bharatiya Nyaya Sanhita, 2023, § 111 (India).
15. Securing Evidence: The Key to Convicting Human Traffickers, UN News, supra note 3; Directive 2011/36/EU of the European Parliament and of the Council, art. 4(2), 2011 O.J. (L 101) 1 (EU).
16. Bharatiya Nyaya Sanhita, 2023, § 143 (India).
17. National Institute of Standards and Technology, Tattoo Recognition Technology - Challenge (Tatt-C): Outcomes and Recommendations, NISTIR 8078 (2015), https://nvlpubs.nist.gov/nistpubs/ir/2015/NIST.IR.8078.pdf; National Institute of Standards and Technology, Tatt-E, supra note 4; Electronic Frontier Foundation, Tattoo Recognition Research Threatens Free Speech and Privacy, supra note 1; Electronic Frontier Foundation, Federal Researchers Complete Second Round of Problematic Tattoo Recognition Experiments, supra note 4.
18. Criminal Procedure (Identification) Act, 2022, §§ 2(1)(b), 4 (India); K.S. Puttaswamy v. Union of India, (2017) 10 SCC 1.
19. State of Bombay v. Kathi Kalu Oghad, AIR 1961 SC 1808 (holding that compelling an accused to furnish material for comparison or identification, as distinct from imparting personal knowledge, does not amount to testimonial compulsion under Article 20(3)).
20. Budhadev Karmaskar v. State of West Bengal, Crim. App. No. 135 of 2010, 2022 SCC OnLine SC 704 (India) (order dated May 19, 2022).
21. Amnesty International, Trapped in the Matrix: Secrecy, Stigma and Bias in the Met’s Gangs Database (May 2018), https://media.amnesty.org.uk/documents/Trapped20in20the20Matrix20Amnesty20report.pdf; Mayor’s Office for Policing and Crime, Review of the Metropolitan Police Service Gangs Matrix (Dec. 2018), https://www.london.gov.uk/sites/default/files/gangs_matrix_review_-_final.pdf.
22. Interpol, Project Millennium, https://www.interpol.int/en/Crimes/Organized-crime/Projects/Project-Millennium.