The Transgender Persons Amendment Act, 2026: Six constitutional vulnerabilities

The Transgender Persons (Protection of Rights) Amendment Act, 2026 received Presidential assent on March 31, 2026. Constitutional challenges are already forming. The Act, which replaces self-perceived gender identity with a medicalised certification process, has attracted attention primarily for its conflict with NALSA v. Union of India and the retrospectivity of its definitional changes.

Those arguments are real and will form part of any challenge. But the Act presents several additional constitutional vulnerabilities that have not received adequate analytical attention. This piece highlights six of them.

The amended definition of "transgender person" under the Act states that the term shall not include, "nor shall ever have been so included," persons who self-identify their gender. This is retrospective legislation: not merely prospective redefinition but a declaration that past recognition was always legally void.

A preliminary clarification is necessary. Parliament can enact retrospective legislation and retrospectivity alone is not constitutionally fatal. The question is whether this particular retrospective measure survives scrutiny. It does not, for two distinct reasons.

First, the manifest arbitrariness standard. In Shayara Bano v. Union of India, the Supreme Court held that legislation enacted capriciously, without adequate determining principle, or in a manner that produces disproportionate consequences, violates Article 14. Over 32,000 persons obtained gender recognition certificates under the 2019 Act through a State-supervised process. The Act now declares those certificates were always invalid. There is no transitional provision. No compensation for reliance. No grace period. No mechanism to challenge the withdrawal of a status on which individuals structured employment, healthcare access and legal identity. Retrospective invalidation of State-issued documents, affecting a vulnerable class, is a textbook case of manifest arbitrariness.

Second, legitimate expectation. The doctrine, recognised across Indian administrative and constitutional law, holds that where the State creates an expectation through its own conduct or representations, it cannot withdraw that expectation without reasonable notice and opportunity. The State examined applicants, issued certificates and recorded legal identities. The Puttaswamy framework requires that any intrusion into a fundamental right be proportionate to its aim. Retrospectively erasing 32,000 identities to achieve "administrative clarity" fails proportionality on any reasonable reading.

Furthermore, this sweeping erasure offends the doctrine of vested rights. When individuals acquired legal recognition under the 2019 framework, they were conferred vested rights governing their employment protections, property ownership and civil identity. Stripping away these rights retrospectively, even as a compelling and overwhelming public interest is absent, renders the amendment highly vulnerable.

The Act reintroduces the word "eunuch" into the statutory definition of transgender persons. This is not a neutral taxonomic choice. In NALSA v. Union of India, the Supreme Court explicitly traced the criminalisation of transgender identities in India to the Criminal Tribes Act, 1871, describing it as "brutal legislation with a vicious and savage mindset" that treated hijras and eunuchs as born criminals. The Court held that the constitutional order established in 1950 categorically repudiated that colonial framework.

Parliament has now reinstated the precise terminology that NALSA condemned. The argument under Article 14 is straightforward: a classification that uses vocabulary carrying active criminal stigma lacks the rational basis that Article 14 demands. The question, thus, is whether the reintroduction of "eunuch" into a statute purporting to protect transgender persons can be characterised as anything other than legislative indifference to the constitutional history the Court took such pains to establish.

There is also a dignity argument under Article 21. In Francis Coralie Mullin v. Administrator, Union Territory of Delhi, the Court held that the right to life includes the right to live with basic human dignity. Using a term historically associated with forced criminalisation of a community in a statute designed for their protection, engages that right in a way that cannot be resolved by legislative intent alone.

Article 15(1) prohibits discrimination by the State on grounds of, among others, sex. In Navtej Singh Johar v. Union of India, a five-judge Constitution Bench held that the expression "sex" in Articles 15 and 16 "is not just limited to biological sex of male or female, but intended to include people who consider themselves to be neither male nor female." Gender identity is discrimination on the ground of sex, as that term has been constitutionally expanded.

The Amendment Act removes self-perceived gender identity as a basis for legal recognition. It denies a defined class, those whose gender identity does not conform to birth assignment and who decline or cannot access medical certification, the legal status that others with conforming identities enjoy as a matter of course. That differential treatment, applied to a class defined precisely by their gender identity, is discrimination on the ground of sex within the expanded Navtej reading.

The Article 15 ground has been underused in commentary on the Act. It deserves prominence in any petition. Unlike Article 21, which requires proportionality analysis and can be qualified by procedure established by law, discrimination on the grounds listed in Article 15(1) is a more categorical prohibition. The State bears a higher burden when Article 15 is engaged.

The amended definition creates a classification that cannot survive rational basis scrutiny. On one hand, the Act excludes persons with self-perceived gender identity. On the other, it includes within the definition "a person who is forced to assume a transgender identity." A person coerced into an identity they did not choose is included; a person who has lived their identity authentically for decades is excluded. The classification lacks any intelligible differentia connecting it to the legislative object.

The vagueness problem compounds this. The Act criminalises "allurement" and "inducement" into assuming a transgender identity. These terms are undefined. In a community where most young transgender persons lose their biological families upon disclosure, the chosen family structures of Hijra guru-chela relationships, the primary social and economic support network available to many, become legally precarious. The guru who welcomes a homeless transgender adolescent into their household is potentially an "inducement." The activist who educates a young person about trans identity may be an "allurement."

This ambiguity also creates a chilling effect on speech under Article 19(1)(a). If basic education, advocacy, or community support can be construed as ‘inducement,’ the provision risks silencing precisely the forms of expression on which transgender communities depend.

The void for vagueness doctrine, applied by Indian courts under Article 14 and Article 19, holds that a penal provision so vague that it does not give fair notice of what conduct it prohibits, is unconstitutional. The Act provides no definition of "allurement" or "inducement" in the context of gender identity. Given the distinct social ecology of transgender communities in India, where chosen family is not a preference but a survival structure, vagueness here is not merely a drafting defect. It is a constitutional defect.

The Act requires a District Screening Committee, constituted substantially of medical professionals, to issue or recommend gender recognition certificates. This requirement applies uniformly to the categories listed in the definition, including, alongside medical and physical categories, the explicitly socio-cultural identities of kinner, hijra, aravani and jogta.

The constitutional difficulty here is structural. How does a medical board verify a socio-cultural identity? The hijra identity is not a diagnosis. It is a community membership, defined by initiation, lived experience and cultural belonging, none of which are accessible to clinical examination. A provision that mandates medical verification as the precondition for recognising an identity that is by definition non-medical is arbitrary in its application to that class. It creates a procedure incapable of achieving its stated purpose for a defined subset of the very persons it purports to cover.

Under Article 14, a law that creates a classification and then applies a mechanism to that classification that is structurally incapable of serving the classification's own stated criteria lacks the rational nexus that equal protection demands. Petitioners representing hijra and kinner communities have a particularly strong Article 14 argument on this ground, one that does not depend on the broader self-identification debate and may succeed independently.

Section 7 of the Act, which has received almost no analytical attention, requires that medical institutions where gender-affirming surgery is performed must furnish information about those surgeries to the district magistrate. The provision creates, in effect, a compulsory State registry of persons who have undergone gender-affirming procedures.

This engages KS Puttaswamy v. Union of India directly. The nine-judge bench held that informational privacy, including the right to control personal medical information, is a component of the fundamental right to privacy under Article 21. Medical procedures undergone by an individual are among the most intimate categories of personal information. Mandatory disclosure to a district magistrate, without the individual's consent, without a specified purpose, without any data protection safeguard and without any provision limiting the use or further disclosure of that information, fails each element of the Puttaswamy test: legality, legitimate aim, proportionality and procedural safeguard.

Section 7 is also constitutionally unusual in that it operates against a third party - the medical institution - to compel disclosure of another person's private medical information. The individual whose information is disclosed has no voice in the provision and no remedy within it. That is a structural due process problem under Article 21 that stands entirely apart from the broader definitional challenges to the Act.

The six grounds mapped above are not exhaustive. The conflict with NALSA and the October 2025 directions in Jane Kaushik v. Union of India will form the spine of any petition. The procedural failures, the absence of meaningful consultation with the National Council for Transgender Persons and the speed of passage will be relevant to questions of legislative intent.

What this analysis suggests is that the stronger challenge is not a single-ground petition but a layered one. The retrospectivity argument challenges the definitional heart of the Act. The Article 15 argument challenges the classification itself. The eunuch provision and the medical board incoherence challenge the internal logic of the definition. The vagueness argument challenges the penal provisions. And Section 7 stands as an independent privacy violation that can be challenged regardless of what happens to the rest.

Each of these grounds has independent constitutional standing. Together, they present a court with a statute that fails at multiple levels of constitutional analysis simultaneously.

Paras Sharma is a practising advocate at the Punjab and Haryana High Court, Chandigarh.

Views are personal.