Ask a room of lawyers to name the law that protects women in India and the answers will arrive as a list of statutes, offered as though the list itself were the answer. It is not. Women’s rights laws in India are not one body of law but four, and they operate on different logics that produce different proceedings, different standards of proof, different forums and different kinds of relief. One family of statutes works by criminal prohibition: it defines conduct as an offence and sets the machinery of police, prosecution and punishment in motion. A second family works by civil remedy: it lets the person harmed apply to a court for orders that protect her, house her, maintain her and compensate her, without sending anyone to prison. A third family works by workplace obligation: it places duties on employers and institutions and enforces them through committees, inspectors and claims authorities rather than through the criminal courts. A fourth family works by welfare entitlement: it promises money, food and support as a matter of entitlement, delivered through schemes and anganwadis rather than through litigation at all.

The practical consequence of this architecture is that the first question in any real situation is never which statute is strongest. The first question is which logic governs the problem in front of the person asking, because the choice of route determines almost everything that follows: who must prove what, to what standard, before which forum, within what time, and what the forum can actually order at the end. A woman subjected to violence in her home may have three routes open at once, under three different statutes, each with a different object. The criminal route seeks the punishment of the person who harmed her. The civil route seeks her safety, her residence and her maintenance. The two are constantly confused with each other, in conversation and sometimes in print, and the confusion costs people the relief they were entitled to pursue. This article maps the whole field at the level of its structure: the constitutional provisions that make protective legislation possible, the four families of statute in the order in which Parliament built them, the institutions that stand behind them, the pattern of the case law that has shaped them, and the honest account of where the field fails the people it was written for.

One claim organises everything that follows, and it is worth stating at the outset because it explains both the reach of this field and its persistent disappointments. India protects women mainly by prosecution. Three of the four families of legislation operate through, or in the shadow of, the criminal law, and the one significant civil statute in the field arrived only in 2005, forty four years after the dowry statute that began the modern sequence. The default Indian response to harm against a woman has been to make the harm an offence, which means the remedy ordinarily on offer is the punishment of the perpetrator rather than relief for the person harmed. That ordering was not an accident. It reflects what the Indian state has historically been better at supplying, which is prohibitions, and what it has been slower to build, which is the civil and welfare machinery that delivers relief without requiring a prosecution to succeed first.

Why does the choice of legal route matter more than the choice of statute?

The choice of route matters because each route carries its own standard of proof, its own forum and its own ceiling of relief, and those three features decide outcomes long before the merits are argued. A criminal prosecution must be proved beyond reasonable doubt and ends, at best, in a conviction that gives the complainant nothing directly. A civil application under the domestic violence statute is decided on a preponderance of probabilities and can produce residence, protection and monetary orders within weeks. A workplace complaint goes to a committee that can recommend compensation but cannot imprison. A welfare entitlement requires no proof of wrongdoing at all, only proof of eligibility. The statute book offers all four; the situation in front of the reader admits only some of them, and choosing wrongly at the start is the most common reason a strong grievance produces no relief.

The sections that follow take the field in the order a reader needs it. The constitutional foundation comes first, because every statute in this field leans on it and because one clause of the equality article is what makes the entire protective edifice constitutionally secure. The four families follow in their historical sequence, which is also the sequence that explains their shape: criminal prohibition first and dominant, civil remedy late and singular, workplace obligation arriving through the courts before it arrived through Parliament, and welfare entitlement newest and thinnest. The institutional layer and the judicial pattern come next, then the enforcement record and the design argument it supports, and finally a navigation section with the decision table that routes five recurring situations to their statutes, forums and orders.

The scope of the map should be stated, since a field this size invites both overclaim and omission. The map covers the general law of the Union as it bears on women’s rights: the constitutional provisions, the penal and civil statutes of general application, the workplace and welfare instruments, the personal law interface where it meets them, and the institutions charged with the field. It does not attempt the statute book of any single state, it does not treat the special position of children, which a dedicated protection code governs and companion articles in this series address, and it does not follow the field into its specialised annexes, education, health and political finance among them, except where a provision’s logic belongs to one of the four families. Within that scope the map aims at completeness of structure rather than completeness of citation: every family is fully drawn, every major statute placed, every institution located, and the decided cases chosen for the turns they mark rather than for the volume of the reports they fill. A reader who finishes it should be able to place any provision encountered afterwards, in a judgment, a news report or a Bill, within the architecture without assistance, and to know which questions to ask of it. That, and not the recitation of the field, is the standard the following sections set themselves.

A word about the shelf itself, because readers searching for this field under its own name will not find it. There is no shelf. The statutes mapped in this article live in the penal codes, in the labour codes, in a food security statute, in the succession and adoption codes of one community, in a workplace Act administered by a ministry of women and child development and enforced by employers, and in a civil statute administered by magistrates. Libraries shelve them by subject of a wholly different order: crime, labour, food, family. The unity of the field is not bibliographic but functional, supplied by the single social fact the statutes all answer, and that is why a thematic map has to be written rather than merely compiled: the connections, the overlaps, the sediment of one family’s failure producing the next family’s statute, are invisible from inside any one code. It is also why the field defeats the casual enquirer in a specific way. Asked which law protects women in India, the honest answer is a routing exercise, and a legal system whose protection requires routing is a system whose accessibility depends on advice, which is to say on money, geography and luck. The fragmentation is not a cosmetic flaw in an otherwise sound architecture. It is one of the architecture’s load bearing characteristics, and every section that follows should be read with it in mind.

The long prehistory: reform statutes before the Constitution

The four families have a prehistory that explains their instincts, because the colonial state legislated about women’s lives for more than a century before the Republic did, and it legislated in a recognisable pattern: a practice would be identified as intolerable, a prohibition issued against it, and the deeper questions of property, marriage and maintenance left largely where custom had put them. The Bengal Sati Regulation of 1829 declared the practice of sati illegal and punishable, and it is the ancestor of every prohibition in this field, including the Act of the Republic that had to be passed on the same subject a hundred and fifty eight years later. The Hindu Widows’ Remarriage Act, 1856, Act XV of 1856, the product of Ishwar Chandra Vidyasagar’s campaign, legalised the remarriage of Hindu widows and the legitimacy of its issue, while forfeiting the widow’s interest in her deceased husband’s property on remarriage, a qualification that shows the era’s method exactly: reform the status, concede the economics. The Age of Consent Act, 1891, Act X of 1891, raised the age below which intercourse with a wife was rape, and the Child Marriage Restraint Act, 1929, Act XIX of 1929, the Sarda Act, penalised the solemnisation of child marriages while leaving the marriages themselves valid, a design weakness its successor statute inherited and the field has never fully repaired.

Property moved later and slower. The Hindu Women’s Rights to Property Act, 1937, Act XVIII of 1937, gave a Hindu widow a limited interest in her husband’s property, the so called widow’s estate, without making her a full owner or a coparcener; the Married Women’s Property Act, 1874, had earlier protected a married woman’s earnings and insurance monies in defined respects, chiefly within the communities it reached. The pattern across the colonial century is the one the modern field would inherit and spend decades correcting. Conduct could be prohibited, sometimes bravely and against strong opposition. Status could be adjusted at the margins. But the economic constitution of the household, who owns the ancestral property, who may demand partition, whose earnings are her own, was the last thing the legislature touched, and the Republic’s first great legislative battle was fought on exactly that ground.

That battle was the Hindu code project. Between 1948 and 1951, Dr. B. R. Ambedkar, as Law Minister, piloted a consolidated Hindu Code Bill that would have recast marriage, divorce, adoption, maintenance and succession together, giving daughters a share in coparcenary property and wives rights the reformers of the previous century had not attempted. The Bill met sustained resistance inside and outside Parliament, was not carried, and Ambedkar resigned from the cabinet in 1951 with the Bill’s failure among his stated reasons. Parliament then enacted the code in pieces during the middle nineteen fifties: the Hindu Marriage Act, 1955, Act 25 of 1955, which introduced divorce and judicial separation into Hindu matrimonial law; the Hindu Succession Act, 1956, Act 30 of 1956, which made women full owners of the property they held, converting the limited estate into absolute ownership, but left the Mitakshara coparcenary itself male, the omission the amendment of 2005 would correct forty nine years later; the Hindu Minority and Guardianship Act, 1956, Act 32 of 1956; and the Hindu Adoptions and Maintenance Act, 1956, Act 78 of 1956, which recognised a wife’s right to maintenance and a married woman’s right to adopt in defined circumstances. The episode matters to this map for two reasons. It shows that the hardest questions in this field have always been the economic ones, property and maintenance, rather than the prohibitory ones. And it shows why the field is fragmented: India has never had its one consolidating statute for women’s rights, because the one attempt at consolidation, in the early nineteen fifties, failed, and legislation has proceeded in pieces ever since.

The constitutional foundation beneath women’s rights laws in India

Every protective statute in this field is written in the shadow of the equality code of the Constitution, and the relationship between the two is the opposite of what a first reading suggests. Article 14 guarantees equality before the law and the equal protection of the laws. Article 15(1) prohibits the state from discriminating against any citizen on grounds only of religion, race, caste, sex, place of birth or any of them. Article 16 guarantees equality of opportunity in public employment and prohibits discrimination on the same grounds in state employment. Read alone, these provisions might appear to threaten any statute that singles women out for special treatment, since a law that protects women and not men classifies on the ground of sex. The Constitution forecloses that objection in terms. Article 15(3) provides that nothing in Article 15 shall prevent the state from making any special provision for women and children. That single clause is the constitutional security of the entire field: it converts what would otherwise be suspect classification into expressly permitted legislation, and it is the provision courts cite when a protective statute is challenged as discriminatory against men. The clause is enabling rather than mandatory. It permits special provisions; it does not command them, and a legislature that enacts nothing for women breaches no duty under Article 15(3) itself.

The foundation has other layers that do quieter work. Article 23 prohibits traffic in human beings and forced labour, and it is the textual anchor for the legislation on trafficking in women. The directive principles of state policy supply the programme that several statutes expressly implement. Article 39(a) directs the state to secure for citizens, men and women equally, the right to an adequate means of livelihood. Article 39(d) directs it to secure equal pay for equal work for both men and women, and the equal remuneration statute of 1976 was enacted to give that direction legal force. Article 39(e) directs that the health and strength of workers, men and women, and the tender age of children are not abused. Article 42 directs the state to make provision for securing just and humane conditions of work and for maternity relief, and it is the constitutional root of the maternity benefit legislation. The fundamental duties add a citizen-facing counterpart: Article 51A(e) makes it the duty of every citizen to renounce practices derogatory to the dignity of women. Directive principles and fundamental duties are not enforceable by a court, and an article of this kind should not pretend otherwise. Their function in this field is different. They supply the stated purpose that courts use when they construe a protective statute generously, and they mark the subjects on which legislation, when it comes, is understood to be implementing the Constitution rather than merely exercising a power.

Reservation completes the constitutional layer, and it operates through amendment rather than through the original text. The Constitution (Seventy-third Amendment) Act, 1992 and the Constitution (Seventy-fourth Amendment) Act, 1992 inserted Articles 243D and 243T, which reserve not less than one third of seats, and one third of the offices of chairperson, for women in panchayats and municipalities, with rotation among constituencies. Many states have since raised the proportion to one half by their own legislation, which the constitutional text permits because one third is a floor rather than a ceiling. The result is the largest single body of elected women representatives anywhere in the world, produced not by any statute in the four families but by the constitutional text itself. The Constitution (One Hundred and Sixth Amendment) Act, 2023, the Nari Shakti Vandan Adhiniyam, extended the same technique upward: it inserted Articles 330A and 332A to reserve, as nearly as may be, one third of seats for women in the Lok Sabha and in the legislative assemblies of the states, including within the seats already reserved for the Scheduled Castes and the Scheduled Tribes. Its operation is deliberately deferred. The reservation takes effect only after a census has been published and a delimitation exercise undertaken on its basis, so the provision sits in the Constitution as an enacted promise whose commencement is tied to two further public acts. That structure matters for a reader of this field, because it is a reminder that even at the constitutional level, the enactment of a protection and the arrival of its benefit can be separated by years and by machinery that has nothing to do with the text.

Where does the Constitution speak about women outside the equality articles?

Outside the equality articles, the Constitution addresses women in three distinct registers. The directive principles instruct the state to secure equal pay for equal work, to protect the health and strength of women workers, and to make provision for maternity relief. The fundamental duties instruct citizens to renounce practices derogatory to the dignity of women. And the reservation provisions inserted by amendment place women in elected office, first in local bodies and, by the amendment of 2023, prospectively in Parliament and the state assemblies. A fourth register is structural rather than textual: the prohibition on traffic in human beings in Article 23, which has anchored the law on trafficking since 1956.

One further feature of the foundation deserves attention because it explains a recurring confusion in the literature of this field. The equality articles prohibit discrimination, and the protective statutes classify. Litigation over that tension is older than most of the statutes: employers and accused persons have repeatedly argued that a provision favouring women offends Article 15(1) or the parallel guarantee in employment. The courts have answered through Article 15(3) for provisions traceable to it, and through the doctrine that a reasonable classification founded on an intelligible differentia survives Article 14 where a naked preference would not. The practical consequence is that the constitutional debate in this field is rarely about whether protective legislation may exist. It is about how far a particular benefit or burden may go, which is a question answered provision by provision. A reader who understands that Article 15(3) exists will not be misled by the common claim that every women-specific law is constitutionally fragile. It is not. The fragility, where it exists, lies in drafting and in administration, not in the constitutional warrant.

The case law on the equality code’s enabling provisions shows the width of the permission in practice. When the reservation of posts for women in state services was challenged as discrimination against men, the Supreme Court in Government of Andhra Pradesh v. P. B. Vijayakumar, in 1995, sustained the reservation by reading Article 15(3) together with Article 16, treating the special provision for women as an affirmative facet of equality rather than an exception grudgingly carved from it, and remarking on the constitutional commitment to women’s participation in public employment that the directive principles declare. The reasoning matters beyond its facts because it fixed the interpretive posture the courts bring to this field: a special provision for women is not a derogation from equality to be narrowly construed, but an application of it, to be read with the grain of the Constitution’s stated purposes. Combined with the reservation architecture in local government, under which women have held elected office in the panchayats and municipalities in very large numbers since 1993, the equality code’s practical footprint in this field is considerably larger than its text, read cold, would suggest, and it explains why constitutional litigation in this area so rarely proceeds as a challenge to the existence of protective legislation and so often as a dispute about its reach, its beneficiaries and its omissions.

One further provision of the equality code repays attention because it is the foundation stone of a live national argument. Article 44, among the directive principles, records that the state shall endeavour to secure for the citizens a uniform civil code throughout the territory of India. For this field, a uniform code is the standing alternative to the personal law settlement described throughout this article: it would replace the community specific rules of marriage, divorce, succession and guardianship with one civil law for all citizens, and its supporters argue from the succession and maintenance disparities the personal law system produces, while its critics argue from minority rights, from the diversity of the family systems a single code would have to flatten, and from the suspicion that uniformity in drafting has often meant the majority’s customs generalised. The courts have periodically commended the aspiration to Parliament while refusing to enact it judicially. Goa administers a civil code of Portuguese origin that is commonly cited in the debate, and Uttarakhand enacted a uniform civil code of its own in 2024, the first state legislation of its kind, whose working will be evidence in the national argument. The map records the debate as unresolved and the provision as a directive principle: a direction to the state, not a right a citizen can enforce, and so a promise of the constitutional layer’s fourth kind, distinct from the permission of Article 15(3), the prohibition of Article 23 and the entitlement language of the maternity provisions.

Where does the constitutional layer stop?

At the door of personal law, by a doctrine the courts built early and have never fully dismantled. In State of Bombay v. Narasu Appa Mali, decided by the Bombay High Court in 1951, the court held that personal laws are not laws in force within the meaning of Article 13 of the Constitution and so are not void for inconsistency with fundamental rights, a reasoning that insulated the personal law systems from the equality challenge that the Constitution’s text appeared to invite. The consequence shaped the field’s development: reform of the rules governing marriage, divorce, succession and guardianship within each community proceeded legislatively and unevenly rather than constitutionally and at a stroke, community by community, which is why the succession reform of 2005 amended a code that applies to Hindus, Buddhists, Jains and Sikhs and left every other community’s law where it stood. Later benches have chipped at the doctrine’s edges, testing specific practices against dignity and arbitrariness rather than against the personal law system as a whole, as the triple talaq decision of 2017 did, but the settlement remains the one the field works with: the protective statutes of general application can be, and are, tested and read against the Constitution freely, while the personal law substrate beneath them moves by legislation, at the pace each community’s politics allows.

The first family: criminal prohibition and the dominance of the offence

The modern sequence begins with a statute most readers could not name and a practice it failed to extinguish. The Dowry Prohibition Act, 1961, Act 28 of 1961, made the giving and taking of dowry an offence and supplied the template the field would follow for decades: identify a harm embedded in a social practice, attach a penalty to it, and rely on the criminal process to change behaviour that families, communities and markets sustained. The statute’s own history is told in the pillar article on the subject, the Dowry Prohibition Act 1961 explained, and the detail belongs there. What matters for the map is the pattern the Act set. The offence was defined around a transaction that both sides to a marriage often had reasons to conceal, the penalties proved light against the size of the transfers involved, and the statute spent its first decades as a provision more cited than convicted under. Parliament returned to it repeatedly, strengthening it by amendment in 1984 and again in 1986, and each return confirmed the template rather than questioning it: when a prohibition underperforms, the legislative instinct in this field has been to raise the penalty and shift the burden, not to change the instrument.

The general criminal code was recruited into the same project in stages, and the stages are the skeleton of the field’s history. In 1983, the Criminal Law (Second Amendment) Act, 1983, Act 46 of 1983, inserted Section 498A into the Indian Penal Code, making cruelty by a husband or his relatives towards a woman an offence in its own right, and adjusted the evidence law so that a court could presume abetment of suicide in defined circumstances of cruelty. In 1986, the same amending exercise that strengthened the dowry statute inserted Section 304B into the Penal Code, creating the offence of dowry death and attaching to it a presumption in the Evidence Act where a woman died in unnatural circumstances within seven years of marriage after cruelty or harassment in connection with a dowry demand. The architecture is instructive. Rather than rely on the ordinary offences of hurt, homicide and abetment, Parliament built bespoke offences around the marital home, because the ordinary offences had proved unable to reach conduct that occurred in private, within families, against a person whose witnesses were usually the accused’s own household. The bespoke offences carried procedural reinforcements: cognizability, restrictions on bail, presumptions and, in the case of cruelty, a definition wide enough to cover conduct that drives a woman to suicide and conduct likely to cause grave injury to life, limb or health, alongside harassment aimed at coercing a property demand.

The third and largest recruitment of the criminal code came a generation later. The Criminal Law (Amendment) Act, 2013, Act 13 of 2013, followed the gang rape and death of a young woman in Delhi in December 2012 and the report of the committee headed by Justice J. S. Verma, a former Chief Justice of India, which Parliament received in January 2013 and legislated on within months. The amendment rewrote the sexual offences in the Penal Code, expanded their definitions, created offences of stalking, voyeurism and acid attack, raised punishments across the chapter, and adjusted procedure and evidence for trials of sexual offences. Delivered Article 28 of this series owns the detail of that amendment, and the map needs only its structural lesson: the response to a catalytic crime was again an expansion of the criminal family, broader offences and heavier sentences, rather than an expansion of the civil remedies available to the person harmed. The Criminal Law (Amendment) Act, 2018, Act 22 of 2018, repeated the movement after another catalytic case, prescribing capital punishment for the rape of a child below twelve years and raising minimum sentences across the sexual offences against children and women. In 2023 the Penal Code itself was replaced by the Bharatiya Nyaya Sanhita, 2023, Act 45 of 2023, which carried the architecture across with renumbering: the cruelty provision sits in Section 85 of the new code, the dowry death provision in Section 80, and the sexual offences in the code’s chapter on offences against women and children. The codes changed the shelf on which the offences sit. They did not change the family’s logic, which remains prohibition enforced by prosecution.

Around the code sits a ring of special statutes, each a prohibition aimed at a specific practice, and their dates tell the field’s story as clearly as their contents. The Immoral Traffic (Prevention) Act began life in 1956 as the Suppression of Immoral Traffic in Women and Girls Act, Act 104 of 1956, enacted to give effect to the international convention signed at New York in 1950, and it was renamed and recast by the amending Act of 1986, Act 46 of 1986, which is why a statute of 1956 carries a title that readers often misdate to the later year. The Indecent Representation of Women (Prohibition) Act, 1986, Act 60 of 1986, prohibits the depiction of women in advertisements and publications in a manner that is indecent or derogatory, and belongs to the family’s periphery: a prohibition enforced by prosecution, directed at representation rather than at violence. The Commission of Sati (Prevention) Act illustrates the drafting trap this field lays for the careless: it is called the Act of 1987 because that is the year in its short title and the year of the events at Deorala in Rajasthan that produced it, but as an Act of Parliament it is Act 3 of 1988, having received assent in January of that year; an account that numbers it as a 1987 Act has confused the title with the enactment. The Pre-Conception and Pre-Natal Diagnostic Techniques (Prohibition of Sex Selection) Act, 1994, Act 57 of 1994, prohibits sex selection and regulates diagnostic techniques, having begun as a statute addressed to pre-natal diagnosis and been retitled and widened by the amendment of 2003 to reach selection before conception as well. It polices clinics, laboratories and practitioners through registration, prohibition and penalty, and its enforcement problem is the family’s enforcement problem in miniature: the conduct is consensual between the parties present, the complainant is absent by definition, and detection depends on the record-keeping the statute mandates rather than on a victim’s report.

The punishment question runs through the family’s modern history as a debate in its own right, and the map should record its terms because the two sides invoke different theories of what this legislation is for. The committee headed by Justice J. S. Verma, whose report drove the amendment of 2013, recommended against the death penalty for rape, reasoning from the evidence available to it that capital punishment had not been shown to deter and that the certainty and speed of punishment matter more than its severity; Parliament accepted most of the report and declined that recommendation in part, prescribing death where the offence caused death or a persistent vegetative state. In 2018 Parliament went further and introduced capital punishment for the rape of a child below twelve years. The abolitionist position in this debate is not softness about the offence: it holds that a capital charge raises the prosecution’s practical burden, lengthens trials, hardens the accused’s incentive to eliminate the only witness, and prices the complainant’s testimony at a level that families and courts hesitate to exact, particularly where the accused is a relative, as the registration data show he very often is. The retentionist position holds that some offences exhaust the community’s vocabulary of condemnation short of death, that the sentence expresses the value the law sets on the victim, and that the deterrent argument demands a proof of a negative that no sentencing policy could supply. Successive governments have sided with retention and extension. The map records the debate because it explains the family’s trajectory better than any other single fact: when the criminal family is asked to respond to horror, the variable Parliament can most visibly move is the maximum sentence, and so the maximum sentence is what moves, while the conviction machinery beneath it, investigation, forensic capacity, trial speed, changes at the pace of budgets rather than at the pace of amendments.

The presumptions deserve a paragraph of their own, because they are the field’s most distinctive legal device and its most contested one. Indian evidence law ordinarily places the burden of proof on the prosecution throughout. The protective amendments of the nineteen eighties reversed that burden at defined points. Section 113A of the Indian Evidence Act, 1872 permitted a court to presume that a husband or his relative abetted a woman’s suicide where she died by suicide within seven years of marriage and cruelty within the meaning of the penal provision was shown. Section 113B required a court to presume that a person had caused a dowry death where the statutory ingredients were established: the death in unnatural circumstances within seven years, and cruelty or harassment soon before her death in connection with a dowry demand. The dowry statute added a burden provision of its own for the offence of taking or abetting the taking of dowry. Both presumptions have been carried into the Bharatiya Sakshya Adhiniyam, 2023, the successor evidence code. The constitutional defence of the device is that it responds to an evidentiary reality the legislature was entitled to recognise: these events occur inside homes, among witnesses aligned with the accused, and an unmodified burden had produced a pattern of acquittals that nullified the offences. The criticism is the mirror image: a presumption of this kind asks the accused to explain a death, and the safeguard built into the drafting, that the prosecution must first prove the foundational facts, including cruelty soon before death, is only as strong as the courts’ insistence on it. The case law has in fact insisted, repeatedly, that soon before is a real condition and not a formula, and the oscillation between the presumption’s text and its judicial disciplining is the most technical expression of the field’s central tension between protection and proof.

The trafficking statute exposes a different fault line within the family. The Immoral Traffic (Prevention) Act is structured around premises and transactions: brothels, living on the earnings of prostitution, procuring and inducing, detaining, and soliciting in a public place. Its enforcement has therefore always been double edged, because the soliciting offence and the raids conducted under the Act fall in practice on the women in prostitution whom the statute’s title describes as its beneficiaries, a criticism made of the Act by commissions, by the courts and by the sex workers’ collectives who have asked to be consulted on any replacement. The Act also carries a rehabilitative apparatus of protective homes and corrective institutions whose record has drawn judicial criticism of its own. The statute’s survival, unamended in substance while comprehensive trafficking legislation has repeatedly stalled, means the field’s instrument against the organised exploitation of women remains one drafted in 1956 for the suppression of immoral traffic in women and girls, retitled in 1986, and administered ever since in a vocabulary the rest of the field has abandoned. It is the clearest instance in the criminal family of the proposition that the age of a provision in this area measures the difficulty of replacing it rather than satisfaction with it.

Two features unify this sprawling family, and a reader who grasps them can predict how any member of it will behave. The first is that the complainant is not the beneficiary of the proceeding. A successful prosecution punishes the accused and may vindicate the complainant, but it does not house her, maintain her, restore her wages or return her property; the compensation provisions in the criminal procedure codes exist, but they are ancillary to sentence and modest beside the civil orders the second family can make. The second is that the family’s offences are proved to the criminal standard, beyond reasonable doubt, about events that characteristically occur in private, inside homes and relationships, where independent witnesses are rare and delay corrodes what evidence exists. Low rates of conviction in this field are therefore partly a measurement of the standard doing its work on a particular kind of fact pattern, a point the design debate later in this article takes seriously, and only partly a measurement of police failure, though the record supplies plenty of that as well. Neither feature is an argument against the criminal family, whose offences mark the community’s judgment about conduct that deserves punishment. Both features are arguments against expecting the criminal family, alone, to deliver relief to the person harmed, which is the expectation Indian legislation loaded onto it for the field’s first four decades.

The second family: the civil remedy that arrived in 2005

The Protection of Women from Domestic Violence Act, 2005, Act 43 of 2005, which came into force on 26 October 2006, is the field’s structural exception, and its exception proves the argument of this article. For forty four years after the dowry statute, a woman facing violence in her home had offences to invoke and almost nothing else. The statute of 2005 changed the question the law asks. Instead of asking only whether a crime had been committed against her, it asks what she needs in order to be safe and to remain housed and supported, and it arms a Magistrate to order exactly that. The full treatment of the statute belongs to its pillar article, the Domestic Violence Act 2005 explained, and what follows is the map-maker’s account of why the statute sits alone in its family and what its design demonstrates.

The Act defines domestic violence widely, reaching physical, sexual, verbal, emotional and economic abuse, and it defines the protected relationship widely too, covering women in relationships by marriage, by blood, by adoption and in relationships in the nature of marriage. Its operative machinery is a set of civil orders. A protection order under Section 18 restrains the respondent from committing or aiding acts of domestic violence and from contacting the aggrieved person or entering her workplace or her child’s school. A residence order under Section 19 protects her right to reside in the shared household, a right the statute confers independently of ownership or tenancy, and can direct the respondent to remove himself from the household, to stay away from her portion of it, or to secure alternative accommodation of the same standard. A monetary relief order under Section 20 directs the respondent to meet the expenses and losses caused by the violence, including loss of earnings, medical expenses, damage to property and maintenance for the woman and her children. A compensation order under Section 22 can add damages for the injuries, including mental distress, that the violence caused, and a custody order under Section 21 settles the temporary custody of children. The orders are civil in character and are decided on the civil standard of proof, and the statute surrounds them with a service architecture of Protection Officers, registered service providers and shelter and medical facilities that the state is obliged to notify and maintain.

Two features of the design repay attention because they expose what the criminal family cannot do. The first is the residence right. The statute separates the right to live in the shared household from the law of property: a woman may have no title to the house and no tenancy in her name, and still hold a statutory right not to be evicted from it except in accordance with law. Property law alone could not have produced that right, and the criminal law could not have produced it either, because no offence is committed by a homeowner who simply wishes a household member gone. The civil statute created a new entitlement precisely where the older families were silent. The second feature is the relationship between the civil orders and criminal liability. Breach of a protection order is itself an offence under Section 31, punishable with imprisonment, so the civil route is backed by a criminal sanction at the point of disobedience; but the proceeding that generates the order remains civil, the relief remains oriented to the woman’s position rather than to the respondent’s punishment, and the same facts can found both a civil application under this Act and a prosecution under the cruelty provision of the criminal code, the two running in parallel without either extinguishing the other. That parallelism is the single most misunderstood feature of the field, and the navigation section of this article returns to it, because readers persistently treat the routes as alternatives between which a choice must be made, when the statute book treats them as cumulative.

Why did the civil family arrive so late, and why has it not grown? The honest answer is institutional rather than ideological. A criminal prohibition costs the state little at the moment of enactment: the police, the prosecutors and the courts already exist, and a new offence is a line in a code. A civil remedy of the 2005 kind requires the state to build and staff a service layer, Protection Officers in every district, notified shelter homes and medical facilities, legal aid that actually reaches applicants, and Magistrates with the time to hear applications quickly, since the statute directs that the first hearing ordinarily be fixed within three days of the application and the application disposed of within sixty days of that hearing. Those directions are honoured unevenly, and the unevenness is the predictable consequence of enacting a service-dependent statute into districts where the service layer is thin. The lesson generalises across the field and it is one of the connective insights this article exists to supply: the age of a family of statutes in this field is roughly inverse to the administrative machinery it needs. Prohibition, which needs almost nothing new, came first and grew fastest. Entitlement, which needs delivery systems, came last and remains thinnest. The civil remedy, which needs a modest but real apparatus, sits between them in both date and depth.

The statute’s singularity also shapes its litigation. Because the civil family has one member, that member carries arguments that in another field would be distributed across several Acts: arguments about the meaning of a shared household, about who may be arrayed as a respondent, about the reach of the Act to relationships outside formal marriage, and about the interplay between its monetary orders and maintenance under other laws. The Supreme Court’s reading of the statute has been correspondingly consequential, and the judicial section of this article takes up the pattern. For the map, it is enough to hold the structural point. When a reader asks which law governs violence at home, the complete answer is a pair: the offence belongs to the first family and the relief belongs to the second, and a guide that names only one of them has answered half the question.

The civil statute also reaches sideways into the rest of the legal system in a way no other member of the field attempts. Section 26 of the Act permits the reliefs it provides, protection, residence, monetary relief and compensation, to be sought within any other legal proceeding pending before a civil court, a family court or a criminal court in which the aggrieved person is a party, so that a woman already litigating a divorce, a maintenance claim or a prosecution need not open a second front in order to obtain protection; the court seised of the existing matter can grant the Act’s reliefs inside it. Section 23 arms the Magistrate to grant interim and ex parte orders on the strength of an affidavit where a prima facie case of domestic violence is disclosed, which is the provision that does the statute’s emergency work, since the typical application arrives at a moment of crisis in which notice and counter affidavit would defeat the purpose of approaching the court. The design as a whole is the most procedurally literate in the field, drafted with an evident knowledge of how litigation actually fails poor and dependent litigants: through delay, through fragmentation across forums, and through the interval between filing and first relief. That the outcomes so often fall short of the design returns the reader to the service layer, and to the evaluations, by the Commission, by parliamentary committees and by the monitoring mechanisms the statute itself attracted, that record Protection Officers holding the charge as an addition to other duties, districts in which the notified facilities exist on paper, and legal aid that arrives late or not at all. The statute remains the field’s proof of concept. It demonstrates that a civil remedy in this territory can be drafted to fit the harm. It equally demonstrates that drafting is the part of the problem Parliament controls, and delivery is the part it does not.

The civil statute’s drafting history is itself part of the field’s record, because it is the one major enactment in this article that was substantially written outside the government’s drafting rooms. Through the nineteen nineties, women’s organisations and lawyers’ collectives pressed for a civil law on domestic violence on the ground the criminal law’s inadequacy had made familiar, and drafts circulated, were revised in consultation, and formed the basis on which the government eventually legislated; the Lawyers Collective’s Women’s Rights Initiative, which had worked on those drafts, then ran the first systematic monitoring of the Act’s implementation, publishing the Staying Alive reports that documented, district by district, the gap between the statute’s service layer and its existence on the ground. The sequence inverted the field’s usual legislative mechanics. Elsewhere in this article, statutes are responses to catalytic events, drafted in weeks under public pressure. The civil statute was a decade of institutional design carried on outside government and adopted by it, and its distinctive features, the Protection Officer, the service provider, the residence right detached from title, the reliefs available inside other proceedings, bear the marks of drafters who had spent years watching litigation fail the women who brought it. That the statute’s implementation then disappointed its own drafters is the field’s oldest lesson arriving on schedule: the quality of a text and the capacity of the machinery that must operate it are independent variables, and a reform movement that wins the text has won the half of the contest that takes place in public.

One design choice inside the civil statute is worth isolating, because it marks the boundary between the civil family and the criminal family more precisely than any commentary has. Breach of a protection order is an offence, and the Act makes that offence cognizable and non-bailable, placing the criminal sanction at the exact point where the civil order is disobeyed and nowhere else. The complainant’s journey through the statute is therefore civil at every step that concerns her position: where she lives, what she is paid, whether she is contacted, who holds the children. The state’s coercive apparatus enters only when the respondent violates what a court has already ordered, which converts the criminal law from the instrument of first resort, as it is everywhere else in this field, into a backstop. Compare the cruelty offence, where the same apparatus is the instrument of first and only resort, and where its failure to convict leaves the woman with nothing in hand. The comparison is the strongest internal evidence for this article’s thesis that the families differ not in the seriousness with which they regard the harm but in the position from which they begin: the first family begins from the wrong and works towards punishment, the second begins from the woman and works towards safety, and the third and fourth, in their different ways, begin from an institution, the employer and the welfare state, and work towards a duty. Everything else in the field, the standards of proof, the forums, the limitation periods and the relief, is a consequence of those starting points.

The third family: workplace obligation and the law that came from a court

The workplace family has a different birth story from every other family in this field, and the story is itself a finding about how Indian protective legislation gets made. Its founding instrument was not a statute but a judgment. In Vishaka v. State of Rajasthan, decided in 1997, the Supreme Court confronted a petition arising from the gang rape of a social worker in Rajasthan and the absence of any legislation governing sexual harassment at work. Rather than wait for Parliament, the Court laid down guidelines, binding as law until legislation occupied the field, holding that sexual harassment at the workplace violates the rights to equality under Articles 14 and 15, the right to practise a profession under Article 19(1)(g) and the right to life with dignity under Article 21, and drawing on the Convention on the Elimination of All Forms of Discrimination against Women, which India had ratified in 1993, to supply the content the domestic statute book lacked. Employers were directed to prevent harassment, to treat it as misconduct, and to constitute complaint committees headed by a woman with an external member drawn from a non-governmental organisation. For sixteen years those guidelines were the law of the workplace in India, which makes this family’s sequence the reverse of the field’s usual pattern: judicial intervention preceded legislation, and the statute, when it came, codified and widened a scheme the courts had already built.

How did the courts expand the idea of dignity under Article 21 in this field?

The expansion ran through the workplace door. In the Vishaka judgment the Supreme Court treated a safe working environment not as a welfare concession but as a component of the right to life with dignity, and grounded the employer’s duty in that right read with the equality provisions. Later benches carried the same move into adjacent territory: the dignity of the person became the premise from which courts reasoned about privacy, about the terms on which the state may inquire into intimate life, and about the construction of protective statutes whose text lagged behind the harm. The mechanism matters more than any single holding. Once dignity under Article 21 is the anchor, a statute is construed to protect the person rather than to favour the institution applying it, and gaps in the text are filled in the direction of the right. That is why the judicial section of this article describes the pattern in this field as purposive: the courts were not merely interpreting a dignity jurisprudence into the statutes, they were supplying the premise on which the statutes were then read.

The statute that eventually occupied the field is the Sexual Harassment of Women at Workplace (Prevention, Prohibition and Redressal) Act, 2013, Act 14 of 2013, which came into force on 9 December 2013 and whose full treatment belongs to its pillar article, the POSH Act 2013: workplace harassment law. Its design is the workplace family’s design in pure form. It does not primarily create an offence. It imposes obligations on the employer: to constitute an Internal Committee at every workplace with ten or more employees, to frame and publicise a policy, to organise awareness, to assist a complainant who chooses the criminal route, and to act on the committee’s recommendations. Complaints from workplaces without an Internal Committee, and complaints by and against domestic workers, go to a Local Committee constituted at district level. The committee inquires with the powers of a civil court for that purpose, can recommend interim measures such as transfer or leave, and can recommend that compensation determined on stated factors be recovered from the respondent’s salary. The proceeding is neither a prosecution nor a civil suit; it is an institutional inquiry backed by penalties on the employer who fails to build the institution. That is the third logic at work: where the first family punishes conduct and the second relieves the person harmed, the third conscripts the organisation in which the harm occurs and makes the organisation answerable for prevention and redress.

The older members of the workplace family came not from the harassment problem but from the economics of women’s employment, and they are the members whose fate a reader most needs to have explained, because the statute book has moved underneath them. The Maternity Benefit Act, 1961, Act 53 of 1961, enacted under the constitutional direction in Article 42, guaranteed paid absence before and after childbirth to women in covered establishments and protected them against dismissal during that absence. The Maternity Benefit (Amendment) Act, 2017, Act 6 of 2017, raised the paid entitlement from twelve weeks to twenty six weeks, added leave for adopting and commissioning mothers, required a creche facility in establishments with fifty or more employees, and enabled work from home by agreement after the leave period. The Equal Remuneration Act, 1976, Act 25 of 1976, implemented the directive principle in Article 39(d) and India’s obligations under the International Labour Organization’s Equal Remuneration Convention of 1951, which India ratified in 1958: it required equal pay for men and women for the same work or work of a similar nature and prohibited discrimination against women in recruitment and in conditions of service after recruitment. Both statutes were pillars of the workplace family for decades. Both have been repealed.

The repeal is the consequence of the consolidation of India’s labour statutes into four codes, and at the point of subsumption the reader should follow the two codes that absorbed this family’s content: the Code on Wages 2019 explained and the Social Security Code 2020 explained. The four labour codes were brought into force on 21 November 2025, and from that date the twenty nine subsumed statutes stood repealed, with savings for actions already taken. The equal pay content travelled into the Code on Wages, 2019, whose Section 3 restates the duty in altered terms: where the Act of 1976 framed the obligation as equal remuneration for men and women, the Code frames it as a prohibition on discrimination on the ground of gender in wages and in recruitment, a formulation that widens the protected ground while relocating the claim into the Code’s general machinery of authorities and claims. The maternity content travelled into the Code on Social Security, 2020, whose maternity benefit chapter carries forward the substance of the amended Act of 1961, including the twenty six weeks at its centre, the provisions for adopting and commissioning mothers and the creche obligation graded by establishment size. The POSH Act was not among the subsumed statutes and continues as a free-standing workplace statute.

This migration creates the single most common currency error in writing about women’s rights in India. Pages prepared before the codes commenced list the Maternity Benefit Act and the Equal Remuneration Act as living statutes, send readers to repealed texts, and quote section numbers that no longer govern anything. The repeal does not diminish the entitlements, which survive in the codes, but it changes where a claim is filed, which authority hears it, and how it is framed. A claim about unequal pay is made under the Code on Wages; a claim about maternity benefit is made under the Code on Social Security; and an adviser who reaches for the repealed Acts will misdirect both. The migration also carries a subtler consequence for this family’s logic. Inside a dedicated statute, a maternity entitlement read as a women’s right. Inside a general social security code, it reads as one benefit among many in a universal scheme, administered through the code’s funds and authorities. Whether that relocation strengthens the entitlement by universalising its machinery or weakens it by dissolving its dedicated forum is a live question of design on which the record is young, and this article records the question rather than pretending to an answer the commencement date cannot yet support.

The workplace statute’s reach deserves its own note, because it is wider than the office employment its name suggests. The definition of an employee extends beyond regular staff to women engaged on a temporary, daily wage, contractual or ad hoc basis, and the definition of an aggrieved woman takes in a woman of any age who alleges harassment at a workplace, whether or not a conventional contract of employment can be produced. The duties, correspondingly, fall on the employer as defined across the public sector, companies, firms, co-operative societies, trusts, educational institutions and the informal arrangements in between, and the Act’s scheme for the unorganised sector, the Local Committee with its district officer and its nodal officers at block level, is an attempt, unique in the field, to build one statute with a formal sector track and an informal sector track inside it. The compliance record of the formal track is imperfect in a documented direction: committees not constituted, external members not appointed, annual reports not filed, and penalties for default that attach to the employer in amounts trivial against an institutional budget. The Supreme Court, years before the statute, had already had to direct in Medha Kotwal Lele v. Union of India, in 2013, that states amend their service rules and standing orders to treat the Vishaka complaints committees as the inquiry authorities contemplated by conduct rules, because the guidelines’ authority on paper was not producing committees on the ground. The statute replaced that improvisation with a uniform duty, and the gap that remains is the field’s recurring one, between the duty as enacted and the institution as staffed. The central government additionally maintains an online complaint mechanism, the SHe-Box portal, through which a complaint of workplace harassment can be lodged and transmitted to the authority concerned, a small piece of delivery infrastructure of the kind the workplace family, alone among the four, has begun to accumulate.

Equal pay, the workplace family’s oldest promise, is also the clearest case in the whole field of a right whose text outran its machinery. The Act of 1976 stated the duty with perfect clarity: no employer shall pay any worker remuneration at rates less favourable than those paid to workers of the opposite sex for the same work or work of a similar nature, and no reduction of any worker’s rate in order to comply. What the Act did not supply was a realistic path for the individual woman who discovered the differential, usually in an establishment where wages were opaque, comparators were few, and the inspectorate charged with the Act was the same stretched labour inspectorate charged with everything else. Claims were rare, awards rarer, and the duty’s real life was lived in the organised sector’s pay structures, where the differential it targeted had largely been formalised away, rather than in the informal employment where the differential by gender is widest and the statute’s definitions least fit. The Code on Wages inherits both the duty and the problem. Its restatement of the duty in the language of gender discrimination is broader, its definition of wages is uniform across the labour codes, which matters because discrimination can hide in the composition of the wage as readily as in its total, and its claims machinery is the labour codes’ general machinery, no more dedicated to this wrong than the inspectorate it replaces. The prediction the map records is therefore cautious: the migration improves the statement of the right and the architecture around it, and the variable that will decide its practical force is the one that decided the force of the Act of 1976, the willingness of the enforcement machinery to treat a pay differential as a case rather than as a market outcome.

The fourth family: welfare entitlement and the promise that needs no defendant

The youngest family in the field operates on a logic the other three cannot imitate: it requires no wrongdoer. A prosecution needs an accused, a civil application needs a respondent, and a workplace inquiry needs an employer. An entitlement needs only a beneficiary who qualifies. The family’s statutory anchor is the National Food Security Act, 2013, Act 20 of 2013, a statute remembered for foodgrains, whose Section 4 entitles every pregnant woman and lactating mother, subject to the schemes framed under it, to a free meal through the local anganwadi during pregnancy and for six months after childbirth, and to a maternity benefit of not less than six thousand rupees in prescribed instalments, excluding women in regular government employment or those receiving similar benefits under another law. The design is significant out of proportion to the sum. It reaches the woman whom the workplace family cannot reach at all: the woman in informal work, in agriculture, in her own household’s economy, who has no employer against whom a maternity claim could run. For her, the third family’s twenty six weeks of employer-paid leave are a description of someone else’s life, and the food security statute’s cash and meal entitlement is the only maternity provision in the entire field addressed to her situation.

Around that anchor sits a scheme layer that is not statute at all, and the distinction matters to a reader trying to enforce anything. The Pradhan Mantri Matru Vandana Yojana, the maternity benefit scheme through which the food security entitlement is principally delivered, pays five thousand rupees in instalments for the first living child, administered as a centrally sponsored scheme with its own eligibility conditions, registration systems and exclusions. The gap between the statute’s figure and the scheme’s figure, and between the statute’s coverage of pregnant women generally and the scheme’s focus on the first child, is the fourth family’s characteristic weakness in concrete form: the entitlement is statutory, but its quantum and reach in practice are set by the scheme framed under it, and a scheme can be redesigned by executive decision in a way a statute cannot. The One Stop Centre scheme, which funds centres intended to give medical, legal, police and shelter assistance to women facing violence under one roof, and the Nirbhaya Fund, the corpus created after 2012 from which such projects are financed, belong to the same layer. They are real and they matter to the women who reach them, but they are programmes rather than rights, and a guide that lists them alongside statutes without marking the difference teaches its reader to expect from a scheme the enforceability of an Act.

The thinness of this family is the field’s most revealing statistic, and it is a statistic about legislative choice rather than about drafting. India built its protective architecture for women over six decades and filled three of its four families with substantial law, while the family that asks nothing of a defendant and everything of the state’s delivery systems acquired one statutory entitlement of general application. The reasons are continuous with the lateness of the civil family. Entitlements are expensive in administration rather than in enactment; they require anganwadis that function, payment systems that reach informal workers, and budgets that recur every year rather than a penalty clause that costs nothing to print. The maternity comparison makes the point sharply. The formal sector woman receives twenty six weeks of wages because her employer can be ordered to pay them. The informal sector woman, who is the large majority of women who work in India, receives a cash benefit measured in thousands of rupees because the state must pay it itself. The field’s generosity tracks the identity of the payer with uncomfortable precision, and any honest map of women’s rights laws in India has to record that the rights are most complete where someone other than the state can be made to fund them.

The financing of maternity protection is the design question inside this family that most repays a reader’s attention, because the Indian model answers it in a way the international instruments were drafted to avoid. Under the maternity benefit statutes and their successor chapter in the social security code, the cost of the twenty six weeks is borne by the individual employer of the woman who claims it: an employer liability model. The international labour standards on maternity protection have long preferred social insurance or public funding precisely because employer liability prices the entitlement into the hiring decision, giving the employer of a woman of childbearing age a visible, individualised cost that the employer of a man does not bear, and thereby building a quiet disincentive into the very statute that prohibits discrimination. Indian policy debate has recorded the argument repeatedly, and the social security code’s architecture, with its funds and its contemplated extension of financing beyond the single employer, acknowledges it without yet displacing the liability model for the established workforce. The welfare family’s cash benefit is the compensating instrument for women outside that workforce, funded by the state because no employer exists to charge. Set side by side, the two instruments describe the field’s settlement with unusual clarity: full wage replacement where an employer can be identified and charged, a flat cash sum where the state must pay, and a financing debate, decades old, about the distortion the first instrument creates and the inadequacy of the second.

Maintenance: the thread that runs across all four families

No subject demonstrates the field’s fragmentation better than maintenance, the obligation to support a wife, a former wife, children and parents, because maintenance is claimed in more forums, under more instruments, than any other relief in this article, and because the instruments overlap without consolidating. The secular route is the oldest of the modern layers: Section 125 of the Code of Criminal Procedure, 1973, empowered a Magistrate to order a person with sufficient means to pay a monthly allowance to a wife unable to maintain herself, to children and to dependent parents, deliberately irrespective of the religion of the parties, and the provision is carried into the Bharatiya Nagarik Suraksha Sanhita, 2023 as Section 144. Its placement in a criminal procedure code is a historical accident with practical consequences: the proceeding is summary, the standard is geared to preventing destitution rather than to dividing matrimonial property, and the order is enforced, in default, by imprisonment, which makes the maintenance jurisdiction a hybrid that belongs fully to none of the four families and borrows the machinery of the first.

Alongside the secular route stand the personal law routes. The Hindu Adoptions and Maintenance Act, 1956 entitles a Hindu wife to be maintained by her husband during her lifetime, to live separately without forfeiting maintenance on stated grounds, and to claim maintenance from the estate of a deceased husband in the hands of his heirs; the Hindu Marriage Act, 1955 adds maintenance pendente lite and permanent alimony within matrimonial proceedings. For Muslim women, the dissolution of a marriage redirected the claim: after the Supreme Court’s decision in Mohd. Ahmed Khan v. Shah Bano Begum in 1985 read the secular provision to cover a divorced Muslim woman beyond the iddat period, Parliament enacted the Muslim Women (Protection of Rights on Divorce) Act, 1986, Act 25 of 1986, requiring a reasonable and fair provision and maintenance to be made and paid within the iddat period, together with mehr and the return of properties; and in Daniel Latifi v. Union of India in 2001 the Court upheld that Act by construing the provision to be made within iddat as one that must extend to the woman’s future needs until remarriage, a purposive reading that preserved the substance of the entitlement inside the statute designed to displace the earlier decision. The civil statute of 2005 then added a fourth door: monetary relief under Section 20 of the domestic violence Act includes maintenance for the woman and her children, expressly in addition to an order under the secular provision or any other law, so that a Magistrate under the civil statute and a Magistrate under the procedure code can both be moved on the same need.

The lesson the maintenance thread teaches is the one this article keeps returning to, stated here in its sharpest form. Where the law offers several doors to the same relief, the outcome depends on which door is chosen and by whom the claimant is advised, because the forums differ in speed, in the quantum they habitually award, in their enforcement machinery and in the defences they entertain. A unified family law would have made maintenance a single claim with a single measure. The accreted field made it four claims with four measures, and the skill the field actually demands of a practitioner is not the recitation of any one of them but the comparison among them. It is also the thread on which the distance between enactment and delivery is most measurable, since a maintenance order is a promise of periodic payment whose value depends entirely on execution against earnings and property, and the arrears litigation that fills the family courts is the standing evidence that in this field, as in the criminal one, obtaining the order and receiving the relief remain two different events.

The institutional layer: a commission that can inquire but cannot decide

Standing behind the four families is an institutional layer that readers persistently overestimate, and correcting the overestimate is one of the useful things a map can do. The National Commission for Women was constituted in January 1992 under the National Commission for Women Act, 1990, Act 20 of 1990, as a statutory body charged with reviewing the constitutional and legal safeguards for women, recommending legislative measures, facilitating the redress of grievances and advising the government on policy affecting women. Its powers are real but of a particular kind. Under Section 10 of its Act, the Commission may investigate and examine matters relating to the safeguards provided for women, take up cases of violation with the appropriate authorities, look into complaints and take suo motu notice of matters involving the deprivation of women’s rights, the non-implementation of protective laws and non-compliance with policy decisions meant for women’s welfare, and it may call for special studies and participate in the planning process for women’s development. When it investigates, it holds the powers of a civil court for summoning witnesses, requiring documents and receiving evidence on affidavit. What it does not hold, anywhere in its Act, is the power to decide a case, to award compensation, to direct a prosecution or to bind any authority by its conclusions. Its findings travel as recommendations. A state commission for women constituted under the corresponding state legislation occupies the same position one level down.

The distinction between inquiry and adjudication is not a technicality, and treating the Commission as a court substitute is among the errors this article is written to correct. A woman who carries her grievance to the Commission gains an investigator with the power to summon records and a public body whose intervention often moves a stalled file in a police station or a department, and those are genuine goods that experienced practitioners use deliberately. She does not gain a judgment. The Commission cannot order the relief that a Magistrate can order under the civil statute, cannot convict as a criminal court can, and cannot compel an employer as the workplace machinery can. Its interventions succeed by authority, publicity and persistence rather than by enforceable command, which is why the Commission’s own reports return so often to the theme of recommendations awaiting action. The institutional design is consistent with the field’s larger pattern. India has been readier to build bodies that inquire into the condition of women than bodies empowered to alter it directly, and the Commission, for all the sincerity of its mandate, belongs to the architecture of attention rather than to the architecture of relief. A reader navigating the field should therefore place the Commission correctly in the sequence of a real case: a parallel channel of pressure and record-making, valuable alongside a proceeding, and never a replacement for one.

The Commission’s working methods explain both its value and its limits in practice. Its complaints and investigation work proceeds through written complaints, hearings at which the parties and the authorities concerned are summoned, fact finding visits, and recommendations addressed to the government or the authority whose action is sought; its annual report, laid before Parliament through the central government, is required to be accompanied by a memorandum of the action taken or proposed on the Commission’s recommendations, and it is in that memorandum, year after year, that the institutional truth of the design is told, in the distance between the recommendation and the action reported against it. The Commission has used its mandate to conduct inquiries into conditions in custodial and shelter institutions, to review proposed legislation clause by clause, and to intervene in litigation touching the safeguards it exists to review. None of this is negligible, and practitioners in the field treat a Commission notice as an event that moves files. But the design choice stands: at the one point where the field might have been given an adjudicatory authority of its own, with the power to award the relief the other families distribute among magistrates, committees and claims authorities, Parliament built a body of inquiry and counsel. The reader who understands that choice understands the institutional layer’s place in the architecture, which is commentary and pressure, arranged around forums that alone can decide.

Around the Commission stands a supporting institutional cast that a complete map should register, if only to prevent the reader mistaking the Commission for the whole of it. The Ministry of Women and Child Development is the nodal ministry for the field, administering the schemes of the welfare family and the funds through which the service layer is financed, and its budget lines, rather than any statute, determine how many One Stop Centres open and how the maternity scheme is paid for. The legal services authorities supply the field’s right to representation: under the Legal Services Authorities Act, 1987, every woman is entitled to free legal aid as a member of a category the Act names, without a means test of the kind applied to the general population, and the district legal services authorities are charged with supplying counsel, including for the civil applications and maintenance claims in which representation decides outcomes. The fast track special courts, funded as a centrally sponsored scheme from 2019 for the trial of rape cases and cases under the child protection statute, are the judiciary’s corresponding piece of delivery machinery, an attempt to answer the pendency problem by ring fencing a class of trials rather than by reforming the trial system entire. None of these institutions adjudicates the field into coherence. Together they form the infrastructure without which the four families’ texts remain what their critics call them, strong laws, and their presence or thinness in a given district predicts the lived content of the statutes there more reliably than any provision in the Acts themselves.

The judicial pattern: purposive reading, and the caution that travels with it

Across this field the courts have followed a recognisable double pattern, and a reader who can state it can predict the direction of most litigation in the cluster. The first movement is purposive construction. Protective statutes are read to advance the protection, ambiguities are resolved in favour of the person the statute exists to protect, and provisions that would defeat the statute’s object on a technicality are read down or struck out. The domestic violence statute supplies the cleanest instance. As enacted, its definition of a respondent was confined to an adult male person, with a proviso touching female relatives in a limited way. In Hiral P. Harsora v. Kusum Narottamdas Harsora, in 2016, the Supreme Court struck the words adult male out of the definition, reasoning that a provision meant to protect women in a household could not rationally exclude complaints against the women of that household, and that the classification had no defensible connection with the statute’s purpose. The statute emerged wider than Parliament had written it, by the removal of two words. The succession field supplies a longer instance of the same movement. The Hindu Succession (Amendment) Act, 2005, Act 39 of 2005, amending the Hindu Succession Act, 1956, Act 30 of 1956, made the daughter of a coparcener a coparcener by birth in her own right, with the same rights and liabilities as a son. For fifteen years the courts divided over whether the amendment benefited only daughters whose fathers were alive on the date it commenced, a reading that would have made a daughter’s birthright depend on the accident of her father’s survival. In Vineeta Sharma v. Rakesh Sharma, in 2020, a three judge bench settled the question the other way: the right is by birth, it does not depend on the father being alive at commencement, and the preliminary decrees drawn on the narrower view did not conclude the matter. Purposive reading, in this field, is not a slogan. It is the documented habit by which the judiciary has repeatedly chosen, between two available constructions, the one that makes the protection real.

The constitutional jurisprudence runs in the same direction and supplies the field’s premises. In Shayara Bano v. Union of India, in 2017, a constitution bench set aside the practice of instant triple talaq by a majority, locating the defect in the practice’s arbitrariness and its incompatibility with the dignity of the women it governed, and Parliament followed with legislation of its own in 2019. In Joseph Shine v. Union of India, in 2018, the Court struck down Section 497 of the Penal Code, the adultery offence, holding a provision that treated a wife as the property of her husband and gave him alone the right to complain to be incompatible with equality and dignity. These decisions belong formally to personal law and to the penal code rather than to the four families, but they are part of this field’s working constitution, because they establish the premise the families presuppose: that provisions governing women’s lives are to be tested against the woman’s standing as a full constitutional person, not against the convenience of the household or the community arranged around her.

The second movement of the pattern is the caution, and it is as well documented as the first. The same courts that read protective statutes generously have recorded, in a line of decisions on the cruelty provision, a judicial anxiety about the provision’s use in matrimonial breakdown. In Preeti Gupta v. State of Jharkhand, in 2010, the Supreme Court observed that complaints under Section 498A were in a large number of cases not bona fide, that relatives including elderly parents and family members living abroad were being implicated on general allegations, and it asked the legislature to reconsider the provision. In Arnesh Kumar v. State of Bihar, in 2014, the Court addressed arrest rather than the offence, holding that arrest in cases punishable with up to seven years, the cruelty offence among them, is not to be automatic, and directing police to satisfy themselves against a checklist drawn from Section 41 of the Code of Criminal Procedure before arresting, with magistrates required to record their satisfaction before authorising detention. In Rajesh Sharma v. State of Uttar Pradesh, in 2017, a bench went further and directed the constitution of family welfare committees to which complaints would first be referred, a direction that a three judge bench in Social Action Forum for Manav Adhikar v. Union of India, in 2018, substantially withdrew, holding that the committee mechanism was a legislative act beyond the judicial province and restoring the position that misuse is to be met by the arrest safeguards and by the ordinary discipline of investigation, not by interposing a screening body the statute never created.

The debate those cases belong to deserves its strongest statement on both sides, because both are made in good faith and the article’s business is to map rather than to settle it. The case for restraint runs as follows, and it is the case the benches in Preeti Gupta and Rajesh Sharma accepted in substance: a non-bailable offence of wide definition, arising inside a failed marriage, is an instrument of exceptional leverage in a matrimonial dispute; the process itself, arrest, bail applications, years of trial, inflicts punishment irrespective of outcome; and a provision enacted to protect women from cruelty loses legitimacy when it is deployed to settle scores, which the acquittal figures, on this view, corroborate. The case against dilution runs as follows, and it is the case pressed by the National Commission for Women and by women’s organisations whenever amendment has been proposed: cruelty inside a marriage is real, widespread and structurally hard to prove; an acquittal records that the prosecution did not prove its case beyond reasonable doubt, not that the complaint was false, and the distance between those two propositions is where most misuse statistics quietly change their meaning; weakening the provision because some complaints fail would withdraw protection from the class the provision exists for, on evidence that measures the standard of proof rather than the incidence of the conduct. Successive law reform exercises have landed between the two positions, keeping the offence and its character while tightening the arrest practice around it, and the dedicated treatment of the provision and its judicial history sits at Section 498A and the judicial response, which examines the evidence in the detail the argument requires. What the map should carry away is the shape of the settlement the courts have produced: the offence stands, its construction remains protective, and the system’s answer to the misuse concern has been procedural, addressed to arrest and investigation, rather than substantive, addressed to the definition of the wrong.

The same generation of benches dismantled protective provisions that protected women out of their livelihoods. In Anuj Garg v. Hotel Association of India, in 2008, the Supreme Court struck down a provision of the Punjab excise law that barred the employment of women in premises where liquor was consumed, holding that legislation of that kind rests on what the judgment called romantic paternalism: the state casting women as a class requiring protection from employments and environments it leaves men free to enter, and thereby entrenching the very inequality the protective frame claims to answer. The decision is the necessary counterweight to this section’s generalisation about purposive reading. The courts in this field are not generous to legislation about women in the abstract; they are generous to legislation that enlarges a woman’s legal position and correspondingly hostile to legislation that restricts it in protection’s name, and the two impulses are one jurisprudence, organised by the premise that the woman is a full constitutional person whose choices are hers. Read with Anuj Garg beside it, even Article 15(3) changes its emphasis in the case law: the provision sustains special provisions for women, and the courts have used it to sustain provisions that confer benefits and open occupations, while provisions that confine women for their own good have been made to justify themselves against a scepticism the text of the exception does not soften.

The purposive pattern is visible in the smaller litigation too, in fields a general reader rarely connects to this one. In Mary Roy v. State of Kerala, in 1986, the Supreme Court held that the Indian Succession Act, 1925 governed succession for Syrian Christians in the former Travancore territory, displacing the Travancore Christian Succession Act, 1916, whose provisions had confined a daughter’s share to a fraction of a son’s; the result, reached by statutory construction rather than by equality review, equalised the inheritance position of a class of women whom the personal law debate had passed by. In Gita Hariharan v. Reserve Bank of India, in 1999, the Court read the word after in Section 6 of the Hindu Minority and Guardianship Act, 1956, which names the father as natural guardian and the mother after him, to mean in the absence of rather than only after the father’s lifetime, so that a mother could act as natural guardian during the father’s life where he was absent from the child’s care; the alternative construction, which would have struck the provision down, was avoided by reading the word against its ordinary place in the sentence, in order to save the statute and the mother’s standing together. In the domestic violence litigation itself, the pattern ran first one way and then the other: an early reading of the shared household definition had confined the residence right in ways the women’s movement regarded as defeating the Act, and in Satish Chander Ahuja v. Sneha Ahuja, in 2020, a three judge bench rejected the narrower construction and restored a reading under which the household where the woman lived or had lived in a domestic relationship could qualify, whether or not she held any right in it. Taken together, the cases show the judicial branch doing, case by case, the harmonising work that a consolidating statute would have done at once, and doing it with the tools of construction because the constitutional route to the personal law substrate remained the blocked path described earlier in this article.

The enforcement record and the design question it raises

The standard summary of this field, repeated in editorials and examination answers alike, is that India has strong laws and weak enforcement. The summary is broadly right about enforcement and incomplete about design, and separating the two is the analytical work this section exists to do. The record first. The National Crime Records Bureau’s Crime in India 2022 report recorded 4,45,256 cases registered as crimes against women in that year, with cruelty by a husband or his relatives the largest single head of registration, followed by kidnapping and abduction and assault; the bureau’s series is the standard source, its figures count registered cases rather than measured incidence, and comparisons across years must allow for changes in registration practice as much as in behaviour. Registration is only the first filter. Charge-sheeting, trial pendency measured in years, and conviction rates that sit far below the rates for offences against property complete a picture no honest account can soften: a complainant entering the criminal route accepts a proceeding whose duration is long, whose attrition is heavy, and whose outcome distribution is dominated by acquittal, compromise and exhaustion rather than by verdict. Dowry deaths continue to be recorded in the bureau’s series in the thousands each year, six decades after the prohibition that was supposed to end the practice that produces them. Whatever the design debate concludes, the enforcement record on its own terms is a record of shortfall, and this article does not qualify that finding away.

Why do so few complaints in this field end in conviction?

Part of the answer is the standard of proof doing exactly what it is designed to do on a difficult class of facts. The offences of this field are committed, characteristically, in private, within households and relationships, where the available witnesses are the accused’s family, the documentary record is thin, and the complainant’s account must carry the prosecution beyond reasonable doubt, often years after the events, against a defence whose task is only to open a doubt. Add the enforcement failures that are genuinely failures, reluctant registration, slow investigation, witnesses who turn hostile under family pressure, trials adjourned across years, and the conviction figures are overdetermined. The analytical mistake is to assign the outcome to one cause. A field that routes most of its harms through the criminal standard will produce low conviction figures even when every official in the chain performs, and the design question is whether the relief the complainant needs should ever have been made to wait on that standard being met.

The design question follows directly. A criminal route requires proof beyond reasonable doubt about events that occur in private, so a low conviction rate is partly a function of the standard rather than only of police failure. A civil route decides on a preponderance of probabilities and aims its orders at the complainant’s safety, residence and support rather than at the respondent’s punishment, so it can succeed for the person harmed in cases where a prosecution, on identical facts, would properly fail. That is the argument the statute of 2005 embodies, and it is an argument about institutional design rather than about the seriousness of the harm: nothing in it denies that cruelty deserves punishment, and everything in it denies that punishment is the only, or the first, thing the law owes the person subjected to it. The counter-argument deserves its place. Civil orders are only as strong as their enforcement, a protection order breached with impunity teaches contempt for the route that issued it, and the civil family’s service layer, Protection Officers and notified facilities, is exactly where the state’s investment has been thinnest; a remedy on paper that a district cannot staff reproduces, in a new form, the gap between enactment and delivery that disfigures the criminal record. Both arguments are true, and they are true about different variables: the first about which standard a claim should have to meet, the second about whether the state funds the machinery either route requires. The field’s tragedy is that the debate is usually conducted as though choosing the better design discharged the obligation to fund it.

The framing that protective laws are widely misused belongs in this section too, because it is at bottom a claim about the enforcement record, and it should be handled with the same separation of variables. The evidence usually marshalled consists of acquittal rates, judicial observations about over-implication, and the undoubted existence of complaints filed as leverage in matrimonial litigation. The evidence on the other side consists of the registration gap itself: the National Family Health Survey’s findings, across its rounds, that a substantial share of women who experience spousal violence neither seek help nor tell anyone, which sits awkwardly beside a picture of the law as primarily an instrument of false complaint, since a system can be simultaneously under-used by most of the class it protects and misused by a minority of complainants within it. Both things can be true at once, and in this field both are. The judicial section of this article set out how the courts have managed the tension procedurally. The map’s contribution is to insist that the misuse debate, conducted as a referendum on whether the offences should exist, is a category error: the live questions are about arrest practice, about the particularity of allegations, and about the speed of trial, and each of those is answerable without withdrawing the substantive protection from a class whose measured experience of violence is not in serious dispute.

A final caution about evidence belongs to the enforcement record, because misuse of the record’s numbers is the most common factual error in writing about this field. Registration figures measure the behaviour of the reporting system as much as the behaviour of offenders. A state in which women are encouraged to report, in which police register complaints as the law directs, and in which the special statutes are publicised will record more crimes against women than a state in which the same conduct meets a closed police station door, and the higher figure is in that respect the better news. Rising registrations after a legislative intervention, as followed the amendments of 2013, are therefore consistent with two opposite stories, an incidence story and a confidence story, and the bureau’s tables cannot by themselves choose between them. The discipline this article has applied, naming the report and the year, quoting the registered total as a registered total, and refusing to convert it into a rate of offending or a trend without the surrounding series, is the minimum the subject admits. What the numbers can support without strain is narrower and still damning enough: the volume of registered harm is large, the cruelty head dominates it, the dowry death count persists across decades of prohibition, and the system’s outputs, convictions delivered within a complainant’s reasonable horizon, lag its inputs by an order of magnitude. A reader who wants a single sentence for the record should take that one, rather than any extrapolation built on figures the source does not contain.

One compensation layer within the criminal process deserves separate mention, because it is the point at which the first family reaches, however imperfectly, towards relief. Section 357A of the Code of Criminal Procedure, inserted in 2009 and carried into the successor procedure code, obliges every state to prepare, in coordination with the central government, a victim compensation scheme administered through the legal services authorities, under which a victim, or the dependents of a deceased victim, may be awarded compensation where the offender is untraced, where no trial takes place, or where the compensation recoverable from a convicted offender is inadequate. Acid attack and rape survivors are the categories the schemes have most visibly addressed. The layer’s significance for the map is conceptual: inside the criminal route itself, the legislature eventually conceded that a proceeding which ends without a conviction may nevertheless leave a person the state acknowledges to have been harmed, and that acknowledgement has a monetary form. The quantum under the schemes is modest, the application process adds its own attrition, and no scheme award substitutes for the civil family’s orders. But the existence of the layer complicates any simple statement that the criminal family offers its complainants nothing, and its placement, attached to the legal services institutions rather than to the trial court alone, is a small instance of the field learning to route relief through the machinery that can actually deliver it.

Overlaps, gaps and the argument for a different drafting

A field built by accretion over sixty years will overlap with itself, and this one does so at its most sensitive point. The same course of conduct in a marital home can be, simultaneously, the offence of cruelty under the criminal code, the economic and emotional abuse that grounds civil orders under the statute of 2005, and, where a property demand drives it, an offence under the dowry statute and the predicate for the dowry death presumption if the woman dies. The overlaps are not drafting accidents. Each layer was added because the layer beneath it had failed at a specific task: the civil statute because the offence could not house or maintain the complainant, the presumptions because the ordinary burden of proof could not be discharged about deaths inside a household, the special offences because the general ones had not reached the conduct. Reading the field as a pile is a mistake; it is better read as a sediment, in which each stratum records the failure that produced it. The practical corollary for a person in difficulty is generous: the routes are cumulative, the choice among them is not exclusive, and the navigation table that follows is a map of combinations rather than of alternatives.

The gaps are as instructive as the overlaps, and three deserve naming. The first is the interface with personal law. The protective statutes are, in the main, laws of general application: the criminal code, the dowry prohibition, the civil statute of 2005 and the workplace statute apply across communities, and the civil statute applies to women of every religion because it governs conduct and relief rather than status. But the adjacent questions that dominate a woman’s legal position in a marriage, the validity and dissolution of the marriage itself, maintenance within the matrimonial codes, succession and guardianship, are governed substantially by the personal law of her community, reformed unevenly: the Hindu codes of 1955 and 1956 recast marriage, succession and adoption for Hindus, Buddhists, Jains and Sikhs, with the succession amendment of 2005 completing the coparcenary reform a half century later, while the corresponding questions for other communities rest on their own statutes and case law, from the Muslim Women (Protection of Rights on Divorce) Act, 1986 to the legislation of 2019 on instant triple talaq. A map of women’s rights that ignored this interface would mislead, because the protective field and the personal law field meet in almost every real dispute, and the practitioner moves between them clause by clause. The second gap is the gendered drafting of the protective statutes themselves, and the recurring proposal that they be drafted in gender neutral terms. The proposal has serious sponsorship: the committee headed by Justice J. S. Verma recommended in 2013 that the sexual offences be framed so that the perpetrator at least be described without gender, and the ordinance that first implemented its report adopted neutral language for a time before the Act as passed restored the gendered frame for the principal offences. The argument for neutrality is that harm, not identity, should define an offence, and that male and transgender victims of some of these wrongs are left to lesser provisions. The argument against is that the offences were drafted against a measured asymmetry of social fact, that the protective statutes answer a distribution of harm which is not symmetrical, and that neutral drafting of a remedy can dissolve the protection of the class the provision was enacted for, as critics argued when the child protection statute’s neutral frame was contrasted with the women specific statutes. The debate is live, the positions are held in good faith, and the existing settlement is a patchwork: the civil statute and the workplace statute are drafted for women, the child protection statute of 2012 is drafted neutrally for children of every gender, and the Transgender Persons (Protection of Rights) Act, 2019 supplies a separate anti-discrimination code for transgender persons rather than admission to the women specific statutes. The third gap, the reach of the field into informal work, has already been marked in the welfare section and is restated here only to complete the inventory: the workplace family presupposes a workplace, and the majority of working women in India do not have one in the statutory sense.

Two further items complete the inventory of the unsettled, because a map that marked only settled law would misrepresent the field’s working condition. The first is child marriage. The Prohibition of Child Marriage Act, 2006, which replaced the restraint statute of 1929, makes a child marriage voidable at the option of the party who was a child, void in defined circumstances of force, fraud and trafficking, punishable for the adults who solemnise, promote or permit it, and supported by child marriage prohibition officers; the marriage age stands at eighteen for women and twenty one for men, the asymmetry resting on premises that the Prohibition of Child Marriage (Amendment) Bill, 2021 proposed to displace by raising the age for women to twenty one, a Bill that was introduced, referred to a parliamentary committee and not enacted. The debate it opened reproduces the field’s central tension in miniature: one side reads a higher age as protection of health, education and autonomy, the other reads it as the criminalisation of the marriages of adult women in poorer communities and a further widening of the gap between the law’s text and social practice. The second unsettled item is trafficking. The Immoral Traffic (Prevention) Act remains the governing statute, organised around the suppression of prostitution in its organised forms rather than around the trafficking of persons as such; a comprehensive Trafficking of Persons (Prevention, Protection and Rehabilitation) Bill was passed by the Lok Sabha in 2018 and not carried through the Rajya Sabha, so the field still lacks the dedicated trafficking code that reform exercises have recommended for decades. In both instances the honest entry on the map is the same: the harm is named, the existing instrument is acknowledged on all sides to fit it imperfectly, and the replacement exists only as a proposal.

The federal dimension: central statutes, state machinery

The map so far has been drawn at the level of the Union, because the field’s statutes are overwhelmingly central, and that fact conceals the field’s most consequential division of labour. Criminal law and criminal procedure stand in the Concurrent List, which is how Parliament built the offences and the presumptions; marriage, divorce, adoption and succession stand there too, which is how the Hindu codes and the amendments of 2005 were enacted for the whole country at once. But police and public order are state subjects, and so are public health, the administration of justice in the subordinate courts in large part, and the welfare delivery systems through which the fourth family operates. The practical consequence follows the Constitution’s text: Parliament writes the protection, and twenty eight state governments and the Union territories staff it. A Protection Officer exists in a district because a state government created and filled the post. A Local Committee sits because a state appointed its district officer and its members. A One Stop Centre opens on a state’s proposal and a centre’s funding formula. The same dowry offence is investigated by a state police, prosecuted by a state prosecution directorate and tried in a state administered court, and the variations in the field’s delivery that the enforcement record displays are, in constitutional form, variations among the states’ machinery under identical central texts.

The states are also legislators in this field in their own right, within their spheres, and the map should not leave them as mere implementers. State commissions for women, constituted under state Acts, replicate the national Commission’s design of inquiry without adjudication one level down. State legislation and state schemes supplement the central floor in the welfare family, where topping up a centrally sponsored maternity benefit or running a state scheme for the girl child is an ordinary exercise of state policy, and the coverage a pregnant woman actually receives depends on the state she lives in, a fact no reading of the central statute alone will reveal. State rules under the workplace statute govern the district machinery through which its informal sector track runs. And the panchayat reservation that transformed women’s presence in local government was raised by many states from the constitutional one third to one half by state law, an exercise of the same design instinct at the level where the states were free to move first. The federal dimension therefore completes the map’s explanation of unevenness. A field can be uniform in its texts and radically uneven in its operation without any actor disobeying the design, because the design itself assigns the text to one level of government and the machinery to another, and funds them from different budgets at different levels of priority. Reform writing about women’s rights in India that addresses itself only to Parliament addresses the author of the promises. The states keep them, or do not.

Reproductive rights sit at the field’s edge and should be placed on the map at headline accuracy, with their detail reserved to the articles that own them. The Medical Termination of Pregnancy Act, 1971, as amended in 2021, governs the termination of pregnancies by registered medical practitioners on the grounds and within the gestational limits the Act prescribes, and a constitution bench’s jurisprudence on privacy and decisional autonomy, together with decisions such as X v. Government of NCT of Delhi in 2022, which rejected a distinction between married and unmarried women for the purposes of the Act’s twenty to twenty four week category, has progressively read the statute through the woman’s autonomy rather than through the status of her marriage. The surrogacy and assisted reproduction statutes of 2021, treated in an earlier cluster of this series, close the field’s reproductive boundary from the other side, prohibiting commercial surrogacy while permitting its altruistic form within a regulated frame. The placement matters: the same legal order that protects women by prohibiting conduct practised upon them also contains a body of law whose organising question is the woman’s own decision, and the tension between the protective frame and the autonomy frame, visible in the drafting of nearly every statute in this article, is nowhere sharper than here. A complete map records both frames and the friction between them, because the next generation of litigation in this field will be argued in the space the friction defines.

How to navigate the field: four routes, one problem

Navigation begins from the situation, not from the statute book, and the table below is this article’s findable artifact because it performs the routing the whole field presupposes. Five situations recur in the questions put to lawyers, to commissions and to search engines, and each routes to a different combination of statute, forum and order. The table states the primary route for each; the prose around it states the combinations and the cautions. Nothing in a general map of this kind substitutes for advice on an individual matter, and limitation periods and forum rules are unforgiving, so a person facing any of these situations should take advice on the facts of the case; the map’s purpose is narrower, to ensure that the first door knocked on is a door the law has actually built. Readers assembling their own file of the field, the texts, the committee reports, the orders in a matter being followed and the notes that connect them, keep that file more safely in an encrypted private workspace such as VaultBook, where source material and working notes can be held under the reader’s own control, than scattered across devices and inboxes.

Situation	Statute and route	Forum	Character	Limitation	What the forum can order
Violence at home	Protection of Women from Domestic Violence Act, 2005, with the cruelty offence in the criminal code available in parallel	Magistrate, approached directly or through a Protection Officer	Civil, with breach of a protection order itself an offence	No limitation period prescribed for the civil application	Protection, residence, monetary relief, compensation and custody orders
Harassment at work	Sexual Harassment of Women at Workplace (Prevention, Prohibition and Redressal) Act, 2013	Internal Committee, or the Local Committee where no internal committee exists or the complaint is against the employer	Institutional inquiry, neither prosecution nor civil suit	Three months from the incident, extendable by a further three months	Recommendations for action against the respondent, compensation recoverable from salary, interim transfer or leave
A dowry demand	Dowry Prohibition Act, 1961, Sections 3 and 4, with the cruelty and dowry death provisions of the criminal code behind it	Criminal court, on a police report or complaint	Criminal prosecution	Three years for the demand offence under the general limitation rules; no limitation for the graver giving and taking offence	Conviction, imprisonment and fine; the demand is punishable whether or not the marriage takes place
Unequal pay	Code on Wages, 2019, Section 3, successor to the Equal Remuneration Act, 1976	The authority appointed under the Code for claims, through the inspector-cum-facilitator	Claims proceeding with penal backing for contravention	Three years for a claim under the Code	Payment of the wages due, with compensation as the authority directs
Denial of maternity benefit	Code on Social Security, 2020, maternity benefit chapter, successor to the Maternity Benefit Act, 1961 as amended in 2017	The competent authority and inspectorate under the Code	Entitlement claim against the employer	No period prescribed in the maternity chapter; the claim follows the confinement to which it relates	Payment of the benefit for the full entitlement period, and protection of employment during the absence

Three cautions govern the use of the table. The first is that its rows are starting points, not compartments: the violence row ordinarily travels with the dowry row’s offences and with maintenance proceedings elsewhere in the statute book, and a competent adviser reads across the rows before filing anything. The second is that the character column is the one to study, because it predicts the experience of the proceeding: a complainant who expects a civil application to punish will be disappointed by its orders, and a complainant who expects a prosecution to house her will be disappointed by its judgment, and both disappointments are failures of routing rather than of law. The third is that limitation in this field is a trap with uneven teeth: the workplace statute’s three months is short and real, the criminal limitation rules bite on the lesser offences while leaving the graver ones untouched, and the civil statute’s silence on limitation is a feature of its design as a remedy for a continuing situation rather than an invitation to treat time as irrelevant, since delay corrodes the evidence on which every route depends.

A last word on the table’s silences, because an artifact teaches by what it omits as much as by what it contains. The rows carry no row for the harms the field has not organised itself around: the street and the public place, where the criminal offences apply but no dedicated machinery receives the complaint; the digital space, where the information technology statute and the penal provisions on voyeurism and stalking meet harms their drafting anticipated only in part; and the custody of children in a contested separation, which the guardianship statutes govern on principles this map has touched only through the courts’ reading of them. A reader whose situation matches no row has not fallen outside the law. The situation has fallen outside the field’s organised routes, and will be handled, as such situations have been handled throughout the field’s history, by the general law applied without a dedicated instrument, until the accumulation of such cases produces, in the pattern this article has traced from 1961 onwards, first a judicial improvisation and then, a decade or a generation later, a statute.

The verdict on the field, after the map has been drawn, can be stated plainly. As an architecture of prohibition, India’s women’s rights legislation is extensive, hard won and, in its coverage of conduct, substantially complete: there is little serious harm in this territory for which the statute book lacks an offence. As an architecture of relief, it is young, thin and unevenly built, because relief requires the civil and welfare machinery that the legislative process finds hardest to fund and slowest to staff. The field’s history is the history of that imbalance being discovered, one failure at a time, and partially repaired: a civil statute in 2005 for the woman the offences could not house, a workplace statute in 2013 for the harm the offences could not prevent, a welfare entitlement in the same year for the woman no employer could be ordered to pay. Whether the next stratum of the sediment is another offence or another remedy will say more about the direction of Indian social legislation than any single enactment in the field has said so far. For the reader, the operative lesson is the one the opening stated: identify the logic that governs the problem, and the statute, the forum, the standard and the relief will follow from it.

How should a new enactment in this field be read?

By asking of it the five questions this article has asked of the whole field, in the same order. First, which logic governs it: does it prohibit, relieve, oblige an institution, or entitle? The answer predicts the forum and the standard of proof before either is looked up. Second, who must act for the promise to be kept: a police officer, a magistrate, an employer, a district administration, a payment system, and how well is that actor staffed and funded in the places where the promise is to operate? Third, what does the claimant receive if she succeeds, in her own hands, as distinct from what happens to the respondent? Fourth, what clock runs against her, and is the period proportioned to the reality that the harm this field addresses usually occurs inside continuing relationships, where the decision to complain is itself delayed by dependence? Fifth, what has the enactment displaced: which older provision has it repealed, renumbered or quietly superseded, and do the commentaries and forms in circulation reflect the displacement? An enactment that survives those five questions with good answers is a rarity in this field’s history, and the habit of asking them is worth more to a reader than any single statute summary, because the field will go on being amended, consolidated and renamed, and the map that cannot be redrawn from its own principles expires with the next amendment. The four families are those principles. The statutes are their applications, and applications change.

Frequently Asked Questions

Q: Which laws protect women’s rights in India?

The protection is spread across four families of statutes rather than one code. The criminal family includes the Dowry Prohibition Act, 1961, the cruelty and dowry death provisions of the penal code, the sexual offences rewritten by the Criminal Law (Amendment) Act, 2013, and special statutes on sex selection, indecent representation, sati and trafficking. The civil family consists principally of the Protection of Women from Domestic Violence Act, 2005. The workplace family comprises the Sexual Harassment of Women at Workplace Act, 2013 and, after the labour codes commenced, the Code on Wages and the Code on Social Security, which absorbed the equal remuneration and maternity benefit statutes. The welfare family’s anchor is the maternity entitlement in the National Food Security Act, 2013. Constitutional provisions, Articles 14, 15, 16, 39, 42 and 51A(e), stand behind all four families.

Q: What does Article 15(3) allow the state to do?

Article 15(3) provides that nothing in Article 15 prevents the state from making any special provision for women and children. It is an enabling provision, not a right in itself: it takes legislation and schemes framed exclusively or preferentially for women outside the prohibition on sex discrimination in Article 15(1), so that a programme reserved for women is not struck down for treating men differently. The Supreme Court has given the provision a wide reading, treating it as an affirmative mandate rather than a grudging exception, and it has been relied on to sustain provisions ranging from reservations of seats and posts to procedural protections in criminal law. Its width has limits of context: the provision authorises special provisions for women, and a measure that is in substance a penalty or a stereotype dressed as protection can still fail constitutional scrutiny on other grounds.

Q: Are women’s rights laws criminal or civil?

Both, in unequal proportions. Counted by statute and by section, the field is predominantly criminal: the dowry prohibition, the cruelty and dowry death offences, the sexual offences and the special statutes on trafficking, sex selection and indecent representation all work by prosecution and punishment. The civil part of the field is concentrated in a single statute, the Protection of Women from Domestic Violence Act, 2005, whose orders protect residence, grant monetary relief and compensation and settle custody without convicting anyone, although breach of a protection order under it is an offence. A third category sits between the two: the workplace statute of 2013 runs an institutional inquiry that is neither a prosecution nor a civil suit. A fourth category is neither: the maternity entitlement in the National Food Security Act, 2013 is a welfare claim that requires no wrongdoer at all.

Q: Which women’s statutes were repealed into the labour codes?

Two. The Equal Remuneration Act, 1976 was subsumed by the Code on Wages, 2019, whose Section 3 restates the equal pay duty as a prohibition on discrimination on the ground of gender in wages and recruitment. The Maternity Benefit Act, 1961, as amended in 2017, was subsumed by the Code on Social Security, 2020, whose maternity benefit chapter carries the twenty six week entitlement and the creche obligation into the code. Both repeals took effect when the four labour codes commenced on 21 November 2025, alongside the repeal of twenty seven other labour statutes, with savings for action already taken under the repealed Acts. The Sexual Harassment of Women at Workplace Act, 2013 was not among the subsumed statutes and continues in force as a free-standing Act, as do the domestic violence and dowry statutes.

Q: What powers does the National Commission for Women have?

The Commission, constituted in January 1992 under the National Commission for Women Act, 1990, may investigate and examine the safeguards for women under the Constitution and the laws, review legislation and recommend amendments, take up violations with the appropriate authorities, and take suo motu notice of the deprivation of women’s rights, the non-implementation of protective laws and non-compliance with policy decisions intended for women’s welfare. While investigating, it has the powers of a civil court to summon and enforce attendance, require the production of documents and receive evidence on affidavit. Its powers stop at inquiry and recommendation. It cannot decide a case, award compensation, order an arrest or bind any authority by its conclusions. A complaint to the Commission is therefore a channel of investigation and pressure that runs alongside a court proceeding, not a substitute for one.

Q: Do these laws apply to transgender persons?

The protective statutes are drafted around women, and their application to transgender persons turns on that drafting. The domestic violence statute protects a woman in a domestic relationship, the workplace statute protects a woman employee, and the penal code’s sexual offences are, in the main, framed around a woman victim and a male perpetrator. A transgender woman seeking the protection of these statutes therefore faces a threshold question about the statutory word woman that the case law has answered unevenly and that has not been settled by a comprehensive ruling. Parliament’s chosen route was separate legislation: the Transgender Persons (Protection of Rights) Act, 2019 prohibits discrimination against transgender persons and makes offences against them, including sexual abuse, punishable under that Act, with penalties markedly lower than the corresponding offences in the penal code. That asymmetry is one of the standing criticisms of the present settlement.

Q: Which women’s laws reach the unorganised sector?

The laws whose operation does not depend on an employer reach furthest. The criminal offences apply everywhere, in every kind of household and work. The domestic violence statute governs the home, not the workplace, so it covers women in informal work on the same terms as any other woman. The maternity entitlement in Section 4 of the National Food Security Act, 2013 was designed for this sector expressly, paying a benefit of not less than six thousand rupees to pregnant women and lactating mothers who are outside regular government employment and similar benefit schemes, delivered principally through the maternity benefit scheme. The workplace statute of 2013 reaches domestic workers through the Local Committee route. The Code on Social Security, 2020 extends its frame, on paper, to unorganised, gig and platform workers, although the schemes that would give that frame content for informal women workers remain the incomplete part of the design.

Q: How do personal laws interact with women’s rights statutes?

The two operate on different questions and meet inside almost every real dispute. The protective statutes are laws of general application: the penal code, the dowry prohibition, the domestic violence statute and the workplace statute govern conduct and relief for women of every community, because they do not depend on religious status. The personal laws govern the adjacent territory of status: the validity and dissolution of marriage, succession, adoption and guardianship, through the Hindu codes of 1955 and 1956 for Hindus, Buddhists, Jains and Sikhs, and through community specific statutes and case law for others, including the Muslim Women (Protection of Rights on Divorce) Act, 1986. Maintenance has long had a secular route in the procedure codes alongside the personal law. In practice a single dispute moves between the fields repeatedly, claiming residence and monetary relief under the general civil statute while the marriage and the property are governed by the personal law of the parties.

Q: What is the difference between Section 498A and the Domestic Violence Act?

The two provisions answer the same facts with different instruments. Section 498A of the Indian Penal Code, carried into Section 85 of the Bharatiya Nyaya Sanhita, 2023, is a criminal offence: cruelty by a husband or his relatives, punishable with imprisonment of up to three years and a fine, tried on proof beyond reasonable doubt, and ending, if it succeeds, in a conviction that orders no housing, maintenance or compensation for the woman as its principal outcome. The Protection of Women from Domestic Violence Act, 2005 is a civil statute: it covers a wider range of relationships and of conduct, including economic abuse, decides on the civil standard, and produces protection, residence, monetary relief, compensation and custody orders addressed to the woman’s position. Breach of its protection order is itself an offence. The routes are cumulative, so the same conduct can support a prosecution and a civil application at the same time.

Q: When did maternity leave become twenty six weeks, and which law grants it after the labour codes?

The Maternity Benefit (Amendment) Act, 2017 raised the paid entitlement from twelve weeks to twenty six weeks, with not more than eight weeks falling before the expected date of delivery, and the amended provisions commenced on 1 April 2017, with the creche requirement for establishments of fifty or more employees following on 1 July 2017. The same amendment added leave for adopting mothers and commissioning mothers and a work from home provision by agreement. The parent statute, the Maternity Benefit Act, 1961, was among the twenty nine labour statutes repealed when the four labour codes commenced on 21 November 2025. The entitlement itself continues in the maternity benefit chapter of the Code on Social Security, 2020, which is the governing text for a maternity claim after the codes, carrying the twenty six weeks, the adopting and commissioning mother provisions and the creche obligation with it.

Q: Can a daughter claim an equal share in ancestral property after the 2005 Hindu Succession amendment?

In a Hindu joint family governed by the Mitakshara school, yes, as a coparcener. The Hindu Succession (Amendment) Act, 2005 substituted Section 6 of the Hindu Succession Act, 1956 so that the daughter of a coparcener becomes a coparcener by birth in her own right, in the same manner as a son, with the same rights in the coparcenary property and the same liabilities. She can therefore seek partition of ancestral coparcenary property rather than wait on a succession that opens on a death. In Vineeta Sharma v. Rakesh Sharma, in 2020, the Supreme Court settled that the right is conferred by birth and does not depend on the father having been alive when the amendment commenced in September 2005. The right concerns coparcenary property; a father’s self-acquired property continues to pass by his will or by the ordinary rules of succession.

Q: What is the difference between a protection order and a residence order under the Domestic Violence Act?

The two orders do different work, and the distinction is the civil statute in miniature. A protection order under Section 18 is directed at the respondent’s conduct: it prohibits him from committing or aiding any act of domestic violence, from entering the aggrieved woman’s workplace or her child’s school, from attempting to communicate with her, from alienating assets the couple holds or operates, and from causing violence to her dependents or to those who assist her. A residence order under Section 19 is directed at the household itself: it restrains the respondent from dispossessing the woman from the shared household or disturbing her possession of it, can direct him to remove himself from the household, can bar him and his relatives from the portion where she resides, and can direct him to secure alternative accommodation of the same standard or to pay its rent. The protection order polices behaviour; the residence order secures the roof, independently of who owns or rents it.

Q: Is there a time limit for filing a complaint under the Domestic Violence Act?

The Act prescribes none. The Protection of Women from Domestic Violence Act, 2005 contains no limitation period for an application under Section 12, and the courts have treated the wrong it addresses as capable of being a continuing one, so an application is not defeated merely because the earliest acts of abuse lie in the past. What the statute does prescribe is speed after filing: the Magistrate is to fix the first date of hearing ordinarily within three days of receiving the application and to endeavour to dispose of it within sixty days of that hearing, directions honoured unevenly in practice. Delay still carries costs even where it is no bar. Evidence of older conduct is harder to assemble, interim protection is directed at the situation the applicant is in, and a long unexplained interval can affect how a court assesses the application’s urgency and particulars.

Q: What can an Internal Committee order under the POSH Act?

The Internal Committee constituted under the Sexual Harassment of Women at Workplace Act, 2013 inquires into a complaint with the powers of a civil court to summon witnesses and require documents. During the inquiry it can recommend interim measures, including transferring the complainant or the respondent, granting the complainant leave of up to three months in addition to her ordinary leave, and restraining the respondent from reporting on her work. On completing the inquiry, it recommends action against the respondent in accordance with the service rules of the organisation, or as misconduct where no rules exist, and it can recommend that compensation, assessed on factors the Act lists including mental trauma, loss of career opportunity, medical expenses and the respondent’s income, be deducted from the respondent’s salary. The employer is required to act on the recommendations within sixty days. The committee recommends; it does not itself convict, fine or dismiss.

Q: Does the PCPNDT Act cover sex selection before conception?

Yes, and the extension is written into the Act’s title. The statute was enacted in 1994 as the Pre-natal Diagnostic Techniques (Regulation and Prevention of Misuse) Act, aimed at diagnostic techniques used on a pregnant woman. The amendment of 2003 renamed it the Pre-Conception and Pre-Natal Diagnostic Techniques (Prohibition of Sex Selection) Act and widened its reach on both sides of conception: Section 3A prohibits any person, including a specialist, from conducting or causing to be conducted or aiding in any manner the technique of sex selection on a woman, a man or tissue or fluid from either, and the Act regulates the clinics, laboratories and practitioners who offer the relevant services through registration, record keeping and prohibition, backed by imprisonment and fine, with suspension of medical registration on conviction. Its enforcement rests on those records rather than on a complainant, since the parties to the selection ordinarily consent to it.

Q: Why does the maternity benefit of a woman in informal work come from the state while an employee’s benefit comes from her employer?

Because the two benefits rest on different designs, and the designs follow the payer. The employee’s maternity benefit is framed as a liability of her employer: the maternity benefit statutes, and the maternity chapter of the Code on Social Security that succeeded them, oblige the employer to pay wages through the leave period, on the theory that the cost of reproduction should fall on the enterprise that engages the woman’s labour. A woman in informal work, in agriculture, in domestic work or in her own household’s economy, has no employer against whom such a liability could run, so her benefit can only be a charge on public funds. That is the design of Section 4 of the National Food Security Act, 2013, which entitles a pregnant woman or lactating mother outside regular government employment and similar schemes to a meal through the anganwadi and a maternity benefit of not less than six thousand rupees. The consequence is the quantum gap this article notes: full wages where an employer pays, a flat cash sum where the state does.

Q: Why did the workplace harassment law begin as a Supreme Court judgment rather than a statute?

Because Parliament had legislated nothing on the subject and a case forced the question. In Vishaka v. State of Rajasthan, in 1997, a public interest petition arising from the gang rape of a grassroots social worker in Rajasthan brought workplace sexual harassment before the Supreme Court. The Court held that such harassment violates Articles 14, 15, 19(1)(g) and 21 of the Constitution, drew the content of the prohibition from the Convention on the Elimination of All Forms of Discrimination against Women, which India ratified in 1993, and laid down guidelines for prevention and redress, including complaint committees headed by a woman with an external member, declaring the guidelines binding as law under Article 141 until Parliament legislated. Parliament took sixteen years to do so. The Sexual Harassment of Women at Workplace Act, 2013 largely codified and widened the Vishaka scheme, converting a judicial floor into a statutory duty owed by every employer.

Q: Can a dowry demand made before marriage be punished if the marriage never takes place?

Yes. Section 4 of the Dowry Prohibition Act, 1961 penalises the demanding of dowry, directly or indirectly, from the parents or other relatives or the guardian of a bride or bridegroom, and the offence is complete on the demand itself; it does not require that a marriage be solemnised, that the demand be accepted or that any property actually change hands. A demand made during negotiations that then collapse is therefore punishable on the statute’s terms, which is deliberate design: the Act aims at the transaction’s opening move, not only at its completion. The graver offence of giving or taking dowry sits in Section 3, carrying a minimum sentence of five years. Because the demand offence falls in the lower punishment band, the general rules of limitation for taking cognizance, in Section 468 of the Code of Criminal Procedure, can bar a stale prosecution of a bare demand, which is one reason the demand is so rarely prosecuted on its own.

Q: What reservation did the 106th Amendment create for women, and when does it operate?

The Constitution (One Hundred and Sixth Amendment) Act, 2023, the Nari Shakti Vandan Adhiniyam, inserted Articles 330A and 332A reserving, as nearly as may be, one third of the seats in the Lok Sabha and in every state legislative assembly for women, including one third of the seats reserved for Scheduled Castes and Scheduled Tribes within that share. It is the legislative counterpart of the reservation the Seventy-third and Seventy-fourth Amendments created in panchayats and municipalities in 1992, which many states have since raised to one half. The parliamentary reservation is deliberately not immediate in its operation: it takes effect after the first census conducted following the amendment’s commencement has been published and a delimitation exercise has been undertaken on its basis, and the reservation then runs for fifteen years from its commencement, extendable by Parliament, with the reserved seats rotating between constituencies as delimitation determines.

Q: Where does a complaint go when a workplace has no Internal Committee?

To the Local Committee, which every district officer is required to constitute at district level under the Sexual Harassment of Women at Workplace Act, 2013. The Local Committee receives complaints from establishments where an Internal Committee has not been constituted because the workplace has fewer than ten employees, complaints against the employer himself, and complaints from domestic workers, whose workplace is a home and who could never have an internal committee. Nodal officers designated in blocks, talukas and wards receive complaints and forward them to the Local Committee within seven days. The Local Committee inquires with the same civil court powers the internal body holds, can recommend the same interim relief and the same compensation recovery, and its recommendations go to the employer or, for a domestic worker’s household, to the district officer for action. The route exists on the statute book throughout India; its visibility to the complainants it was built for is the weaker part of the design.