Introduction
It is one of the quieter tragedies of Indian family law that almost no one who passes through it, neither the wife who alleges cruelty, nor the husband who denies it, nor the extended family drawn in as collateral respondents, emerges from the process feeling that justice was either swift or proportionate. A woman genuinely subjected to cruelty, harassment or economic deprivation within marriage frequently finds that the legal remedies available to her are slow, evidentially demanding and, even when finally granted, poorly enforced. At the same time, a man wrongly implicated in a matrimonial complaint discovers that the very features that make the law protective (its cognizable and non-bailable character, the presumption of urgency attached to a cruelty complaint, and the reluctance of courts to appear unsympathetic to a complainant) make it correspondingly difficult to establish his innocence without years of litigation, reputational damage and financial cost, even where he is eventually vindicated.
This paper proceeds from the premise that there are not two separate problems to be solved by pulling the law in opposite directions, but a single structural failure. A legal architecture built primarily around a criminal complaint, a maintenance petition or a divorce suit, each requiring years of trial before final relief, cannot adequately serve either the protective or the adjudicatory function it is meant to perform.
The statutory architecture of matrimonial remedies
The Indian matrimonial litigant confronts not one remedy but several, often pursued simultaneously and before different fora. Cruelty is addressed criminally under section 498A of the Indian Penal Code, 1860, now sections 85 and 86 of the Bharatiya Nyaya Sanhita, 2023,1 and civilly through the Protection of Women from Domestic Violence Act, 2005, which additionally provides for residence orders, protection orders and monetary relief. Maintenance may be claimed under section 125 of the Code of Criminal Procedure, 1973, now section 144 of the Bharatiya Nagarik Suraksha Sanhita, 2023,2 and, for Hindu couples, under sections 24 and 25 of the Hindu Marriage Act, 1955. Divorce itself is sought under the relevant personal law, such as the Hindu Marriage Act, 1955, the Special Marriage Act, 1954, or the applicable religious or customary law, each with its own grounds, procedural timelines and evidentiary thresholds.
The consequence is that a single marital breakdown routinely generates four or five parallel proceedings: a criminal complaint under section 498A, a domestic violence petition, a maintenance application, a custody petition where there are children, and the divorce suit itself. Each proceeding has its own timeline, its own standard of proof and its own court. For a complainant, this multiplicity can appear to offer a menu of protection; for a respondent, it can appear to multiply the opportunities for harassment; and for the system as a whole, it multiplies the points at which delay, non-appearance and procedural manoeuvring can occur. The remainder of this paper examines how this architecture performs, and fails, in practice.
The genuine victim’s struggle for relief
For a woman who has genuinely suffered cruelty, whether physical, verbal, economic or the sustained mental cruelty that the courts have increasingly recognised as a valid ground for relief, the path to relief is neither swift nor certain. Mental and economic cruelty, unlike a visible physical injury, is inherently difficult to prove through documentary and medical evidence; it typically depends on oral testimony, which is tested, and often diluted, through prolonged cross-examination years after the events complained of. Interim maintenance, intended precisely to provide immediate relief pending final adjudication, is itself frequently contested at length, with respondents filing objections regarding income, assets and need that can take months or years to resolve.
Even when an order is eventually secured, its practical value depends entirely on compliance. It is not uncommon for a woman to obtain an order for interim maintenance only to find that actual payment requires a fresh round of litigation: an execution petition, a contempt application, or an application under section 125(3) of the Code of Criminal Procedure (now section 144(3) of the Bharatiya Nagarik Suraksha Sanhita) for non-payment, each of which is, once again, a proceeding requiring notice, hearing and time. The genuinely aggrieved spouse is thus placed in the position of having to litigate not merely for substantive relief but for the enforcement of relief already granted, a burden that falls disproportionately on those least able to sustain protracted litigation.
There is, in addition, a subtler cost. Family and social pressure to settle, to withdraw complaints, or to avoid “dragging the family through court” weighs heavily, particularly where the woman remains economically dependent on the very household against which she has complained. Genuine claims are often abandoned not because they lack merit but because the cost of pursuing them to conclusion, social, financial and emotional, exceeds what the litigant can bear.
The wrongly accused: vindication at a price
The converse difficulty is by now well documented in Indian jurisprudence. As early as 2005, in Sushil Kumar Sharma v. Union of India,3 the Supreme Court, while upholding the constitutional validity of section 498A, observed that “by misuse of the provision a new legal terrorism can be unleashed”, and that the provision’s constitutionality did not “give a licence to unscrupulous persons to wreck personal vendetta or unleash harassment”. The Court treated this not as a reason to strike the provision down but as a reason for the courts, until the legislature acted, to “take care of the situation within the existing frame work”. The observation has since been echoed and expanded upon repeatedly.
In Preeti Gupta v. State of Jharkhand,4 the Court took particular note of the tendency to implicate the husband’s relatives along with him, observing that “[t]he tendency of implicating husband and all his immediate relations is also not uncommon”, and that the allegations in such complaints “are required to be scrutinized with great care and circumspection”, particularly where the husband’s relations had been living in different cities and had never, or rarely, visited the place where the complainant resided. Four years later, in Arnesh Kumar v. State of Bihar,5 the Court confronted the practice of automatic arrest upon registration of a section 498A complaint. Noting National Crime Records Bureau data that showed a conviction rate of only 15 per cent despite a very high volume of arrests, it directed police officers not to arrest automatically but to satisfy themselves of the necessity for arrest against a checklist of the conditions in section 41(1)(b)(ii) of the Code of Criminal Procedure, and directed magistrates to peruse the officer’s reasons and record their satisfaction before authorising detention.
The difficulty is that even where these safeguards operate as intended, and an accused is never arrested, is later acquitted, or has the proceedings quashed, the underlying litigation itself imposes a cost that vindication does not erase. An offence under section 498A is non-compoundable in most States, meaning that even where the parties wish to settle, they must approach the High Court under its inherent jurisdiction to have the proceedings quashed,6 a further proceeding with its own delay and expense. Parallel proceedings for maintenance, custody and divorce continue independently of the criminal case and are not automatically resolved by an acquittal. The result is that an accused who is ultimately found to have done nothing wrong may nonetheless have spent years defending multiple simultaneous cases, incurred substantial legal costs, lost employment or professional standing, and endured the reputational consequences of a criminal charge that a court later found to be unfounded. Vindication, in other words, arrives, but often at a cost that the system does little to compensate.
Landmark precedents and the limits of blanket application
The tension between protecting genuine complainants and preventing misuse is nowhere better illustrated than in the short and troubled history of the Family Welfare Committee mechanism. In Rajesh Sharma v. State of Uttar Pradesh,7 a two-judge Bench, responding to what it described as the “serious concern” that a large number of cases continued to be filed under section 498A, directed District Legal Services Authorities to constitute Family Welfare Committees, drawn from para-legal volunteers, social workers, retired persons and other suitable citizens, to which every section 498A complaint was to be referred, and directed that, until a committee had reported, no arrest should normally be made. The intention, plainly, was protective of the accused: to introduce a screening mechanism before the coercive machinery of arrest was engaged.
Little more than a year later, a three-judge Bench in Social Action Forum for Manav Adhikar v. Union of India8 held these directions impermissible. It observed that they introduced “a third agency which has nothing to do with the Code”, with committees “empowered to suggest a report failing which no arrest can be made”, and held that the direction in paragraph 19(i) was “not in accord with the statutory framework”, having said that some of the directions had “the potential to enter into the legislative field”. When issuing notice in the matter in October 2017, the Court had already recorded that it was “not in agreement with the decision rendered in Rajesh Sharma” because it was “disposed to think that it really curtails the rights of the women who are harassed under Section 498A”.
The episode is instructive for reasons beyond its immediate subject matter. A guideline crafted with the entirely legitimate aim of preventing misuse, on closer examination, imposed a uniform delay on every complainant, including the substantial number whose complaints were genuine and urgent. Equally, the case-by-case scrutiny urged in Preeti Gupta and the arrest checklist required in Arnesh Kumar, while more calibrated than a committee mechanism, still depend for their effectiveness on the diligence of individual police officers and magistrates across thousands of police stations and courts, with compliance that is, in practice, uneven. The broader lesson is that a fact-sensitive category of dispute, in which the same statutory provision can protect a woman facing genuine violence and can also be invoked strategically in a contested divorce, resists a single procedural rule applied uniformly from the top and expected to filter down predictably to every trial court and police station in the country.
The enforcement deficit: orders without teeth
Perhaps the least discussed but most corrosive feature of matrimonial litigation is the gap between the grant of relief and its actual enforcement. An order for maintenance, a protection order under the Domestic Violence Act or a residence order is only as effective as the mechanism available to enforce it, and that mechanism in India remains comparatively weak. A defaulting spouse can decline to pay, understate income, change employment, transfer assets to relatives or simply relocate, leaving the party in whose favour the order was passed to initiate fresh execution proceedings under Order XXI of the Code of Civil Procedure, 1908, a complaint for breach of a protection order under section 31 of the Domestic Violence Act, or contempt proceedings before the very court that passed the original order. Each avenue requires the aggrieved party to re-enter the litigation process, with fresh notice, fresh hearings and fresh delay, in order to obtain compliance with an order that ought, in principle, to require no further adjudication at all.
This enforcement deficit produces a curious asymmetry: obtaining an order is treated as the end point of litigation, when in practice it is often only the midpoint. A husband determined to avoid payment has, as a practical matter, numerous avenues open to him, such as repeated adjournments, revision petitions against interim orders, appeals that stay execution, and the simple fact that contempt is rarely pursued to conviction because courts are, understandably, reluctant to imprison a party over what is fundamentally a civil obligation. The cumulative effect is that even a legally sound and well-reasoned order can be rendered practically meaningless for years, while the party entitled to relief bears the cost of repeated enforcement litigation on top of the cost of the original proceeding.
Conclusion
The Indian matrimonial justice system is often discussed as though it presents a choice between protecting women and protecting the falsely accused, with landmark judgments read as swinging the pendulum first one way and then the other. This paper has argued that the framing is mistaken. The genuine victim who cannot secure timely, enforceable relief and the wrongly accused respondent who cannot extricate himself from years of multiplying litigation are both casualties of the same underlying weaknesses: excessive reliance on criminal and quasi-criminal process for what are fundamentally evidentiary and relational disputes, and an enforcement mechanism too weak to give orders practical effect. Blanket judicial guidelines, however well intentioned, have struggled to correct these weaknesses precisely because they are applied uniformly to a category of disputes that is inherently fact-specific. A more durable reform would treat the protection of genuine victims and the prevention of misuse not as opposing interests to be balanced against each other, but as two symptoms of the same structural problems, to be addressed through the same set of institutional fixes: speed, enforceability, and case-specific rather than blanket procedure.
*****
Footnotes
1. The Indian Penal Code, 1860, No. 45, Acts of Parliament, 1860 (India), § 498A. The Code has been replaced, with effect from July 1, 2024, by the Bharatiya Nyaya Sanhita, 2023, and the substance of section 498A now appears in The Bharatiya Nyaya Sanhita, 2023, No. 45, Acts of Parliament, 2023 (India), §§ 85–86.
2. The Code of Criminal Procedure, 1973, No. 2, Acts of Parliament, 1974 (India), § 125, now The Bharatiya Nagarik Suraksha Sanhita, 2023, No. 46, Acts of Parliament, 2023 (India), § 144. Maintenance may also be claimed under The Hindu Marriage Act, 1955, No. 25, Acts of Parliament, 1955 (India), § 24 (maintenance pendente lite) and § 25 (permanent alimony and maintenance), or under the personal law applicable to the parties.
3. Sushil Kumar Sharma v. Union of India, (2005) 6 SCC 281 (India).
4. Preeti Gupta v. State of Jharkhand, (2010) 7 SCC 667, ¶ 35 (India).
5. Arnesh Kumar v. State of Bihar, (2014) 8 SCC 273 (India). Section 41 of the Code of Criminal Procedure, 1973, construed there, now corresponds to The Bharatiya Nagarik Suraksha Sanhita, 2023, No. 46, Acts of Parliament, 2023 (India), § 35.
6. See Social Action Forum for Manav Adhikar v. Union of India, (2018) 10 SCC 443, ¶¶ 34, 40 (India) (on settlement, a non-compoundable proceeding may be quashed by the High Court under The Code of Criminal Procedure, 1973, No. 2, Acts of Parliament, 1974 (India), § 482, now The Bharatiya Nagarik Suraksha Sanhita, 2023, No. 46, Acts of Parliament, 2023 (India), § 528).
7. Rajesh Sharma v. State of Uttar Pradesh, (2018) 10 SCC 472, ¶ 19(i) (India).
8. Social Action Forum, supra note 6, ¶¶ 16, 34, 36, 38, 39.