Cruelty As Ground For Divorce: Doctrinal Paradox At Heart Of Hindu Matrimonial Law
Anubha Shukla
18 Sept 2026 11:00 AM IST

Cruelty is, without question, the single most litigated fault ground under the Hindu Marriage Act, 1955. It is also the least defined. Section 13(1)(ia) permits a spouse to seek divorce if the other has “treated the petitioner with cruelty”, but Parliament never told courts what cruelty actually means. Fifty years after the concept was written into the divorce provision by the Marriage Laws (Amendment) Act, 1976, that silence has produced something genuinely paradoxical: a ground of divorce that has become both the most useful tool in matrimonial litigation and the least predictable one. The very quality that makes cruelty valuable, its open texture, is the same quality that has made it doctrinally incoherent. That paradox, rather than any single case or statute, is really what deserves attention.
From Sanskara to Statute
Under classical Hindu law, marriage was never meant to be dissolved. It was a Sanskara, a sacrament undertaken for dharma, procreation and companionship, and regarded as binding across lifetimes. Cruelty existed as a social wrong, but it gave an aggrieved spouse nothing more than a right to live separately when there was danger to life. The Hindu Marriage Act, 1955 broke with that tradition in a limited way. In its original form, cruelty was a ground for judicial separation under Section 10, not for divorce under Section 13. The legislature wanted to acknowledge marital suffering without abandoning the ideal of permanence.
That changed with the 1976 amendment. Cruelty became a direct ground for divorce, and, just as significantly, the requirement that the cruelty create a reasonable apprehension of injury to life or limb was dropped. This one deletion pushed the law from an objective, physical test toward a subjective test of marital well being. It also handed courts an almost impossible task: deciding, case by case, where the ordinary wear and tear of married life ends, and actionable cruelty begins, with no legislative yardstick to guide them.
The Dastane Foundations
The modern architecture of cruelty was largely built by the Supreme Court in N.G. Dastane v. S. Dastane (1975), decided just before the 1976 amendment but still treated as foundational. Dastane did three things that continue to shape every cruelty case decided today. It held that matrimonial proceedings are civil in nature, so the standard of proof is the preponderance of probabilities rather than proof beyond reasonable doubt, a point courts continue to rely on even now, as seen in a recent Himachal Pradesh High Court ruling that reiterated the Dastane standard while dismissing a husband's cruelty petition. It rejected an objective “reasonable man” test in favour of assessing cruelty from the petitioner's own sensitivities. It developed the doctrine of condonation under Section 23(1)(b), later refined by the Kerala High Court's observation that condonation does not erase a matrimonial wrong the way a presidential pardon would; the wrong is, in the court's words, merely obscured, not obliterated, and can revive.
These three pillars gave cruelty its procedural skeleton. What they did not give it was a definition. That gap is where the paradox begins to take shape, because the very tools that made cruelty workable, a lower standard of proof and a subjective test tied to the petitioner's own experience, are also what allowed the concept to expand almost without limit in the decades that followed.
The Rise of Mental Cruelty
As physical violence receded as the dominant fact pattern in matrimonial litigation, mental cruelty took its place. In V. Bhagat v. D. Bhagat (1994), the Supreme Court described mental cruelty as a state of mind incapable of being measured by any single yardstick, built instead out of a sustained course of conduct assessed cumulatively rather than incident by incident. This “cumulative effect” test is intuitively fair. It recognises that a marriage can be destroyed by a hundred small cruelties that would mean nothing individually. But it also means that no two courts, hearing similar facts, are bound to reach the same conclusion, because the assessment depends entirely on the totality of a relationship that only the parties themselves fully know.
Samar Ghosh v. Jaya Ghosh (2007) tried to bring some order to this. The Supreme Court, dealing with the collapse of a marriage between two senior bureaucrats, listed fourteen illustrations of conduct that might amount to mental cruelty, ranging from a unilateral refusal of sexual relations to a unilateral decision against having children. Crucially, the Court insisted these examples were illustrative and not exhaustive, and that a marriage reduced to nothing but a legal tie could itself be a source of cruelty. This is precisely the paradox in miniature. Samar Ghosh was meant to standardise the law, yet by expressly refusing to be prescriptive, it left the underlying subjectivity untouched. Lower courts got a checklist, but not a rule.
New Faces of an Old Ground
Cruelty has kept expanding to absorb new social realities. Economic cruelty is now well recognised; the Supreme Court held as far back as Shobha Rani v. Madhukar Reddi (1988) that persistent dowry demands amount to cruelty even without physical violence, and the Delhi High Court has since gone further, holding that mental cruelty is wide enough to include the financial instability inflicted on a spouse. Digital life has produced its own variant. The Telangana High Court recently observed that depriving a spouse of access to Facebook or Instagram, in a manner that damages reputation or social standing, may itself amount to cruelty, a striking illustration of how far the doctrine has travelled from its origins in physical safety.
Each of these extensions is defensible on its own facts. Taken together, they show a doctrine with no outer boundary, only an ever-growing list of behaviours that some court, somewhere, has found intolerable enough to qualify.
Where Civil Meets Criminal
The paradox deepens at the intersection of matrimonial and criminal law. Section 13(1)(ia) of the HMA is gender neutral; either spouse may allege cruelty. Section 498A of the erstwhile Indian Penal Code, now retained in near-identical form as Section 85 of the Bharatiya Nyaya Sanhita, 2023, protects only a married woman against cruelty by her husband or his relatives. These two provisions were designed for different purposes, one to dissolve a marriage, the other to punish an offence, yet courts have increasingly folded the criminal provision back into the civil one. In K. Srinivas Rao v. D.A. Deepa (2013), the Supreme Court held that filing false or exaggerated criminal complaints under Section 498A can itself amount to mental cruelty entitling the husband to a divorce, a principle since applied repeatedly, including in a matter where the Orissa High Court granted divorce after a wife lodged forty-five FIRs against her husband. Section 85 BNS, meanwhile, remains the subject of its own reform debate, with the Supreme Court recently urging Parliament to revisit its drafting given how frequently it is invoked and how rarely it results in conviction.
A provision meant as a shield against dowry violence has, in a significant number of cases, become evidence of the very cruelty it was designed to punish. That is not a flaw in either statute individually. It is a product of asking one word, cruelty, to do civil and criminal work at the same time.
Cruelty as a Proxy for a Dead Marriage
Perhaps the clearest expression of the doctrinal paradox lies in how cruelty has been used to smuggle in a ground Parliament has never enacted: irretrievable breakdown of marriage. The Law Commission recommended this as an independent ground as early as 1978. Parliament has never acted on it. In its place, the Supreme Court has repeatedly held that a marriage kept alive only in name, with no prospect of reconciliation, itself inflicts cruelty on both spouses, most notably in Naveen Kohli v. Neelu Kohli (2006), and has used its extraordinary powers under Article 142 to dissolve such marriages directly, as the Constitution Bench did in Shilpa Sailesh v. Varun Sreenivasan (2023). A recent LiveLaw analysis of this line of cases frames the resulting problem well: irretrievable breakdown now depends less on the facts of a marriage than on which court happens to be hearing it, since only the Supreme Court can invoke Article 142.
This produces a genuinely strange result. A marriage the Supreme Court would readily call dead can remain legally alive in a family court simply because the petitioner could not prove a specific act of fault-based cruelty and had no means of reaching the apex court. As one recent commentary on the subject put it, the absence of irretrievable breakdown as a statutory ground has created a troubling inconsistency, where relief is often available only to litigants who can reach the Supreme Court while similarly placed parties remain trapped below.
The Paradox Restated
Put all of this together and the shape of the problem becomes clear. Cruelty has been made flexible precisely so that the law can keep pace with changing marital realities, from dowry harassment to digital surveillance to the simple fact of a marriage that has quietly died. That flexibility is not an accident; it is the deliberate consequence of Parliament's refusal to define the term and the judiciary's willingness to fill that vacuum on a case-by-case basis. But the same flexibility has meant that outcomes turn on the sensibilities of the particular bench, the social background of the parties, and even the level of court a litigant can afford to approach. A doctrine built for adaptability has, in practice, produced unpredictability, and a ground of fault has quietly become the judiciary's preferred instrument for recognising marriages that have simply ceased to exist.
This is the doctrinal paradox that any serious study of cruelty under the HMA must confront: the flexibility that has kept the law relevant is inseparable from the uncertainty that now defines it. Legislative recognition of irretrievable breakdown as an independent ground, alongside clearer statutory guidance on what constitutes mental cruelty, would not remove the subjectivity that cruelty inherently carries. But it would at least stop courts from using one doctrine to do the work of two, and would restore some of the predictability that litigants on both sides of a broken marriage are currently denied.
Author is a third-year Law student at Gujarat National Law University, Gandhinagar. Views are personal.

