Burden Of Proving Cruelty in Matrimonial Litigation

Our courts have said, repeatedly and without much dissent, that a false allegation raised by either spouse of adultery, of dowry harassment, or through a fabricated criminal complaint is itself cruelty, independent of whatever other physical or mental cruelty has been pleaded. What has troubled me across fifteen years of handling these disputes is a narrower and more practical question: what degree of proof does the law actually demand before such allegations can be said to be established, when in the overwhelming majority of cases there is neither an independent witness nor documentary evidence in support? And critically, does that evidentiary bar vary depending which forum a spouse chooses, and if so, why, given that intimate partner violence is specifically characterized by its invisibility? 

The civil standard, and why it matters

The starting point is that matrimonial causes are not decided on the standard of proof in criminal standard. In Dastane v. Dastane (1975), the Supreme Court held that a matrimonial petition, like any civil case, may be proved by a preponderance of probabilities, and that the High Court below had erred in demanding proof beyond reasonable doubt. This single holding carries most of the weight in every contested cruelty trial: a spouse's own testimony, if consistent with the pleading and unshaken in cross-examination, is treated as substantive evidence rather than an allegation waiting for external corroboration because the law does not presume that domestic conduct occurs in the presence of witnesses.

Samar Ghosh v. Jaya Ghosh (2007) built on this by rejecting any demand for a single, dramatic, independently provable act. The Court held that instances of alleged cruelty should not be assessed in isolation, and that the correct approach is to take the cumulative effect of the facts and circumstances emerging from the evidence on record, drawing a fair inference from the whole of the matrimonial life rather than testing each incident against a standalone threshold of proof. For a practitioner, this is doctrinally significant: it converts what might otherwise be an impossible task proving each individual incident to a fixed evidentiary standard into an exercise of building a consistent, cumulative narrative across the pleading, the evidence, and the cross-examination, where the overall pattern carries more weight than any single unwitnessed event.

The evidentiary shortcut for false-allegations cruelty

Where the cruelty alleged is the making of a false accusation rather than an act of physical or verbal abuse, the Supreme Court has gone further and largely removed the corroboration problem. In Vijay Kumar Ramchandra Bhate v. Neela Vijaykumar Bhate (2003), the Court held that allegations of unchastity or an illicit relationship, made in the written statement or suggested in cross-examination, are by themselves sufficient to constitute cruelty the making of the allegation is the wrong, not merely evidence of one. K. Srinivas Rao v. D.A. Deepa (2013) extended the same logic to false criminal complaints, holding that a spouse subjected to prosecution later shown to be false suffers cruelty by that fact alone, and that physical cohabitation is no precondition for mental cruelty to be inflicted. Rani Narasimha Sastry v. Rani Suneela Rani (2020) reaffirmed the principle where a false prosecution ends in acquittal. In each of these, the proof problem largely dissolves: the FIR, the chargesheet or its absence, and the order of acquittal form the record, and the petitioner does not have to relitigate the truth of the original allegation from scratch.

It is only when a party sets out to prove the underlying abuse itself not a false complaint but the actual instances of physical or verbal cruelty that the absence of a witness becomes a genuine evidentiary difficulty, and it is here that the discipline of pleading and the survival of testimony under cross-examination, rather than any external corroboration, tends to decide the case.

Where the Domestic Violence Act sits differently

The Protection of Women from Domestic Violence Act, 2005 does not import even the civil standard at the threshold stage. Under Section 12, read with Section 18, a Magistrate may pass a protection order on being merely prima facie satisfied that domestic violence has taken place or is likely to a lower bar than preponderance of probabilities, deliberately so. But this is not a shortcut born of skepticism toward matrimonial litigation generally. It is calibrated to domestic violence specifically, because intimate partner violence operates through control and isolation, and corroboration is structurally unlikely. The lower threshold responds to this reality. The relief sought is urgent protection, not punishment or dissolution, justifying the different standard. 

That design, however, is also where the asymmetry becomes structural rather than incidental. Section 2(a) of the Act defines an "aggrieved person" as a woman in a domestic relationship with the respondent; as the statute was designed to address a specific harm pattern. The protective architecture of the Act simply does not extend to a husband, a point the Delhi High Court confirmed in 2023 when it observed that the protection of the Act is not available to a male member of the family, and more particularly the husband. A husband alleging physical or verbal abuse by his wife is consequently left entirely within the ordinary matrimonial and criminal framework. For the Dastane standard in a divorce petition or an FIR, he must first persuade a police station to register. A man alleging abuse faces an access barrier before he reaches a court at all; police have historically been reluctant to register complaints, treating domestic violence as a private matter. This is an institutional failure, not a judicial one, and it means his case may never generate the records any court, civil or criminal, would later examine. The asymmetry in access to forums is real.

What this means in practice

Put together, the doctrine gives a practitioner three distinct routes to work with, and it is worth being deliberate about which one a given case calls for. Where the allegation is of a false complaint or a false accusation of infidelity, the paper trails the FIR, the chargesheet, and the order of acquittal do most of the evidentiary work, and the task is largely one of assembling and presenting that record. Where the allegation is of the underlying conduct itself, the verbal and physical abuse that leaves no institutional record, the task shifts entirely to the discipline of the pleading: specific, dated, internally consistent instances rather than generalized characterizations, tested and re-tested for whether they survive cross-examination unshaken, because Samar Ghosh tells us the cumulative pattern is what a court will ultimately weigh. And for a wife proceeding under the Domestic Violence Act specifically, the statute itself lowers the threshold in a way ordinary matrimonial proceedings do not, which is worth remembering when advising a client on which forum to approach first and in what order.

None of this resolves the underlying unease a practitioner is right to feel that two structurally similar accounts, one from a wife and one from a husband, can travel through the legal system on very different evidentiary tracks not because judges are applying different standards of proof to similar facts, but because the surrounding architecture, statutory and institutional, was not built symmetrically to begin with and we need to act and advise our clients accordingly.