Brij Bhushan acquittal: What the order leaves open to scrutiny

On August 3, 2026, a Delhi trial court acquitted Brij Bhushan Sharan Singh, the former Wrestling Federation of India president, of charges under Sections 354, 354A, and 506(I) of the Indian Penal Code in the sexual harassment case brought by six women wrestlers. Vinod Tomar, the WFI’s former assistant secretary, was acquitted of the remaining charge against him. The court held that the prosecution had failed to prove its case beyond reasonable doubt.

That conclusion is not, by itself, unusual. Criminal courts are required to acquit where the prosecution evidence leaves a reasonable doubt. What deserves closer examination is the route this particular judgment took to get there.

The case so far

The allegations went back to January 2023, when some of India’s most decorated wrestlers, including Vinesh Phogat, Sakshi Malik, and Bajrang Punia, sat down in protest at Jantar Mantar, accusing Singh of sexually harassing junior wrestlers over several years, at training camps, tournaments, and federation events. Singh, then a six-time BJP MP, denied everything and dared the wrestlers to prove it in court.

It took months of protest, and a nudge from the Supreme Court, before Delhi Police registered two separate FIRs against Singh in April 2023 – one under the IPC for the adult complainants, and a second under the POCSO Act for the seventh, minor complainant. The POCSO case never made it to trial: the minor later withdrew her allegations, police filed a cancellation report, and the case was formally closed in May 2025. The IPC case, brought by the six adult complainants, is the one that reached trial. A chargesheet followed in June 2023, and in May 2024 a magistrate ordered charges framed against Singh under Sections 354, 354A, 354D, and 506 IPC. That trial ran for more than two years, held partly in camera, before Additional Chief Judicial Magistrate (ACJM) Ashwani Panwar delivered the acquittal on August 3. 

Two complainants had turned hostile at trial, telling the court they had been “forced” or “pressured” into making the allegations. The judgment leaned heavily on this, concluding that the case looked “false and fabricated, made collectively” in what it called a “deep conspiracy” that seemed “motivated politically.” Singh, for his part, called the verdict a vindication and said the judiciary had used the word “honour” while acquitting him.

What’s worth sitting with is how the 239-page judgment got there. Across its length, the court repeatedly treated discrepancies in dates and places, continued contact with the accused, the absence of immediate retaliation, photographs, and the conduct of people around the complainants as evidence that the alleged incidents could not have happened. 

A wrong country, a wrong year, and two eyewitnesses

Take the first complainant, referred to as V P. Her allegation is that Singh groped her while she was dining with the Indian contingent in Mongolia in 2016. That account was corroborated in court by two witnesses, and the judgment itself records that corroboration. But V P’s earlier statement before the government-appointed Oversight Committee had named Turkey, 2015, not Mongolia, 2016. She explained in court that she had inadvertently written the wrong country and year, and had since corrected the mistake.

The court treated this as fatal. It called the failure to recall the location “abnormal conduct,” and held that the contradiction was “fatal to the prosecution case.”

That conclusion sits uneasily against the Supreme Court’s own approach to discrepancies in witness testimony. In State of UP v MK Anthony, the Court held that evidence must be assessed as a whole, and that discrepancies which don’t touch the core of the case shouldn’t automatically sink the testimony. Bhoginbhai Bharwada Hirjibhai v State of Gujarat (1983) goes further, cautioning against over-indexing on minor errors of date, location, or peripheral detail when the central narrative of assault holds firm, because ordinary human memory doesn't function with precision under stress.

None of this means a wrong location is always insignificant. Sometimes it’s decisive. But the question the court has to ask is what the discrepancy actually disproves. Here, the country and year were wrong in an earlier, non-judicial document; the allegation itself was later backed at trial by two independent eyewitnesses. 

“Not at all understandable”

The most troubling reasoning in the judgment concerns what the complainants did after the alleged incidents.

The court repeatedly asks why the women stayed on cordial terms with Singh if he had harassed them. It points to photographs from family functions and weddings. It accepts that the complainants might have stayed silent for years out of fear that Singh, then WFI president, could damage their careers. But in the same breath, it says something else: that it was “not at all understandable” why they maintained cordial relations with him for years after.

Para 73 of the judgement read: 

“It is understandable as alleged that Al being the then WFI President would have spoiled their careers if they had made complaint against him, however, it is not at all understandable as to why cordial relations were maintained with him for years. Victims did not complain, they gave justification for the same, but everyone who witnessed was either shocked, disturbed or felt sad for the victims, nevertheless, none was courageous enough to proceed against Al for the alleged acts? Since nobody made any formal complaint or retaliated to the alleged acts of Al, it is highly improbable that the alleged acts are true.”

This is exactly the terrain the Supreme Court warned trial courts of before. In Aparna Bhat v State of Madhya Pradesh, the court cautioned that judicial reasoning in sexual-offence cases must not lean on stereotypical ideas of how a woman “should” behave. That principle was reinforced in State of HP v Prem Singh (2009) and State of Punjab v Gurmit Singh (1996), both of which recognised that delayed reporting, social pleasantries, or unconventional post-assault conduct can’t be judged against a rigid, traditional benchmark, not when trauma and complicated socio-economic dependence are in play.

The judgment accepts that Singh’s position could have made the wrestlers fear for their careers, and then treats their continued contact with him as inexplicable, for the very same reason. The two things aren’t in tension. A person can fear someone professionally and still keep interacting with them, smiling for photographs at public functions included, when their livelihood depends entirely on the institution that person runs.

The same reasoning shows up again around a photograph from Jakarta. The court notes that the image shows a celebratory moment, and that the complainant doesn’t look uncomfortable in it. But a photograph captures a fraction of a second. It can’t establish what an interaction felt like before or after the shutter clicked. Nor does it disprove unwelcome physical contact happening in a crowded athletic arena.

Two hostile witnesses, four who weren’t

Two of the six complainants, referred to as P and S P, turned hostile or retracted core parts of their testimony at trial. That damaged the prosecution’s case, no question. But under Khujji v State of Madhya Pradesh, a hostile witness doesn’t automatically wipe out the entire prosecution case. The portions of testimony that survive scrutiny are still supposed to count.

What the trial court did instead was move from the retraction of two complainants to a much bigger conclusion: that the entire joint case was a false, fabricated “deep conspiracy,” and a politically motivated one at that. In the judge’s own words, the surviving evidence showed the allegations were “false and fabricated made collectively,” in a conspiracy that “seems to be motivated politically.”

An appellate court examining that leap can reasonably ask what independent, objective material connects the retraction of two witnesses to a systemic conspiracy involving the four who didn’t retract, and whose accounts weren’t identical to begin with. The Mongolia allegation, again, was corroborated by two separate eyewitnesses, neither of whom turned hostile or retracted a word of their testimony. 

The constitutional question underneath it all

There is a broader tension sitting under all of this. Section 354 IPC requires proof of assault or criminal force intended, or known to be likely, to outrage a woman’s “modesty.” The Bharatiya Nyaya Sanhita carries the same formulation forward into Section 74. Meanwhile, Indian constitutional jurisprudence, most prominently KS Puttaswamy v Union of India, has moved toward reading Article 21 in terms of bodily integrity, personal autonomy, and decisional freedom.

Against that backdrop, the trial court’s preoccupation with whether the complainants behaved “naturally,” whether they retaliated at once, whether bystanders reacted, whether husbands intervened, whether they looked comfortable in photographs, deserves close appellate scrutiny. 

An acquittal means the prosecution didn’t establish guilt beyond reasonable doubt. Appellate courts are, rightly, cautious about disturbing that under Chandrappa v State of Karnataka, given the double presumption of innocence an acquittal carries. But Ravi Sharma v State (NCT of Delhi) (2022) makes clear that appellate interference is warranted where a trial court’s appreciation of evidence is legally flawed, rests on speculative generalisations, or ignores material corroboration on record.

The real question the Brij Bhushan judgment raises isn’t only whether every charge was proved. It’s whether, in deciding what counted as credible, the trial court was actually answering the legal question in front of it: what does the evidence establish? Or whether it quietly swapped that question for a different one: would a real victim have behaved this way?

Small teams can do great things. All it takes is a subscription. Subscribe now and power Newslaundry’s work.