On March 24, a Sessions Court in Delhi denied bail to Umar Khalid as part of a set of cases that have commonly come to be known as “the Delhi riots cases”. The case of the police was that Mr. Khalid was one of the conspirators behind the February 2020 violence in Delhi, which had claimed more than 50 lives. For this, Mr. Khalid, along with many others, was charge-sheeted under the Unlawful Activities (Prevention) Act (UAPA), 1967, and jailed pending trial. Mr. Khalid has been in jail for over 500 days. The trial has not yet begun.
Much has been written about the serious problems with the manner in which the Delhi Police has conducted its investigation, and prosecution of the Delhi riots: in particular, its selective targeting of activists who were involved with the protests against the Citizenship (Amendment) Act of 2019, which was the alleged trigger for the violence, while refraining from prosecuting individuals who are on record delivering incendiary speeches. The denial of bail to Mr. Khalid highlights an equally serious problem: the broken nature of India’s criminal justice system.
Bail hearing becomes trial
First, consider these facts. Mr. Khalid’s bail application was filed in July 2021. The order denying bail was passed eight months later, after multiple hours-long hearings, multiple adjournments, and three deferrals of the order itself. It is important to ask why an application for bail took so many hearings and eight months to decide: in criminal law, the purpose of bail is to ensure that an individual is not unjustly denied their liberty while the trial against them is still proceeding and their guilt has not yet been established. As such, in normal circumstances, courts are supposed to consider whether an accused is a flight risk, or is likely to tamper with evidence or intimidate witnesses. If neither of those dangers exist, there is no purpose in denying an individual their freedom before their guilt has been established in a court. This, in other words, is the real meaning of the hoary phrase ‘innocent until proven guilty’.
This is where the notorious UAPA comes in. Shorn of legalese, the UAPA prohibits a judge from granting an individual bail if, on a perusal of the police diary or the police report, the judge is of the opinion that there are “reasonable grounds for believing that the accusation is … prima facie true.” The effect of this, as the criminal legal scholar Abhinav Sekhri has pointed out, is that the UAPA introduces elements of the criminal trial into the question of bail. There are traces of this in the Indian Penal Code as well, for bail under serious non-UAPA offences. This hints at a larger problem with the criminal justice system, of which the UAPA is only the starkest example. Questions of guilt or innocence are meant to be determined at the end of a trial, after evidence has been sifted, witnesses examined and cross-examined, and arguments completed. The question of guilt or innocence at the stage of bail short-circuits that essential procedure.
But that is not the only problem with turning bail hearings into mini-trials. The problem is also that this mini-trial — to borrow a colourful phrase from the U.S. Supreme Court — licenses “one side … to fight freestyle, while requiring the other to follow the Marquis of Queensberry Rules (i.e., the rules of professional boxing)”. What the judge has before them is entirely one side of the case: the police version. In a trial, the defence would be entitled to cross-examine the prosecution’s witnesses, determine inconsistencies in their testimony, examine its own witnesses, present its own evidence, and otherwise demonstrate that the case against the accused has not been made out beyond reasonable doubt. In a bail hearing, the defence can do none of that. The starting point of the bail hearing is the presumption that everything in the police report is true. Based on that presumption, all the two sides can then argue about is whether according to these “facts”, the legal ingredients of the offence are fulfilled — or, in some rare cases, about whether the facts themselves are self-contradictory or flat-out implausible, so that no reliance can be placed on them even at the stage of bail. To use an analogy, it is like holding a debate between two sides, stopping it after one side finishes, allowing the other side to pose two or three questions but not say anything more, and then deciding whether the motion passes or falls.
Such a system might possibly be defensible in a situation where criminal justice was swift, efficient, and trustworthy. If, for example, criminal trials habitually concluded within six months, it might just be possible to argue that in terrorism cases, six months of pretrial incarceration is a painful but proportionate price to pay (in my opinion, it is still unjustifiable, but there is at least a case to be made). However, that is not the case in India: a UAPA trial takes years — often more than 10 years. In such a situation, the court’s decision on bail, de facto, becomes the decision on the case: the denial of bail means that a person is likely to spend a decade or more behind bars, as the trial winds on. And given the UAPA’s abysmally low conviction rates, the trial will likely end in acquittal.
This, thus, explains why bail hearings take so long, and are so convoluted (although there is still little excuse for the eight-month-long process in Mr. Khalid’s case). Both the defence and the prosecution know that the outcome of the bail hearing is, for all practical effects, the outcome of the case itself. The result of the denial of bail is, functionally, the same as the result of a finding of guilt: a decade-plus in jail. But, as we have seen, while the denial of bail is effectively a finding of guilt, it has none of the safeguards that the criminal law puts into place before an actual finding of guilt. The accused is first gagged from contesting the police’s version and is then condemned for not being able to disprove the police’s case.
In a notorious judgment in National Investigation Agency v. Zahoor Ahmad Shah Watali (2019), the Supreme Court made a bad situation even worse by forbidding the lower courts from scrutinising in depth even the police case. This leads to absurd situations like Mr. Khalid’s bail order. A reading of the bail order shows that the court reproduces various allegations against Mr. Khalid — some of them hearsay, and therefore inadmissible during the trial, and some extremely implausible; dismisses the defence’s challenges to them without any engagement; and then denies bail. Lawyers and legal scholars may disagree over whether the UAPA actually requires the courts to become stenographers for the prosecution, even under existing legal doctrine. The point, however, is that for all the reasons we have discussed above, the result is rank injustice.
Reforming the criminal justice system is the task of many years. In the immediate future, however, it is at least possible to curtail the manner in which the UAPA plays havoc with the lives of so many individuals. Striking down or reading down its bail prohibitions and subjecting the police case to stricter scrutiny during bail hearings would be a start. It remains to be seen whether the judiciary has the will and the inclination to do so.
Gautam Bhatia is a Delhi-based lawyer