Plea bargaining in India is a statutory scheme under which an accused person may apply for a mutually satisfactory disposition of the case, which may include compensation to the victim and the meeting of expenses, in exchange for a reduced sentence. It is available only on the accused's own voluntary application, only for offences punishable with imprisonment not exceeding seven years, and it is excluded for offences affecting the socio-economic condition of the country as notified by the Central Government, and for offences committed against a woman or against a child below the age of fourteen. It is not available where the accused has previously been convicted of the same offence. The scheme was introduced into the Code of Criminal Procedure 1973 by amendment in 2005 and is carried forward in the plea bargaining chapter of the Bharatiya Nagarik Suraksha Sanhita 2023 (BNSS).
One clarification is essential before anything else, because it explains a great deal about why the scheme underperforms. This is sentence bargaining, not charge bargaining. The charge does not change. The accused is not negotiating what they are accused of; they are seeking a lighter sentence on the offence as charged, together with a disposition that compensates the victim. The comparison with plea bargaining in the United States, which people reach for immediately, is therefore misleading in its most important respect.
The second thing worth stating at the outset is that uptake has remained low since the scheme was introduced. The interesting question is why, and the answer is more useful than the statute.
Who Qualifies, and Who Does Not
| Requirement | The position |
|---|---|
| Who may apply | The accused, and only the accused. Neither the prosecution nor the court can initiate it. The application must be voluntary and the court must satisfy itself that it is |
| Punishment ceiling | Offences punishable with imprisonment not exceeding seven years. Offences punishable with death or with imprisonment for life are outside the scheme entirely |
| Socio-economic offences | Excluded where the offence affects the socio-economic condition of the country, as notified by the Central Government |
| Offences against women and children | Excluded where the offence has been committed against a woman, or against a child below the age of fourteen years |
| Prior conviction | Not available where the accused has previously been convicted by a court of the same offence |
| Timing | The BNSS introduced a defined window within which the application must be made after charges are framed, which is a change from the earlier position and a discipline on when the option can be exercised |
| Voluntariness | The court examines the accused in camera, in the absence of the other party, to satisfy itself that the application has been filed voluntarily. This is a substantive safeguard, not a formality |
The in camera examination is the heart of the safeguard
The scheme is built around a single risk: that an accused, particularly one in custody or without means, is pressured into an application. The answer the statute gives is that the court examines the accused privately, in the absence of the other side, to satisfy itself that the application is voluntary. An accused who says in that room that they are applying because they were told to should not have their application entertained. That is the mechanism, and it is worth knowing about before walking into it.
How the Procedure Runs
The accused files an application
In the court in which the offence is pending trial, containing a brief description of the case and accompanied by an affidavit sworn by the accused stating that the application has been filed voluntarily, after understanding the nature and extent of the punishment provided for the offence, and that the accused has not previously been convicted of the same offence.
Notice issues to the other participants
The court issues notice to the public prosecutor or the complainant, as the case may be, and to the accused, to appear on a fixed date. The victim's participation is built into the scheme rather than incidental to it.
The court satisfies itself on voluntariness
The accused is examined in camera, in the absence of the other party. If the court is satisfied the application is voluntary, it proceeds. If not, the matter goes back to the ordinary trial from the stage the application was filed.
The parties work out a mutually satisfactory disposition
The court provides time to the parties to work out a disposition, which may include compensation to the victim and other expenses. The court's role here is to facilitate, and the scheme provides for the accused to participate with their advocate.
A report is prepared and signed
Where a satisfactory disposition is worked out, a report is prepared, signed by the presiding officer and the participants. Where no disposition is worked out, the court records that and the matter reverts to ordinary trial.
The court disposes of the case
The court awards compensation as per the disposition and passes sentence, applying the reduced sentencing framework and considering release on probation or after admonition where the accused is eligible.
Limited challenge thereafter
The scheme restricts appeals against a judgment delivered under it, which is the logical consequence of a disposition the accused themselves applied for. The constitutional remedies are not thereby displaced, and that is a question for counsel.
What the Accused Actually Gets
The sentencing framework sits in Section 293 of the BNSS, and it is where the deal is actually priced. Where a minimum punishment is provided for the offence, the court may sentence the accused to half of that minimum. Where no minimum is provided, the court may sentence the accused to one fourth of the punishment provided or extendable for that offence. That much the CrPC scheme already did.
The BNSS added a first time offender tier, and it is the largest change in the chapter
Section 293 goes further than the provision it replaced. Where the accused is a first time offender who has not been convicted of any offence in the past, the figures drop a second time: one fourth of the minimum where a minimum punishment is prescribed, and one sixth of the punishment provided or extendable where no minimum is prescribed. The old scheme stopped at half the minimum and one fourth of the term. For a first time offender this roughly halves an already reduced sentence, and it does so as a matter of the court's power under the section, not as an act of grace that has to be begged for.
First time offender status therefore does two distinct things, and treating it as relevant only to the first is the commonest way the value of the scheme gets understated. Before the court reaches the sentencing clauses at all it hears the parties on whether the accused should instead be released on probation of good conduct or after admonition, or dealt with under the Probation of Offenders Act 1958 or any other law in force. That route remains available. But where the court does sentence, being a first time offender changes the arithmetic of the sentence itself, and any advice on whether to apply that leaves out the one fourth and one sixth tiers is pricing the deal too low.
Alongside the reduced sentence sits the compensation element, which is a genuine feature rather than a decoration. The disposition may provide for compensation to the victim and for the expenses incurred, and the court awards it as part of the disposal. For a victim, a plea bargained outcome can deliver compensation years earlier than a contested trial followed by an appeal would.
The bargain on offer is real: a substantially reduced sentence, a resolved case, and compensation to the victim, all years earlier than a contested trial would deliver any of it. The problem is not the offer. It is that nobody can tell you what it is worth.
Why Almost Nobody Uses It
The common explanations are cultural: that Indian criminal practice is adversarial by temperament, that pleading guilty carries stigma, that defence practice is built around delay. There is something in each of these and none of them is the main reason. The main reason is informational, and it is the reason worth dwelling on because it is fixable.
You cannot price what you cannot compare
To decide rationally whether to accept a bargain, an accused and their counsel need three numbers. What is the realistic sentence if convicted after trial? How long will the trial actually take, including appeals? What have similar accused, on similar facts, in this court, actually received? Without those, the choice is between a known concession and an unknown alternative, and human beings decline that trade almost universally.
In practice those numbers are extremely hard to obtain. Sentencing outcomes are not published in a form that supports comparison. Time to disposal varies enormously by court and case type. A defence counsel with twenty years in a particular court carries an intuition about all three, and that intuition is neither transferable nor auditable, and it is not available at all to the counsel appearing in that court for the first time.
The other reasons, in order of weight
There is no charge concession
The accused pleads to the offence as charged and gets a lighter sentence. In systems where uptake is high, the accused is often also negotiating what they are convicted of, which is a much larger benefit. Comparisons to those systems overstate what is on offer here.
Fear that the application is read as an admission
The statute provides for the matter to revert to ordinary trial where no disposition is worked out, and the statutory protection around statements made in the process exists for exactly this concern. The fear nevertheless deters applications, and perceived risk operates whether or not it is well founded.
Counsel incentives
A contested trial is a longer engagement than an application resolved in a few appearances. This is not stated as an accusation against the Bar, and it is a structural feature of how criminal defence is remunerated, and structural features shape behaviour.
No institutional push
Nobody's performance is measured by plea bargaining uptake. Courts, prosecutors and legal services authorities all have other priorities, and an option nobody actively surfaces stays unused.
The accused has often never heard of it
Particularly where the accused is represented through legal aid or is unrepresented at the early stages. An option that is never explained is not an option, whatever the statute says.
Competing routes already absorb the volume
Compounding of offences, Lok Adalats and settlements in cheque dishonour matters resolve enormous numbers of cases, precisely because in those routes the value of the deal is legible: a stated sum, paid, in exchange for a known end.
What Compounding and Lok Adalats Get Right
The contrast is instructive. Compounding of offences, where the law permits it, and settlement in dishonour of cheque matters through Lok Adalats and mediation, move very large volumes of cases in India every year. Both are voluntary. Both require the accused to give something up. Both work.
The difference is that in those routes the deal is legible. The complainant wants a sum of money. The accused pays it. The proceeding ends. Everyone can evaluate the offer in thirty seconds because it is denominated in a unit everyone understands. Plea bargaining asks the accused to evaluate an offer denominated in sentencing outcomes and time to trial, neither of which they can observe.
That is why the argument of this piece is that plea bargaining in India fails for information reasons rather than design reasons. The design has a genuine offer in it. What is missing is the comparative data that would let anyone value it.
What Would Actually Change the Numbers
- Comparable sentencing outcomes, by offence and by court, so that counsel can say what similar accused have actually received rather than what the section provides for.
- Realistic time to disposal, by court and case type, drawn from case management data rather than from impression. The NJDG publishes pendency information by court and by age of case, and that is the direction this has to come from.
- Visibility of the option itself, at the point where it can still be exercised. The BNSS window after framing of charges makes this sharper: the option now has an expiry date, and an accused who does not know it exists during that window has lost it.
- Better information for victims, since the scheme's compensation element is one of its genuine strengths and is almost never explained to the person it is designed to benefit.
- Institutional attention, particularly from legal services authorities, whose clients are precisely the accused least likely to hear about the option otherwise.
General information, not legal advice
This article describes the plea bargaining scheme in general terms. It is not legal advice. Whether the scheme is available in a particular matter, the applicable timing, and whether it is in an accused person's interests are questions that depend entirely on the offence, the statute and the facts, and must be taken to a criminal law practitioner. The provisions of the BNSS and any applicable special statute govern.
Where CourtMesh Fits
The argument above is that this is a data problem, so it is worth being precise about what data actually helps and what a platform can honestly claim.
CourtMesh covers roughly 310 million cases sourced only from official government portals, spanning the Supreme Court, all 25 High Courts, the district judiciary and tribunals. What that supports is finding comparable matters: judgments and orders in similar offences before the same court, how long comparable matters have taken, and what courts have said when disposing of them. Search retrieves by meaning as well as by keyword, which matters because sentencing reasoning is written in widely varying language and rarely uses the phrases you would think to type.
What it does not do is predict a sentence, and any tool claiming to should be treated with suspicion. Semantic retrieval and deep analysis cover a subset of the corpus rather than all of it, a clean result means the tool found nothing rather than that there is nothing, and no software is enrolled under the Advocates Act 1961. The advice about whether a particular accused should apply is a professional judgement that stays entirely with counsel.
An option nobody can price is an option nobody takes
India's plea bargaining scheme offers a genuinely reduced sentence, compensation to the victim, and an end years earlier than a contested trial. It goes unused largely because neither the accused nor their counsel can say what the alternative is worth. CourtMesh makes comparable matters findable across roughly 310 million cases from official government portals, searched by meaning as well as keyword. It will not predict a sentence, and it will show you what this court has actually done.
Explore CourtMesh
