Author(s): Chhavi Agrawal
Paper Details: Volume 4, Issue 4
Citation: IJLSSS 4(4) 13
Page No: 114 – 127
ABSTRACT
The provision which states that bail shall be granted shall be an exception in cases of emergency”, has been reiterated in various decisions of the Indian Supreme Court since 1977. Today, however, the figure stands at over 75% of the population of all prisoners in India are those who are not charged with any offence but are merely awaiting trial, according to government data, almost 50 years after that. The present paper he has traced in the bail law and the evolution of law from State of Rajasthan v. Balchand (1977) to Sunny Chauhan v. State of Haryana (2026) and has come to know the difference between the judicial principle and practice. The paper discusses structural challenges to the implementation of bail measures, based on data sourced from the National Crime Reporting Bureau (NCRB) Prison Statistics India 2022, Supreme Court of India Report regarding the bail, Law Commission Reports on Prison Reform and Comparative Study of Prison Practice of various International standards for the Pretrial detention. The national statistics for three years cover a decline in the overall occupancy rate from 131.4 per cent (2022) to 112.7 per cent (2024) but the percentage of undertrials has fallen marginally by only 3.2 per cent (from 75.8 per cent in 2022 to 72.6 per cent in 2024), and number of admissions has risen by 4.2 per cent in 2024. The paper explains a “greater harm” test created with the precedents of the Supreme Court, and this is the test that may only be used when the granting of bail would cause a “specific and Articulated harm.” That demonstrates the lack of use of this test in practice at the magisterial level. The conclusion of the paper is that systemic change in the judiciary needs to be achieved through providing structured systems of accountability in a structured way that has been successful in international courts.
Keywords: Undertrial Prisoners, Bail Reform, Section 479 BNSS, Article 21, Pretrial Detention, Criminal Justice Reform, Greater Harm Test, NCRB Prison Statistics 2024
I. INTRODUCTION
One of the cornerstones of any democratic criminal justice system is the presumption of innocence. However, the latest government stats provide a very different picture in India, given that the right to life and personal liberty is guaranteed under the country’s Constitution; and that a long line of Supreme Court decisions have established that bail is the rule, and jail the exception.[1] The NCB’s Prison Statistics India 2024 (Ministry of Home Affairs, Government of India, 2026) state that currently India’s prisons are holding 5,11,542 inmates while the sanctioned number of prisoners in the prisons was approximately 4.53 lakh, with 3,71,440 inmates being undertrials accounting for 72.6 percent of inmates.[2] As per the World Prison Brief, the percentage of India’s pretrial population is one of the highest among the countries in the world, which, according to a report by United Nations Office on Drugs and Crime (UNODC), in 2023 recorded approximately 11.7 million persons detained across the world of which, about one-third were detained in the pretrial settings[3]
This dissonance between legal principle and empirical reality is not new. The Supreme Court first articulated the bail-not-jail doctrine in State of Rajasthan v. Balchand in 1977,[4] and reinforced it through landmark decisions including Hussainara Khatoon v. State of Bihar (1979),[5] Satender Kumar Antil v. CBI (2022),[6] and most recently, Sunny Chauhan v. State of Haryana (2026).[7] The Bharatiya Nagarik Suraksha Sanhita 2023 introduced Section 479, replacing Section 436A of the Code of Criminal Procedure, with relaxed bail provisions for first-time offenders.[8] Despite these cumulative legislative and judicial interventions spanning nearly five decades, the undertrial crisis has persisted with remarkable consistency.
Mandated by the Supreme Court’s recent interventions in 2024-2026, literature has not evaluated the persistence of the crisis from the perspective of reforms under the NCRB and the latest of the Supreme Court’s verdicts (2024–2026) nor the three years of NCRB data before these reforms (2022–2024). It traces the history of the development of bail jurisprudence, aims to identify a “greater harm” test emerging from the Supreme Court case law, breaks down the latest empirical data for the last three reporting years and places the Indian experience in a comparative international perspective. The aim is to create an overview of the mismatch between jurisprudential ideals and actual practices, as well as to document structural factors that can be seen in judicial observations and parliamentary committee reports, in government data, and in empirical studies, that support this gap.
II. LEGISLATIVE FRAMEWORK ON BAIL AND PRETRIAL DETENTION
The constitutional framework for securing the undertaking rights in India is mainly based on this provision of Article 21 which states that no person shall be deprived of life or personal liberty except according to procedure established by law.[9] The Supreme Court’s expansive interpretation of this provision, beginning with Maneka Gandhi v. Union of India (1978), established that the procedure must be just, fair, and reasonable a standard that directly governs the conditions under which pretrial detention may be imposed.[10]
The statutory framework governing bail has recently undergone significant reform through the Bharatiya Nagarik Suraksha Sanhita 2023, which replaced the Code of Criminal Procedure 1973 with effect from 1 July 2024. Section 479 of the BNSS, which supersedes Section 436A of the CrPC,[11] establishes two tiers of mandatory bail eligibility. Under the general provision, an undertrial who has served one-half of the maximum period of imprisonment specified for the alleged offence must be released on bail.[12] For first-time offenders, this threshold is reduced to one-third of the maximum sentence. A significant procedural addition requires the Superintendent of the jail to proactively apply to the concerned court for the release of eligible undertrials.[13]
However, the BNSS also introduced certain restrictions absent from the CrPC. Persons accused of offences punishable with death or life imprisonment are excluded from the benefit of Section 479 a notable departure from Section 436A, which had permitted even such accused to seek bail. Additionally, Section 479(2) bars bail where investigations in multiple cases are pending against the accused.
In August 2024, the Supreme Court in Re-Inhuman Conditions in 1382 Prisons, a public interest litigation originating from a letter by former Chief Justice R.C. Lahoti, ordered that Section 479 shall apply retrospectively to all undertrials regardless of whether their cases were registered before or after 1 July 2024.[14] The Court directed jail superintendents across the country to process bail applications for eligible undertrials immediately.
III. EVOLUTION OF SUPREME COURT BAIL JURISPRUDENCE
The judicial articulation of bail principles in India can be traced through a remarkably consistent line of authority spanning nearly five decades. In Gudikanti Narasimhulu v. Public Prosecutor (1978), Justice V.R. Krishna Iyer described the bail determination as involving a consideration of liberty, justice, public safety, and the burden on the public treasury.[15] He held that judicial discretion in bail matters must be exercised with a predisposition towards liberty, and that the denial of bail must be substantiated by compelling and constitutionally sound reasons.[16] In State of Rajasthan v. Balchand (1977), the Supreme Court distilled the principle into its most enduring formulation: the basic rule is bail, not jail.[17]
In Hussainara Khatoon v. State of Bihar (1979), the Court confronted the reality of mass undertrial detention. Discovering that thousands of prisoners in Bihar had been detained for periods exceeding the maximum punishment for their alleged offences, the Court read the right to a speedy trial into Article 21 and directed the release of eligible undertrials.[18] Gurbaksh Singh Sibbia v. State of Punjab (1980) further clarified that the object of bail is to secure the attendance of the accused at trial, not to impose preventive punishment.[19]
The Supreme Court in Prahlad Singh Bhati v. NCT Delhi (2001) enumerated the specific factors that courts must consider when determining bail applications: whether there is prima facie ground to believe the accused committed the offence; the nature and gravity of the charge; the severity of potential punishment; the danger of the accused absconding; the likelihood of the offence being repeated; reasonable apprehension of witness tampering; and the danger of justice being thwarted.[20] This framework effectively established a structured test a set of specific grounds that must be assessed before bail is denied.
The cumulative effect of these decisions establishes what may be described as a “greater harm” test for pretrial detention bail may be denied only where the court identifies a specific, articulable harm that would result from the accused’s release, namely the likelihood of absconding, tampering with evidence, influencing witnesses, or repeating the offence. Where none of these grounds exist, the constitutional mandate under Article 21 requires that bail must be granted. Subsequent decisions reinforced this framework. In Sanjay Chandra v. CBI (2012), the Court held that the primary purpose of bail is to ensure the accused’s presence at trial, and that where the accused is not likely to abscond or tamper with evidence, bail should ordinarily be granted.[21] Arnesh Kumar v. State of Bihar (2014) restricted arrests for offences punishable with seven years or less, requiring the police to record reasons for arrest and mandating magistrates to verify compliance before authorising detention.[22]
This “greater harm” principle found its most explicit judicial articulation in the decisions of Justice D.Y. Chandrachud. In Arnab Manoranjan Goswami v. State of Maharashtra (2020), Justice Chandrachud observed that courts must remain the first line of defence against deprivation of liberty, and that such deprivation even for a single day is one day too many.[23] The Court criticised the Bombay High Court for failing to conduct a prima facie evaluation of the case before declining bail, characterising this as an abdication of constitutional duty.[24] In Ramesh Bhavan Rathod v. Vishanbhai Hirabhai Makwana, his bench held that bail orders must not be casual, arbitrary, or mechanical, and must reflect a proper application of mind to the specific grounds of the “greater harm” test establishing that the structured assessment of harm is not merely aspirational but judicially mandated.[25]
Satender Kumar Antil v. CBI (2022) represents the most comprehensive judicial intervention on bail reform. The Court categorised offences into four groups and issued detailed guidelines directing lower courts to avoid unnecessary arrests, prefer summons over warrants, and grant bail where statutory conditions are met.[26] It demanded a cultural shift from reflexive custody to reasoned restraint and invoked data showing that over two-thirds of prison inmates were undertrials, many of whom need not have been arrested at all.[27] When Justice Chandrachud became Chief Justice of India in November 2022, the institutional expectation was that these principles would translate into systemic change. Yet during his tenure as CJI (2022–2024), the NCRB data showed undertrial percentages declining only from 75.8 percent (2022) to 72.6 percent (2024), and decisions such as Javed Gulam Nabi Shaikh and Jalaluddin Khan still had to restate the very principles his earlier benches had established.[28]
In 2024, the Supreme Court delivered a series of decisions further strengthening undertrial rights. Javed Gulam Nabi Shaikh v. State of Maharashtra (2024) held that Article 21 applies irrespective of the nature of the crime, and that prolonged incarceration without trial converts pretrial detention into punishment.[29] Jalaluddin Khan v. Union of India (2024) reaffirmed that bail is the rule even under special statutes such as the Unlawful Activities (Prevention) Act.[30] In Manish Sisodia v. Directorate of Enforcement (2024), the Court emphasised that bail should not be withheld as punishment.[31]
Most recently, in Sunny Chauhan v. State of Haryana (2026), a bench of Chief Justice of India Surya Kant and Justice Joymalya Bagchi, while hearing a case concerning the pendency of bail applications before the Punjab and Haryana High Court, pulled up the High Courts and issued six binding structural directives.[32] These included mandatory weekly or fortnightly listing of bail matters, same-day pronouncement and uploading of bail orders, mandatory filing of status reports before the first hearing, mandatory service of bail applications on the Advocate General, prohibition on casual adjournments to Union or state governments, and a direction to High Court Chief Justices to take up with state governments the persistent delays in forensic science laboratory reports which continue to hamper bail proceedings.[33] The Court described liberty under Article 21 as the most precious fundamental right an extraordinary intervention that itself evidences how deep the institutional resistance to timely bail disposal runs.
The chronological trajectory from Balchand (1977) to Sunny Chauhan (2026) reveals a notable pattern: the same constitutional principle and the same “greater harm” test has been restated with increasing urgency by the very institution entrusted with enforcing it. That the same court, and in several instances the same judge, had to reiterate identical principles across a span of years itself demonstrates that the failure is not doctrinal but institutional. The judiciary has diagnosed the disease accurately and repeatedly, but lacks the institutional machinery to administer the cure at the district and magistrate level where bail decisions are actually made.
IV. UNDERTRIAL DETENTION: A THREE-YEAR EMPIRICAL ANALYSIS (2022-2024)
The NCRB Prison Statistics India reports for 2022, 2023, and 2024 provide the most comprehensive longitudinal picture of India’s undertrial crisis. In 2022, India’s prisons held 5,73,220 inmates, of whom 4,34,302 approximately 75.8 percent were undertrial prisoners.[34] By 2023, the total prison population declined to 5,30,333, with 3,89,910 undertrial prisoners constituting 73.5 percent.[35] The most recent data for 2024 records a further decline to 5,11,542 total prisoners, of whom 3,71,440 or 72.6 percent were undertrials. Convicts numbered 1,36,138 (26.6 percent) and detenues 3,048 (0.6 percent).[36]
The three-year trend reveals an important pattern. The total prison population declined by 10.8 percent over two years (from 5,73,220 to 5,11,542), and the national occupancy rate fell from 131.4 percent to 112.7 percent a decade low. However, the undertrial proportion declined by only 3.2 percentage points. More significantly, total admissions to Indian jails actually increased in 2024: 17,97,486 inmates were admitted during the year, a rise of 4.2 percent over the 17,24,522 admissions in 2023.[37] This admission data reveals a revolving-door dynamic more individuals are entering the prison system than before, but releases are slightly outpacing admissions, producing a modest net decline. Yet the undertrial proportion remains virtually unchanged, demonstrating that India’s decongestion efforts address the symptom of overcrowding while leaving the structural over-reliance on pretrial detention untreated.
The undertrial population decreased by 4.7 percent from 3,89,910 in 2023 to 3,71,440 in 2024.[38] During 2024, approximately 1.5 million undertrial prisoners were released on bail, alongside 1,50,000 convicts.[39] Yet the year-end undertrial population settled at 3,71,440, indicating a continuous cycle of new admissions replacing releases.
The demographic composition of the undertrial population reinforces the structural dimension. Nearly half of all undertrial prisoners are under the age of 30, with another 41 percent between 30 and 50 years. Approximately 63 percent did not study beyond Class X.[40] Among states that reported caste data covering 3,06,391 undertrials or 82 percent of the total persons from Scheduled Castes, Scheduled Tribes, and Other Backward Classes together accounted for 67.3 percent. The share of OBC undertrial prisoners has increased consistently from 29 percent in 2001 to 36 percent in 2024, while SC and ST prisoners together accounted for 31.3 percent.[41]
The duration of detention presents an equally concerning picture. While 69.9 percent of undertrials were confined for up to one year, nearly 30,000 prisoners constituting 8.1 percent had spent more than three years in custody. Of these, 9,028 individuals had been in jail for more than five years without conviction.[42] This figure has risen from 2.5 percent in 1998 to 8.1 percent in 2024.
Regional disparities further illuminate the crisis. Uttar Pradesh reported the maximum number of undertrials at 59,194 (15.9 percent), followed by Bihar at 45,339 (12.2 percent) and Maharashtra at 31,523 (8.5 percent).[43] Delhi recorded the highest occupancy rate at 194.6 percent,[44] with an undertrial share of 88 percent. Bihar recorded the highest undertrial share among major states at 87.2 percent.[45] As of 31 December 2023, 24,879 accused persons who had been granted bail continued to remain in prison due to their inability to furnish bail bonds,[46] while bail rejection rates stood at 32.3 percent in Sessions Courts and 16.2 percent in Magisterial Courts.[47] Staffing data compounds the picture: against 99,758 sanctioned jail staff positions, only 63,004 were filled a vacancy rate of approximately 37 percent.[48]
V. INTERNATIONAL STANDARDS AND COMPARATIVE PRACTICES
International human rights law establishes a clear framework on pretrial detention. Article 9(3) of the International Covenant on Civil and Political Rights provides that it shall not be the general rule that persons awaiting trial shall be detained in custody.[49] The United Nations Human Rights Committee has stated that bail should be granted except where the accused would abscond, destroy evidence, influence witnesses, or flee.[50] The Tokyo Rules provide that pretrial detention shall be used as a means of last resort.[51]
The UNODC Handbook on Strategies to Reduce Overcrowding in Prisons identifies pretrial detention as a primary driver of prison overcrowding globally.[52] The United Nations System Common Position on Incarceration identifies the overuse of pretrial detention as a systemic global crisis linked to poverty and inequality, and connects prison reform to SDG 16.[53] The ABA’s Handbook documents convergence across international systems: pretrial detention is permissible only to prevent absconding, serious offences, interference with justice, or threats to public order.[54] The Council of Europe’s jurisprudence provides that seriousness of the alleged offence alone cannot constitute an absolute bar to release.[55]
Two national reforms offer instructive models. The UK Bail Act 1976 establishes a statutory presumption in favour of bail, with courts required to record reasons for refusing bail from an exhaustive statutory list of exceptions.[56] [57] This mandatory recorded-reasoning requirement ensures accountability in every bail decision. New Jersey’s Criminal Justice Reform Act (2017) replaced cash bail with a risk-based assessment system.[58] The pretrial jail population declined by approximately 40 percent; by 2021, 80 percent of detained defendants were charged with serious offences compared to 35 percent before reform.[59] Defendants must be indicted within 90 days and tried within 180 days.[60]
VI. IDENTIFYING THE IMPLEMENTATION GAP
Given the feeble response to such a well-developed statutory framework and consistent judicial rulings, one possible explanation for this issue is that there are structural factors working against the engagement of these actors.
First of all is the concern about the vacant seats in the Judiciary and pendency of cases. In the cases of Uttar Pradesh and Bihar, with an NCRB data record of maximum number of undertrials in Allahabad High Court (59,194) and Patna High Court (45,339), the pendency aggravated in these Courts aired in the einige words of the Supreme Court in Sunny Chauhan (2026), requiring necessary remedial measures in the listing system.[61]
The second factor concerns policing practices. The Supreme Court in Arnesh Kumar (2014) acknowledged that police routinely arrest individuals even where arrest is neither necessary nor warranted.[62] The PSI 2024 admission data corroborates this: 17,97,486 inmates were admitted during 2024, a 4.2 percent increase over 2023.[63] Even as the year-end population declined, the system admitted more people than before indicating that the rate of unnecessary arrests has not reduced. The system continually replenishes its undertrial stock through fresh arrests.
The third factor is the dysfunction of legal aid services. Free legal aid is a constitutional right under Article 39A read with Articles 14 and 21, and NALSA operates extensive schemes for legal assistance to undertrial prisoners.[64] However, the fiscal underpinning of legal aid reveals a systemic failure. According to the India Justice Report 2025, legal aid receives less than one percent a mere 0.49 percent of the total justice budget.[65] NALSA’s own funds declined from ₹207 crore in 2017-18 to ₹169 crore in 2022-23, and fund utilisation dropped from 75 percent to 59 percent due to rigid expenditure restrictions imposed by the NALSA Manual 2023, which bars State Legal Services Authorities from spending without prior central approval.[66] The number of para-legal volunteers the frontline workers who visit prisons and assist undertrials fell by 38 percent between 2019 and 2024, with most receiving honorariums below minimum wage. Between April 2023 and March 2024, only 15.5 lakh persons accessed legal aid despite nearly 80 percent of India’s population being eligible. National per capita spending on legal aid stands at merely ₹7 a fraction of the ₹225 per day the State spends on housing each undertrial in custody.[67] The Supreme Court’s Centre for Research and Planning report revealed bail rejection rates of 32.3 percent in Sessions Courts and 16.2 percent in Magisterial Courts, with over 52 percent of cases where the accused had been in custody for more than a year still pending at the evidence stage.[68]
The fiscal irrationality extends beyond legal aid. The Rajya Sabha Department-Related Parliamentary Standing Committee on Home Affairs, in its 245th Report on Prison Conditions, Infrastructure and Reforms, observed at paragraph 4.10.6 that more than 70 percent of prisoners in Indian jails are undertrials who are not being released for want of sureties or inability to pay fine amounts, and that prison administrations are spending far more money on keeping such prisoners inside jails than the bail money required for their release. The Committee recommended creating a fund on the lines of the “Cheyutha Nidhi” started by the Andhra Pradesh Prison Department for payment of fine amounts for poor prisoners in all States and Union Territories.[69] Yet the Central Government’s own Support to Poor Prisoners Scheme designed precisely to address this bail-poverty nexus reached only 273 prisoners across 16 states, with total expenditure of merely ₹71.7 lakh, while 20 states did not utilise the scheme at all.[70] The State thus spends an estimated ₹8 crore per day on housing undertrials while spending ₹71.7 lakh on the entire bail assistance scheme a ratio that encapsulates the fiscal dimension of the implementation gap.
Judicial recognition of the fourth factor, institutional reluctance. Denying bail does not mean that the magistrate’s job is getting perilous so that there is no risk incentivizing the denial of bail, granting it renders the bowel of the judicial officer accountable; if the accused commits further offences or absconds, the officer will attract the cynosures of the institutions. The NCRB data shows that a majority of India’s 3,71,440 undertrials are arrested for crimes where there exists no plausible “greater harm” argument but are still denied bail with no specific reasoning for the court to consider on bail and for the assessment of the evidence tampered with or likelihood of repeat offences. The formulated “greater harm”” test prescribed by Supreme Court in the case of Prahlad Singh Bhati, Arnab Goswami & Satender Kumar Antil is yet to be pressed at the magisterial level, where most of bail applications are processed.
However, it is also noted that the rate of decline of the proportion of the under-trials reduced from 77.1 per cent in 2021 to 72.6 per cent in 2024 showing some measure of impact of reforms. At this rate of decline, however, it will take more than 30 years to reach the international standard – 20-30 percent of prisoners in pretrial conditions. The BNSS’ own structural decisions have also muddied the waters: the decision to exclude those facing life-sentences from Section 479 is a step backwards for individuals caught with Section 436A effects and the ban on bail for people who have more than one investigation pending allows for the possibility of sequential registrations.
VII. CONCLUSION
The jurisprudence of the Indian bail from Balchand’s case (1977) to the case of Sunny Chauhan (2026) demonstrates a paradox – the articulation of undertrial rights has strengthened over time, while the real-world evidence shows resistance to change. The 3 years uncovered the fact that although the rate of occupants was 131.4% and 112.7%, the proportion on the undertrial had only shifted by 3.2% from 75.8% to 72.6%, while the number of persons entering into jail in a year had risen by 4.2%. More than 9,028 people have been detained for over five years without being convicted, 24,879 were released on bail, but were still detained without being able to provide bail bonds, and 273 were able to take advantage of the government’s own bail program nationwide.
Such a comparative analysis has indicated that it is not doctrine that is missing, but rather institutions. All of the foregoing point to the same conclusion: the presumption of bail must be put into practice by putting mechanisms in place that change its nature from an exhortation to that of a mechanism of accountability for failure to release someone.It is from the above that the following conclusion can be drawn: the presumption of bail must be turned into a mechanism that entails accountability for failing to release a person, not just on appeal after somebody has been arrested.
The data systematically reveals that the SBF bearers are predominantly young, poor-educated and from socially and economically underprivileged communities. Two-thirds are from SC, ST and OBC communities. Almost half of them are less than 30 years old. More than one out of three did not even finish secondary school. While the State funds legal aid at just ₹7 per person per year, it allocates ₹225 to house an undertrial for a day, turning the ‘undertrial situation’ from being an issue of rights to being an issue of finances! They can’t just imagine that something like “judicial memory” would keep them from being reunited.Being separated is not something which can be explained to them with a doctrinal nuance but the lost years away from loved ones and friends only because they lacked a conviction.
“Setting the perimeter of the rule” is not the challenge of criminal justice reform that is the preoccupation of Indian minds now. These two have been carefully constructed, reiterated, implemented and strengthened through almost five decades of uniform court decisions. The difficulty, instead, lies in developing the institutional framework for the judicial capacity, legal aid financing, policing accountability, mechanisms of the bail fund, as well as the transparency of the procedures, within which this standard would be uniformly applied. Till it is achieved, the bail-not-jail principle will merely be considered a corrective when asked in an appeal and the system of failure to live up to its constitutional ideals will continue to make India’s undertrial situation an indictment.
[1]State of Rajasthan v. Balchand, AIR 1977 SC 2447.
[2]National Crime Records Bureau, Prison Statistics India 2024 (Ministry of Home Affairs, Government of India, 2026). Due to non-receipt of data from West Bengal for 2024, the NCRB used data furnished for 2023. All PSI 2024 figures cited in this paper are subject to this caveat.
[3]United Nations Office on Drugs and Crime (UNODC), Prison Matters 2025: Global Prison Population and Trends; A Focus on Rehabilitative Environments (United Nations, July 2025) 7, 13, 17.
[4]State of Rajasthan v. Balchand, supra note 1.
[5]Hussainara Khatoon v. State of Bihar, (1979) 3 SCC 532.
[6]Satender Kumar Antil v. Central Bureau of Investigation, (2022) 10 SCC 51.
[7]Sunny Chauhan v. State of Haryana, SLP (Crl.) No. 1613 of 2026, Order dated 11 May 2026 (Supreme Court of India).
[8]The Bharatiya Nagarik Suraksha Sanhita, 2023, § 479.
[9]The Constitution of India, art 21.
[10]Maneka Gandhi v. Union of India, AIR 1978 SC 597.
[11]The Code of Criminal Procedure, 1973, § 436A.
[12]The Bharatiya Nagarik Suraksha Sanhita, 2023, § 479(1).
[13]The Bharatiya Nagarik Suraksha Sanhita, 2023, § 479(3).
[14]In Re-Inhuman Conditions in 1382 Prisons, W.P. (C) No. 406 of 2013, Order dated 23 August 2024 (Supreme Court of India)
[15]Gudikanti Narasimhulu v. Public Prosecutor, (1978) 4 SCC 47.
[16]Id.
[17]State of Rajasthan v. Balchand, supra note 1.
[18]Hussainara Khatoon v. State of Bihar, supra note 5.
[19]Gurbaksh Singh Sibbia v. State of Punjab, AIR 1980 SC 1632.
[20]Prahlad Singh Bhati v. NCT Delhi, (2001) 4 SCC 280.
[21]Sanjay Chandra v. Central Bureau of Investigation, (2012) 1 SCC 40.
[22]Arnesh Kumar v. State of Bihar, (2014) 8 SCC 273.
[23]Arnab Manoranjan Goswami v. State of Maharashtra, (2020) 10 SCC 616.
[24]Id.
[25]Bhavan Rathod v. Vishanbhai Hirabhai Makwana (Koli), (2021) 6 SCC 230.
[26]Satender Kumar Antil v. CBI, supra note 6.
[27]Satender Kumar Antil v. CBI supra note 6, para 72.
[28]Arnab Manoranjan Goswami v. State of Maharashtra, supra note 23; Satender Kumar Antil v. CBI, supra note 6; National Crime Records Bureau, Prison Statistics India 2022, supra note 34; National Crime Records Bureau, Prison Statistics India 2024, supra note 2.
[29]Javed Gulam Nabi Shaikh v. State of Maharashtra, (2024) 9 SCC 813.
[30]Jalaluddin Khan v. Union of India, (2024) 10 SCC 574.
[31]Manish Sisodia v. Directorate of Enforcement, 2024 INSC 595.
[32]Sunny Chauhan v. State of Haryana, supra note 7.
[33]id..
[34]National Crime Records Bureau, Prison Statistics India 2022 (Ministry of Home Affairs, Government of India, 2023): Total prison population 5,73,220; Undertrial prisoners 4,34,302 (75.8%); Convicts 1,33,415; Detenues 4,324.
[35]National Crime Records Bureau, Prison Statistics India 2023 (Ministry of Home Affairs, Government of India, 2025): Total prison population 5,30,333; Undertrial prisoners 3,89,910 (73.5%); Convicts 1,35,536; Detenues 3,916.
[36]National Crime Records Bureau, Prison Statistics India 2024 supra note 2: Total prison population 5,11,542; Undertrial prisoners 3,71,440 (72.6%); Convicts 1,36,138 (26.6%); Detenues 3,048 (0.6%). All figures as on 31 December 2024.
[37]National Crime Records Bureau, Prison Statistics India 2024, supra note 2: During 2024, a total of 17,97,486 inmates were admitted in various jails, an increase of 4.2% over 2023 (17,24,522 inmates).
[38]National Crime Records Bureau, Prison Statistics India 2024, supra note 2: Undertrial prisoners decreased from 3,89,910 in 2023 to 3,71,440 in 2024, a decrease of 4.7%.
[39]National Crime Records Bureau, Prison Statistics India 2024, supra note 2, ch. 7, tbl. 7.3.
[40]National Crime Records Bureau, Prison Statistics India 2024 (Ministry of Home Affairs, Government of India, 2026), ch. 2, at 38, 40, tbls. 2.11A & 2.15, chart 2.8.
[41]National Crime Records Bureau, Prison Statistics India 2024, supra note 2, ch. 2, tbl. 2.11D.
[42]National Crime Records Bureau, Prison Statistics India 2024, supra note 2. 69.9% of undertrials confined for up to one year; nearly 30,000 (8.1%) had spent more than three years; 9,028 had spent more than five years without conviction.
[43]National Crime Records Bureau, Prison Statistics India 2024, supra note 2. Uttar Pradesh reported the maximum number of undertrials (15.9%, 59,194), followed by Bihar (12.2%, 45,339) and Maharashtra (8.5%, 31,523) at end of 2024.
[44]National Crime Records Bureau, Prison Statistics India 2024, supra note 2. Delhi recorded the highest prison occupancy rate at 194.6% in 2024.
[45]IndiaSpend supra note 34: Among major states, Bihar recorded the highest undertrial share at 87.2%, Delhi at 88%, Jammu & Kashmir at 84.6%, and Maharashtra at 80.8%.
[46]Supreme Court of India, Centre for Research & Planning, Prisons in India: Mapping Prison Manuals and Measures for Reformation and Decongestion (November 2024, Revised Edition November 2025): As of 31 December 2023, 24,879 accused persons who were granted bail continued to remain in prison due to their inability to furnish bail bonds.
[47]Supreme Court of India, Centre for Research & Planning, supra note 46, Bail rejection rate in Sessions Courts is 32.3% and in Magisterial Courts is 16.2%. Over 52% of cases where accused were in custody for more than a year were still pending at evidence stage in 2023.
[48]National Crime Records Bureau, Prison Statistics India 2024, supra note 2, Sanctioned strength of jail staff was 99,758; actual strength was 63,004 (vacancy rate approximately 37%).
[49]International Covenant on Civil and Political Rights, art 9(3).
[50]United Nations Human Rights Committee, General Comment No. 35 on Article 9 (Liberty and Security of Person), CCPR/C/GC/35 (2014) para 38.
[51]United Nations Standard Minimum Rules for Non-Custodial Measures (The Tokyo Rules), G.A. Res. 45/110, Principle 6.1 (Dec. 14, 1990).
[52]United Nations Office on Drugs and Crime, Handbook on Strategies to Reduce Overcrowding in Prisons, Criminal Justice Handbook Series (Vienna, 2013).
[53]United Nations System, Common Position on Incarceration (Apr. 2021).
[54]American Bar Association Rule of Law Initiative, Handbook of International Standards on Pretrial Detention Procedure 7 (2010).
[55]Council of Europe, Recommendation Rec(2006)13 of the Committee of Ministers on the Use of Remand in Custody.
[56]Bail Act 1976, c. 63, § 4 (U.K.).
[57]Bail Act 1976, c. 63, sch. 1, pt. I (U.K.).
[58]New Jersey Criminal Justice Reform Act, NJ Rev Stat § 2A:162-15 (2017).
[59]New Jersey Judiciary, Administrative Office of the Courts, Criminal Justice Reform: Annual Report to the Governor and the Legislature 2018 (Trenton, N.J.: The Judiciary, 2019).
[60]Third Way, Analyzing Cash Bail Reform (July 11, 2023); Arnold Ventures, New Jersey Bail Reform Fact Sheet (2020).
[61]Sunny Chauhan v. State of Haryana, supra note 7.
[62]Arnesh Kumar v. State of Bihar, supra note 22.
[63]National Crime Records Bureau, Prison Statistics India 2024, supra note 2.
[64]National Legal Services Authority, Annual Report 2022-23 (NALSA, New Delhi).
[65]India Justice Report 2025 (Tata Trusts & Partners, 2025); India Justice Report Budget Study 2025 (noting legal aid constitutes 0.49% of total justice budget for 11 major states in fiscal 2024-25 BE).
[66]india Justice Report 2025, supra note 66. NALSA’s own funds declined from ₹207 crore (2017-18) to ₹169 crore (2022-23); fund utilisation dropped from 75% to 59% due to rigid expenditure restrictions under the NALSA Manual 2023.
[67]India Justice Report 2025, supra note 66.Para-legal volunteer density fell by 38% between 2019 and 2024, from 5.7 to 3.1 per lakh population. Most PLVs receive honorariums below minimum wage. Only 15.5 lakh persons accessed legal aid in 2023-24 despite nearly 80% of India’s population being eligible. National per capita spending on legal aid: ₹7.
[68]Supreme Court of India, Centre for Research & Planning, Prisons in India: Mapping Prison Manuals and Measures for Reformation and Decongestion, supra note 46.
[69]Department-Related Parliamentary Standing Committee on Home Affairs, Rajya Sabha, 245th Report: Prison-Conditions, Infrastructure and Reforms (Rajya Sabha Secretariat, New Delhi, Sept. 2023).
[70]Department-Related Parliamentary Standing Committee on Home Affairs, Rajya Sabha, supra note 70; Lok Sabha Unstarred Question No. 501, Ministry of Home Affairs, Government of India.