Friday, March 20, 2015

Reinventing the criminal justice system (Part 2 of 2)

by Nandkumar Saravade.


In Part 1 of this article, we looked at the problems of investigative capacity. This yields a picture of a case load that has overwhelmed the criminal justice system, poor performance, and an overall lack of effectiveness in criminal investigation and prosecution. We now turn to the ways to remedy the situation.

The key issues which must be understood are capacity and resource constraints, clarity and measurement of the output of the Indian Police, the required structural changes in Police organisations the required accountability framework for that resourcing.

Guiding Principles

Record all crime: The purpose of criminal investigation is to generate deterrence by ensuring that the perpetrator is identified and brought before the court to face trial, while adhering to due process. All citizens have recourse to the law. The first step towards ensuring the rule of law is to record all complaints, irrespective of what further action taken.

There is a bias in public systems to turn away complainants. In addition, not all victims may want to report. These two problems generate bias in the official statistics. This calls for a three-pronged approach with (a) Cultural change that removes the stigma associated with being the victim of a crime; (b) Improved technology and process design through which all crime is reported and (c) Parallel measurement of crime incidence using Crime Victimisation Surveys (CVS).

Investigate selectively: Since resources are limited, not all instances of violation of laws can be investigated. The purpose of the criminal justice system is deterrence, not retribution. If perpetrators feel there is a sufficiently high probability of getting caught, and if the punishments are sufficiently severe, this will generate deterrence. For offences involving loss of property, insurance and other compensation mechanisms may be operated to substantially comfort the victim. However, all instances of offences involving bodily hurt must be investigated. Indian law separates offences into cognisable (where police register a First Information Report (FIR) and investigate) and non-cognisable (where permission of a court is required). There is no merit in this distinction.

Analyse for prevention: Data collected through direct reporting and CVS can be used to improve the allocation of scarce police presence, channeling this to locations and timepoints which maximise deterrence for the same expenditure.

Speed up prosecution: As we saw earlier, delayed trials are not only ineffective, but also counterproductive for subsequent cases. The judges:population ratio needs to be substantially higher. The Law Commission (120th Report) recommended in 1987 that this ratio needs to go up from 12.5 judges per million to 50, which is a four-fold increase (2009 data). This would still be half of that found in the US in 1999. At the same time, dramatic increases in the productivity of judges are feasible through better process engineering of courts. There may be a case for large-scale withdrawal of unviable and old cases, as a one time measure, to debottleneck courts.

Agenda for State Governments

The most important change that is required is to improve the working of the police. The police machinery is currently controlled by the party in power and manipulated for their short term objectives of 'managing the crime numbers.' The National Police Commission (NPC), set up in 1977, went into all aspects of policing in the country and gave detailed recommendations through several reports, which have not been implemented. Most of the findings and the recommendations of the NPC remain valid even today. The issue of police reforms got revived after the Supreme Court judgement on a case filed by Prakash Singh, former DGP, UP and others. It is making slow progress, though not always in the right direction.

The recommended governance model for the police involves setting up of State Security Commission (headed by the Chief Minister, the Leader of the Opposition, the Chief Secretary, members of the civil society and subject matter experts) to act as the highest policy-making and oversight body. Further, an empowered Police Establishment Board is required, to insulate police officers from political interference. A Police Complaints Authority is required, to look into allegations of unprofessional/illegal behavior of police officials.

As the police is a subject in the state list under the Constitution, all the issues of police reform -- structural change, and better resourcing -- have to be carried out by the state governments. There is a remarkable opportunity to re-ignite this agenda, owing to the recent decision of the Central Government to increase the states' share of tax revenues to 42%, which is going to dramatically enhance budgets at the level of states. As part of the devolution of greater resources to the state level, the Central Government is also exiting its interventions in state subjects, which has led to discontinuation of the Police Modernisation Scheme. The facts need to be taken together by the leadership of states, and a sharp increase in the resourcing of police is required, alongside improvements in the organisational effectiveness and accountability mechanisms. These three elements -- increased resourcing, institutional reform, and enhanced accountability -- should be seen as a package.

Vision for the Police

For police organisations themselves, the mission to improve investigation effectiveness starts at creating an effective reporting mechanism, in addition to the local police station. Currently, there is a common telephone number (Dial 100) in most states, which does not always work. However, this is meant only for emergency responses. For filing a complaint, the complainant has to travel to the police station. With appropriate amendment in the Code of Criminal Procedure (CrPC), modified processes to mandate registration of crime must be brought about, including: (a) Reporting on the spot with the police official visiting the place of occurrence and (b) offering multiple channels (telephone/email/web/paper) of interaction. Will these changes reporting will improve, fulfilling the first requirement of police being a responsive organisation.

With a business triage approach enabled through a suitable CrPC amendment, high priority investigations can be taken up and the rest, especially those relating to minor property offences, only acknowledged.

From the organisational side, it would be important to devote a sufficient number of skilled police officers, give them standard operating procedures and measure their performance more accurately, with commensurate incentives for better achievement. To improve the supply of investigators without excessive cost escalation, the current heavy bias towards low-output constabulary should shift to officer-oriented composition. At present, 15% of the staff are officers; this needs to go up to 25%. This will also improve promotions and, consequently, motivation and morale of the force, through aspiration based efforts and more meaningful job content.

Separating the investigation function from public order maintenance to promote specialisation has been a long-standing recommendation. This was (recently done in Punjab). This involves earmarking a fixed number of officers for investigative work at the police station level. These detectives cannot be deployed for public order management or security without the orders of the district police chief and unless there is a dire emergency. This protects the prioritisation of investigations.

A national Police Standards Board needs to be created, to improve police investigative processes with clear Key Performance Indicators (KPIs) and an external audititing mechanism, which can directly report to the State Security Commission (SSC), as is done by Internal Audit to the Audit Committee in a listed company.

To enable the SSC to gauge the level of crime afflicting the society, independent surveys are required to be carried out, as a standard practice. Called Crime Victimisation Surveys, these are invaluable for measuring trends on crime reporting and comparative performance of different police units. They constitute reporting of the performance of the police that is completely independent of the police, and serve as an accountability mechanism in the eyes of society, which would see performance in these surveys as the outcome of resources put into the police.

For keeping the knowledge, skill and ability of the investigators at the desirable level, modern training infrastructure is essential. The current training mechanism emphasises a long initial training period over several months, but little subsequent training. However, there is a need for continuous training all through the working years, through which experience and knowledge interventions come together to induce a spiral of capability. This can be addressed through higher budgets for training and developments of a comprehensive Learning Management System (LMS) and community building through professional certifications, Continuing Professional Education (CPE) and a Code of Ethics to promote independence and integrity.

The technology infrastructure of the police department needs to be transformed. An enterprise-scale Information Technology backbone is required, with seamless flow of communication across the police, judiciary, prisons, forensic department and prosecution. Currently, investigation generates a lot of manual and tedious paperwork. With modern devices and well-designed software, the investigator can be freed to concentrate on the actual work. Digital tools for data visualisation and case management, as well as forensic databases for fingerprints, DNA, ballistic markings, paint/glass, shoe prints and tyre marks will improve the success rate of investigations.

It would simultaneously be important to build in accountability mechanisms for the improved resourcing so that the changes in the system stay put, or become feedback loops for the next stage of improvements.

The Supreme Court has passed an order asking for mandatory registration of criminal cases without any discretion to the police officer, as well avoiding indiscriminate arrests. Adherence to this on the ground is inconsistent and low.

Summing up

In summary, it is important to treat the Criminal Justice System holistically. It is possible to get some early wins, by fast tracking trials of cases involving elected representatives and creating specialised units and courts dealing with economic crimes and violent crimes against women. However, for long-term success in establishing the rule of law, there is no alternative to improving governance and making adequate investments in quantity and quality of investigative resources and trial courts.


  1. Excellent attention to an oddly neglected topic. Thanks for such comprehensive suggestions. I always want to back up one step and ask, why has so little been done in this regard? Why is there such similarity of neglect across such a diverse set of states? What are the political economy obstacles that have prevented state legislatures from implementing important and impactful reforms that have been hanging fire since (in at least one of your examples) 1977? Maybe you would need part 3 of 2 to answer so many questions.

    1. Thanks for your comments. A nexus of vested interests (some of them in police itself) has thwarted any attempts to break the status quo. It is only now that there is some realisation that rule of law is an essential requirement for economic progress. Sustained pressure from civil society, media and judiciary will probably move the needle over the next few years.


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language. I delete any comment which does not contribute to the intellectual discussion about the blog article in question.

Please note: LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.