The conviction rate for every kind of crime is in free fall, engendering a breakdown of law that no republic can survive

Even criminals, back in 1953, seemed to be soaking in the warm, hope-filled glow that suffused the newly free India. From a peak of 654,019 in 1949, the number of crimes had declined year-on-year to 601,964. Murderers and dacoits; house-breakers and robbers — all were showing declining enthusiasm for crime. Large-scale communal violence, which had torn apart the nation at the moment of its birth, appeared to be a fading memory. Bar a Calcutta tram workers’ strike, which had paralysed the city for three weeks, there was no large-scale violence at all.

The sun wasn’t shining in the stone-clad corridors of New Delhi’s North Block, though, where police officials had just completed the country’s first national crime survey — the National Crime Records Bureau’s now-annual Crime in India.

India, they concluded, faced a crisis of criminal justice. For one, India faced a crippling shortage of police officers. Then, poor training standards meant “there had been no improvement in the methods of investigation”. “No facilities exist in any of the rural police stations and even in most of the urban police stations for scientific investigation,” the report went on, “there had been a fall in the standard of work”. The result, Crime in India, 1953 recorded, was plain: intelligence capacities had diminished; cases were failing; criminals walking free.

Stinging indictment

Last month, the Jamia Teachers Solidarity Association, a New Delhi-based human rights advocacy group, brought out a stinging indictment of policing in independent India. In studies of 16 cases involving the Delhi Police’s élite counter-terrorism unit, the Special Cell, the report found evidence of illegal detention, fabricated evidence and torture. Each case, the report states, ended in acquittal — but not before protracted trials destroyed the lives of suspects.

Delhi Police officials have responded by arguing that the report cherry-picks cases where the prosecution collapsed. Sixty eight per cent of the 182 individuals tried for terrorism-related crimes from 1992 have been convicted. In addition, they claimed to have secured convictions in six of the 16 cases of illegal possession of arms and explosives.

This line of defence is profoundly wrong-headed. Even if there was evidence that even one of the 16 suspects was framed or wrongfully prosecuted, that in itself would constitute a scandal. The Delhi Police’s failure to initiate an independent review of the cases does the force no credit.

In using the cases to argue that terrorism related prosecutions are driven by communal malice, though, the JTSA study falls into serious errors of its own. The stark truth is that convictions for every kind of crime are in free fall, engendering a state of criminal injustice no republic can tolerate and hope to survive.

Figures on rape prosecutions graphically demonstrate the need for caution before making the deductive leap that the police are simply framing innocents to serve communal biases, or hide their incompetence. In 2003, less than a quarter of alleged rapists were eventually convicted. In most Indian rape prosecutions, the testimony of victims is key. To suggest that the high levels of acquittals are evidence of the framing of suspects by the police would be to suggest that a large percentage of women who file rape complaints are lying — a self-evidently ridiculous proposition in our social context.

It is entirely possible that another kind of police bias — against women — might account for this high level of acquittals; male-chauvinist police officers would, after all, conduct poor investigations. It isn’t only alleged rapists, though, who are being acquitted in record numbers. Kidnapping convictions have fallen from 48 per cent in 1953 to 27 per cent in 2011; successful robbery prosecutions from 47 per cent to 29. In 2003, less than a third of completed murder trials ended in a conviction; in 2011, the last year for which data has been published, the figure remained under 40 per cent (see table).

Thus, a 30 per cent conviction rate in terrorism cases — a widely-used figure, albeit of uncertain empirical provenance, that also finds mention in the JTSA report — would be entirely consistent with the overall police record. From the NCRB’s crime data, this much is evident: if the Delhi Police have indeed secured convictions in 68 per cent of terrorism cases since 1992, it is a sign of stellar competence.

Interestingly, the police have had stellar results in explosives and arms cases, where cases revolve around material recoveries: half or more of all prosecutions since 1983 have ended in convictions. This points us in the direction of the real malaise. Investigators seem good at sniffing out hidden guns and bombs, sometimes after crudely beating information out of suspects, but not so competent in the complex process of marshalling a chain of credible evidence.

This is not to suggest that there is no bias in policing. In 2010, the last year for which NCRB data on India’s prison population is available, 17.74 per cent of the 125,789 convicts in the country’s prisons were Muslims — somewhat higher than their share of population, which the 2001 census put at 13 per cent, and is now estimated to be over 14 per cent. The overrepresentation of Muslims among prisoners facing trial was even more marked: 22.2 per cent of 240,098 that year shared this religious affiliation.

Evidence of bias

These figures make clear that Muslims are not only disproportionately likely to be convicted for an offence but also more likely to be arrested for a crime for which they are eventually acquitted.

For two reasons, though, it is unclear that communal chauvinism alone accounts for this overrepresentation. First, Muslims were significantly overrepresented among the prison population in some States with a record of non-communal administration. In West Bengal, for example, 5,722 of 12,361 prisoners under trial were Muslims — a staggering 46 per cent, against a share of the general population of around 25 per cent. Uttar Pradesh had 15,510 Muslim prisoners under trial, out of a total of 55,872 — 27 per cent — though the religious community made up less than 20 per cent of its population in 2001.

West Bengal’s numbers were not dramatically different from highly communalised Gujarat. There, 18.3 per cent of prisoners under trial in 2010 were Muslims, who make up just 9 per cent of the population. Even Gujarat did better than Maharashtra, where a staggering 32 per cent of prisoners under trial, and almost 31 per cent of convicts, are Muslims — though just over 10 per cent of the population are of that faith.

Muslims, secondly, aren’t the most overrepresented category in Indian jails. Just over 1 per cent of India’s 365,887 undertrial prisoners and convicts held post-graduate qualifications; three quarters were either illiterate or had failed to pass the 10th grade. Three-fifths of Bihar’s 5,260 convicts serving time in 2010, for example, belonged to the Scheduled Tribes, the Scheduled Castes or the Other Backward Classes — a pattern evident in many States. Muslims are among the poorest and most educationally deprived segments of India’s population, a fact of significance.

Finally, local factors — for example, the historic character of organised crime in Mumbai or Ahmedabad — might have played a role in the making of these figures, too. It ought to be no surprise that Hindus would account for a high share of terrorism suspects in Manipur or Assam; nor that Muslims might be overrepresented in Jammu and Kashmir or Sikhs in Punjab.

High quality empirical studies to establish just how much communal bias influences the criminal justice system are desperately needed — and their absence is evidence of the chronic deficits in the policing system as a whole.

The bottom line is this: even as far too many innocent people are ending up suffering punishment for crimes they never committed, even greater numbers are walking free after perpetrating hideous acts of violence.

Every word of the authors of the 1953 Crime in India could be republished in a crime survey today without emendation. The police remain understaffed, under-trained and under-equipped to conduct meaningful investigations.

Dismal ratio

Little has been done to address chronic deficits in staffing either. Last year, the Union Home Ministry told Parliament it had pushed up the ratio of police officers to the population to 174:100,000, inching towards the global norm of 250:100,000 or more, and up from a painfully-anaemic 121:100,000. It neglected to note, though, that its claims were based on the 2001 census; adjusted for population growth, the ratio is still an appalling 134:100,000.

Forensic facilities also remain rudimentary and training is cursory: it bears recalling that the forensic evidence that links 26/11 gunman Muhammad Ajmal Kasab to his handlers in Pakistan emerged as a result of work by the Federal Bureau of Investigation, not India’s police. The central academy for police investigation skills proposed in 1953 still doesn’t exist.

Ever since 26/11, there has been big talk on police reform — but precious little action. The results are evident, scarring the life of every citizen.

praveen.swami@thehindu.co.in

More In: Lead | Opinion