Handling communal riots


back to issue

THE National Police Academy, Hyderabad, awarded me a fellowship to conduct a study on the neutrality of the police during communal riots, especially between Hindus and Muslims in India. During the course of my study, I encountered some disturbing trends in its behaviour. In most parts of the country, the relationship between the police and Muslims was inimical and community perception of the police in situations of communal tension was that of an enemy. This is true not only for the post Independence period; pre-Partition Indian society too expected the police to behave in a communal fashion. For a policeman, Hindu or Muslim, continued to be looked upon primarily as a protector of his own community.

While working on the project I came across two interesting incidents from pre-Partition days. The riots in East Bengal during the ’20s and ’30s were abetted by Muslim policemen spreading rumours among the Muslim peasants that attacks on Hindus would be considered as acts of loyalty to the Raj. Further, there was an agreement between the Nawab of Dhaka and the Emperor of Britain that attacks on Hindus would not attract any punishment. I also came across a petition submitted in the ’30s by one Pandit Raghuvar Dayal of Kanpur that Hindu citizens of the town felt insecure because of a lower representation in the Kanpur police. These two instances exemplified the dominant trend in Indian society of the time.



Unfortunately, the situation has not changed significantly and the relationship between citizens of a particular religion with policemen of the other religion remains more or less the same. We need to examine minority fears regarding the behaviour of the police keeping these ‘facts’ in view. Is the current behaviour of the police and the reaction of the minorities just an extension of the earlier trend? Is the Muslim perception regarding the police based on certain realities or is their behaviour too responsible? Why are the perceptions of Hindus and Muslims about the Indian Police so diametrically opposed to each other? Muslims in India consider the police as their enemy; the Hindus see them as friends and protectors. The answers to these questions have to be sought in the behaviour of the police combatting communal riots, the representation of minorities in the police, and conflicting expectations of different segments of society in any given situation.

We should first analyse the efforts of the police to quell incidents of communal violence. Like with any other law and order problem, police efforts to cope with the situation can also be divided into many stages. Collection of intelligence and preventive actions – detention of anti-social and communal elements, execution of bonds, instilling fear in the minds of mischief-mongers through show of force, and diffusing tension through reconciliatory measures, form the first stage of police strategy.



The second stage of police action begins with the eruption of violence. This includes actual use of force – lathi-charge, firing, arrests, imposition of curfew and extension of protection to the victims of violence. The third and final stage involves measures like investigation and prosecution of riot cases, rehabilitation of riot victims, necessary arrangements to ensure that there is no recurrence of communal violence and rebuilding of confidence among the people.

The neutrality of police behaviour and its relationship with members of different communities can be understood better only after analysing police actions during the above three stages. It is basically the overall behaviour of the police in situations of communal strife which pushes members of a minority community, like the Muslims, into viewing it as an enemy.

I was stunned to discover that in most major communal riots in the country, Muslims were the worst sufferers, both in terms of loss of life and property. Often, the percentage of Muslim casualties was more than 60% of the total. Their losses in terms of property were in similar proportion. Given these facts, it is not unnatural to expect that the law enforcing agencies would react in a manner commensurate with this reality.

Unfortunately, the real picture is quite different. Even in riots where the number of Muslims killed was many times more than the Hindus, it was they who were mainly arrested, most searches were conducted in their houses, and curfew imposed in a harsher manner in their localities. This observation holds good for even those riots where almost all killed were Muslims, e.g., Ahmedabad (1969), Bhiwandi (1970) or Bhagalpur (1989). This phenomenon can be better understood through the accompanying table.


Arrests and Casualties Hindu Muslim

Bhiwandi Riots Arrested in cognisable/

(1970) substantive offences 21 901

Casualties 17 59


Meerut Riots (1982) Arrested in cognisable/

upto15 September substantive offences 124 231

Casualties 2 8




Similarly, Muslims are often at the receiving end during house searches. The general pattern during a communal riot is that a Muslim mohalla is cordoned off with the help of the army or para-military forces after which the houses are searched indiscriminately. Such acts only result in injuring the pride of the entire community. What is more disturbing is the mind-frame of the civil and police administration. While the curfew is enforced with all strictness in the Muslim localities, it is virtually confined to the main roads in Hindu areas, with normal activity in the lanes and by-lanes remaining unaffected.

In interviews with the riot victims of Ahmedabad, Meerut, Bombay and Allahabad, this single factor came across as the most important in explaining Muslim anger towards the police. This complaint of discrimination was more bitter in areas of adjoining Hindu and Muslim residential townships. Further, the experience of curfew was different for the poor residents of slum areas belonging to the two communities. Most houses lack basic facilities such as drinking water and lavatories. The Muslims invariably complained that while they were not permitted to move out of their houses to fetch water from public taps, which happen to be the main source of water supply in such areas, the Hindus were rarely subjected to such restrictions.



An analysis of the number of victims of police firing in communal riots reveals a similar trend. Normally, Muslims suffer the brunt of police firing. The table below shows that Muslims suffer differentially in police firing even in those riots where they have already suffered far more than Hindus in the violence.


Number of persons killed in police firing

Place Hindu Muslim

Bhiwandi (1970) nil 9

Firozabad (1972) nil 6

Aligarh (1978) nil 7

Meerut (1982) nil 6


It is not difficult to identify the reasons behind the discriminatory behaviour of the police. The conduct of an average policeman is guided by the same predetermined beliefs and misconceptions which influence the mind of an average Hindu. Not unlike his average co-religionist, an average Hindu policeman too believes that Muslims by nature are generally cruel and violent. In the course of my study, I spoke to a large number of policemen of various ranks. Most held the view that apart from being cruel and violent, Muslims were untrustworthy, anti- national, easily influenced by a fanatical leadership, and capable of rioting at the slightest provocation. Further, most policemen believed that riots are initiated by the Muslims. Even when confronted with evidence that it was not in the interest of Muslims to start a riot, the arguments rarely changed.



It stands to reason that since policemen are convinced of the mischievous role of Muslims in riots, they rarely entertain doubts regarding the modalities required to check them. They believe that the only way to control riots is to crush the mischief mongering Muslims. Instructions from the state government or senior police officials to deal firmly and ruthlessly with the rioters are interpreted in a prejudiced and biased way. Being firm and ruthless with rioters is interpreted as firmness and ruthlessness towards Muslims, arrests means arrests of Muslims, search means search of Muslim houses, and police firing means firing on Muslims.

Just how strongly the subconscious is affected by the prejudices and predetermined beliefs we hold, and the degree to which our conduct is influenced by them, can be discerned from the actions of policemen during communal riots. Even in situations where Muslims were at the receiving end from the very outbreak of rioting or where the killing of Muslims was totally one-sided, the police did not hesitate in claiming that the Muslims had caused the riot. Even subsequently, after it was established that the Muslims had suffered most, they continued to argue that Muslims were responsible for the outbreak of riots.

In my conversations with some of the policemen posted in Bhagalpur (1989) and Bombay (1992-93), it became clear that their perception about Muslims as violent and cruel was so deeply embedded in their psyche that even after admitting the disproportionate destruction of Muslim life and property, they continued to ‘discover’ many ‘reasons’ to dismiss the suggestion that the ‘naturally non-violent and pious Hindus’ could in any way have been responsible.

It is this psychology that governs police reactions during communal strife. While combatting riots, they look for friends among Hindus and foes among Muslims. It is a common sight in the towns of North India that outside forces sent to control communal tensions make their lodging arrangements in temples, dharmashalas and parks in Hindu localities or the space available in Hindu homes and shops. When shops are shut during curfew, food, tea and snacks are supplied to them by Hindu homes. Members of the majority community, who in normal times may maintain a distance from the police just like members of the minority communities, suddenly perceive policemen as friends. This is their ‘natural’ expectation from a ‘friendly’ police – that it will not use force against them. Whenever the police has used force against Hindus, they have reacted in amazement and behaved as though cheated.



The first information report (fir) lodged by Ajit Dutta, dig, during the Bhagalpur riots (1989), candidly underscores this mentality. He writes about the dismay and anger expressed by a mob of law-breaking Hindus when confronted by the police. Obviously, for them this was just not done. This reminds me of a similar experience at Gadiwan Tola in Allahabad (1980). I warned a Hindu mob that we would open fire if they did not disperse. The crowd refused to take the warning seriously believing it was a joke. Subsequently, when they heard the order to open fire, there was the unambiguous reaction of disbelief and surprise.

How far this deeply entrenched perception of Muslims as being solely responsible for riots and strictness towards them as the only way to quell a riot affects, the reaction of a policeman may be illustrated by the example of Hashimpura, where the savagery and horrifying lack of professionalism of the police became a matter of national shame.

The Meerut riots (1987) were unprecedented in the toll of human life and for the long period of continued and unabated violence. The magnitude of the riots can be gauged by the fact that the services of about 50 gazetted police officers and magistrates along with more than 70 companies of PAC, para-military forces and army had to be pressed into ser-vice. The policemen deployed here harboured all the above-mentioned beliefs and prejudices. When their round-the-clock vigil failed to control the violence, some of them went berserk.



Fully convinced that the only way to quell riots in a civilised society was by teaching the Muslims a lesson, one section of the PAC picked up more than two dozen Muslims from Hashimpura. They were transported in police trucks and killed at two places in Ghaziabad. I was SP Ghaziabad at the time and after receiving the information registered two cases against the PAC. The cases were handed over to the Uttar Pradesh cid and after eight years of investigations a charge sheet was reportedly filed against the erring personnel of the PAC.

Why should the PAC have committed such a detestable act? I talked to a number of policemen deployed in Meerut in this period during my tenure as sp Ghaziabad (1985-88) as well as during the course of my study. An understanding of the psychology of these men may help us better appreciate the relationship between the police and members of the minority communities.

Most of the policemen posted in Meerut thought that the riots were a result of Muslim mischief. They also believed that Meerut had become a mini-Pakistan because of Muslim intransigence; that it was necessary to teach the community a lesson in order to establish permanent peace in the city. They were deeply affected by rumours which suggested that Hindus in Meerut were totally vulnerable to Muslim attacks.



Instances like Hashimpura only worsen the already strained relationship between Muslims and the police. We find that some riots did start with a Muslim attack on the police. Often, in a surcharged atmosphere, the presence of police angers people. For instance, reacting to the demolition of the Babri mosque in Ayodhya, angry mobs of Muslims in different cities initially chose the police rather than the Hindus as a target. There are many other examples of communal rioting in which trouble started as a clash between the police and Muslims and only subsequently turned into a Hindu-Muslim conflict. The Idgah incident in Moradabad (1980) is a case in point.

The clearest reflection of the hostile relationship between Muslims and the police can be witnessed in the behaviour of the police entering a Muslim locality during communal tensions. The briefing, preparation and weaponry of the police party before entering a Muslim locality for arrests, searches or even normal patrolling is such that it thinks it is entering enemy territory. I have encountered many such groups and invariably found them comprising of people full of apprehension and fear. Their behaviour is not inexplicable. It is necessary for them to be alert, as they could be the target of attack. Who is responsible for this feeling of distrust and enmity? Perhaps, the seeds are to be found in the terms ‘we and they’ used by police officials for Hindus and Muslims during conferences organised to devise ways and means to deal with a communal situation.



The reporting of facts, the investigation into and prosecution of those involved in communal riots, are other aspects where a clear communal bias in police behaviour can be discerned. Facts are reported at various levels. Intelligence reports prepared at the police station to be sent to government and senior police officials are normally affected by this bias. For example, a list of habitually communal agitators, maintained at various police levels in Uttar Pradesh, is dominated by Muslim names. Even during the days when Hindu communal forces were active in the Ram Janmabhoomi agitation, it was difficult to find the name of Hindu inciters in the list. Perhaps the same perception which holds that to be communal is the prerogative of the Muslims was at work here.

The damage this bias does to police professionalism can be understood from the incidents that led to the destruction of the Babri mosque. It is evident from the chargesheet filed by the CBI that the demolition of the mosque was the result of a well-planned conspiracy. However, none of the intelligence agencies actually discovered this fact before 6 December 1992.

A heinous example of this bias in reporting facts is available from Bhagalpur (1989). 116 Muslims were killed in Logain village on 27 October 1989. This brutal massacre was enacted by the Hindus of Logain and other neighbouring villages. Logain stands 26 kilometers from the district headquarters of Bhagalpur, with the police station only 4 kms away at Jagdishpur. The Muslims killed were buried in the fields. The 65 Muslim survivors went to many places, including Bhagalpur town, and reported this ghastly incident. Details were published in local and national newspapers. Despite this, the district and police administration of Bhagalpur continued to deny any such incident till a police party led by dig Ajit Dutta dug out some bodies from the fields on 8 December1989.



The Justice D.P. Madon Commission which enquired into the riots of 1970 at Bhiwandi-Jalgaon cited similar examples of bias in reporting. His analysis about the failure of the police to take effective measures at Jalgaon, even after receiving the report of Bhiwandi troubles, is scathing: ‘The real reason for the inadequacy of the measures taken by the authorities was the communal bent of mind of some officers and incompetence of others. Unfortunately, SP S.T. Raman appears to have possessed a communal bent of mind and perhaps a pro-Jan Sangh bias. As shown by some of his own reports and his notings on the reports of Inspector Sawant, incharge of the Jalgaon city police station, he fully realised the seriousness of the situation. He, however, chose to turn a blind eye to it and even to mislead the government and the IGP about the true state of affairs in his report dated 29th March 1970.’

The commission found a similar bias in the conduct of PSI Bhalerao, who did not include incidents of brick-batting by Hindus in the records of the police station. The officials of the intelligence department displayed a similar bias. PSI Badgoojar sent an entirely false report to DIG (Int.) that the riot was caused by Muslims throwing burning torches on Hindu houses.

Investigating agencies too are afflicted by a communal bias while looking into riot cases. There is the classic case of Hashimpura, Meerut, cited earlier, in which the Uttar Pradesh cid took eight years to complete its investigation. Another example relates to the cases registered during the 1984 anti-Sikh riots. In most of these situations, the police organisations failed to book the culprits.

The role of the special investigation squad, set up to investigate the riot cases of Bhiwandi, was focussed on by the Madon Commission as a glaring example of communal bias. The squad’s effort to establish the theory of a Muslim conspiracy was ridiculed by the commission, finding it totally untenable. The commission highlighted many examples of investigators trying to fabricate evidence against Muslims and shielding Hindu culprits. It also cited many instances of tampering with official records in a communally biased manner.



The same communal bias on the part of state agencies is evident in their treatment of arrested persons. In any civilized society it is a well-established norm that once a person is taken into custody, it becomes the duty of the state to protect his life and provide him facilities to which he is entitled as part of his human rights. Unfortunately, there are numerous instances when the basic human rights of persons under custody are violated by police and jail officials, solely because of their communal bias. Nowhere is evidence of this bias better described than in the Justice Joseph Vithayathal Commission of Inquiry Report on the Tellicherry Disturbances (1971).

How one wishes that the above examples were simply aberrations and exceptions, and not reflective of the general behaviour of our law enforcing agencies.