The price we pay

Naomi Hossain looks at the consequences of coping with crime and violence in Bangladesh

One of the great tragedies of Cyclone Sidr was that many of those who died knew they were at risk, yet stayed, fearful of looters. I wondered, as you may have done (safe in my brick-built building screened by iron grilles and 24-hour security personnel) why those who stayed risked their lives to protect their cows and crops and sticks of simple furniture. I doubt I will ever understand those calculations, but based on research I have been part of over the last year, am gradually concluding that Sidr syndrome is characteristic of how we cope with threats to our security in Bangladesh.

My colleagues and I spent much of 2007 studying crime and violence in Bangladesh: how people talk about it; the experiences and concerns of women, the urban poor and religious and cultural minorities; what people and communities do to protect against and cope with the problems. This has all been part of our attempt to get to grips with what we now know as "human security" -- particularly what Amartya Sen has called the "downside risks" that afflict the poorest most. Pending the results of a large survey that will help us pin down more precisely how prevalent crime and violence are in Bangladesh (due in early 2008), one conclusion has been that people's solutions to crime and violence in Bangladesh are almost as detrimental as the problems themselves.

Photo: Tanvir Ahmed/ DRIKNEWS

For that, we have to blame the inadequacy of crime prevention and the remoteness of possibilities of redress. Responses to persistent criminal threats are hard-wired into our social structures and coping mechanisms, so deeply ingrained that we don't even notice them for what they largely are: the unbearable cost of preventing and coping with the threat of violence and unlawful loss. Crucially, the high cost of coping with crime and violence reflects the fact not so much that we are a particularly criminal or violent society -- Bangladesh is no Sicily or Waziristan -- as two other facts. First, the near certainty that criminal and violent acts will be punished by the law rarely and at great cost, and then only if victims are, and perpetrators not, well-connected. Second, that victims will rarely if ever receive compensation or restitution that they often desperately need. The poor have particularly strong reasons for guarding against crime with all their might, because any losses they suffer are likely to be significant and unrecoverable blows.

Because our responses are so embedded in social behaviour, crime and violence are a bit like the elephant in the room: too big and too close to make it easy to focus on. A colleague, the researcher Tariq Ali, told me that when he visited poor villages in Nilphamari to study rural theft in 2004 he was repeatedly told "we don't have a crime problem here." This was followed swiftly by variants of "we used to have a crime problem, but the thieves have all died, become good, or gone away." Tariq persisted, until he found that only between one-tenth and one-quarter of all the rural households he interviewed had not fallen victim to some form of theft, ranging from petty to major, in the previous two years. Hardly the crime-free sanctuary of Domar villagers' imagination.

But nor was this precisely a fabrication: people appeared to be inured to the fact of what were mainly tiny opportunistic thefts during the hungry or wet seasons. They had developed a whole battery of systems for preventing, coping, and punishing that were part of their daily routines. These were costly and time-consuming, often led to risk-averse and low-return economic activities, and the punishments for repeat offenders were gory, humiliating, and barely shy of the lynch-mob.

But they were not interesting enough to discuss, and there was no wishful thinking around "if only the police would do a better job." (Having the police visit your village remains a source of deep shame). This, it seems, is how you live in rural Bangladesh. You sleep unhygienically next to your cow so it is not stolen overnight. You marry your daughter off before she sprouts breasts so as to avert her rape. You flatter and pay off the local chairman so next hungry season he will recover your stolen crop for you. You never really get ahead of the game because you have cut off your nose to spite your face.

Even as I write this, I feel I may be unhelpfully overstating the problem, and need to claw it back. Is crime really such a threat that the choices of ordinary Bangladeshis place them between the rock of losing their worldly goods and the hard place of risky acts of protection? The answer, during Cyclone Sidr, was yes. But its not that simple: the threats people perceive do not follow directly from the percentage of muggings or dacoities per capita per annum (for example). The answer, more generally, is that we don't know. But -- with the notable exception of violence against women in which evidence suggests we are world leaders -- the prevalence of crime and violence is probably not all that much greater than in other poor countries at similar levels of development to Bangladesh. And there are many places -- richer and poorer -- where it is worse.

On what basis do I make such a claim? Other than official crime statistics, and I suspect even the police don't think those get within a whiff of the real situation, there have been two recent attempts to assess the prevalence of crime. Both were urban studies. One by Care, in which the eminent Bangladeshi researcher Akhter Ahmed of IFPRI was involved, found that 16 per cent of poor urban households in Dinajpur had been victims of a crime in the previous 12 months -- a standard measure of crime prevalence. This is our best estimate of crime victimisation in Bangladesh, and suggests we may not have an unusually significant problem: comparable recent figures are 40 per cent for Nigeria; 29 for Malawi; 34 for Chile; 30 for Nicaragua; and 15 per cent across 18 European Union countries.

A study produced by the World Bank in 2007 suggested a far scarier scenario: that an astonishing 93 percent of the urban slum population had been "affected by crime or violence" in the previous year. On reflection, the surprise is more that 7 per cent had escaped such impact -- surely everyone comes into contact with crime and its victims and is by that definition "affected by crime"? The really interesting finding of that research was only revealed under closer scrutiny (so often the case with World Bank research), when it emerged that an undisclosed yet apparently low percentage of slum-dwellers had reported having personally been victims of crime in the previous year. Obviously dissatisfied with their research findings about low crime rates in Bangladesh (possibly these were at variance with the policy prescriptions of the report) the 93 per cent "affected by crime" figure was given greater prominence, until -- as was intended -- it slid into the popular imagination as the shocking finding that 93 per cent of urban slum-dwellers had personally experienced crime!!!

But sensationalism does not detract from the important finding in the World Bank report that concerns and insecurities about crime are extremely widespread. The question remains, what is the source of insecurity, if the likelihood of falling victim to crime is not all that great? The answer, I believe, must be that if crime levels are lower than we expect, it is because people make great and costly sacrifices to protect themselves; if they feel insecure, it is because they know they must continue to do so, because if they fall victim to crime, their chances of redress or justice are slim.

The situation is different when it comes to violence against women, as a recent study by WHO shows that Bangladesh has particularly high levels of domestic violence, with well over half of all women reporting having experienced some form of violence at the hands of their husband. The only good news appears to be the growing willingness of Bangladeshi women to talk about the problem. It would be ironic were it not plainly tragic that marriage is the solution parents alight on for their daughters as a sanctuary against the pervasive problem of sexual harassment and the ever-present fear of sexual violence. Early marriage, withdrawing girls from education and preventing them from working are some of the unbearable costs of coping with threats of harassment and violence. In our research, we were particularly struck by the unquantifiable psychological costs young women bear from protecting their reputation while also trying to get ahead in school or work, with the mobility and public interaction that entails.

What both the Care and the World Bank study also found was echoed in the research my colleagues have been involved with. This is that crime prevalence is highly localised: some areas have a real problem, and others, quite close by, do not. Why should this be the case? The most obvious explanation has to do with how local community action takes up the slack of state failure. In some villages and slums, there are extraordinary stories of villagers pulling together to present a united front against drug abuse and criminality, raids by bandits and dacoits. There are heart-warming stories of people uniting against a common enemy; of respected leaders -- even Union Parishad chairmen -- doing the right thing; of relatively safe spaces, where people do not fear their children's journey to school and women can use toilets without fear of harassment or rape. But community-based prevention only goes so far: it is in the roads and markets, the spaces between villages and paras where local authority has no weight, where people feel least secure. And the chairman may be a figure of great authority with the local village thief, but he is powerless against criminals from further afield. Most seriously of all, local politicians and police may not wish to take sides to protect ethnic minorities under attack when they have seen the strength of the local mob. There are clear limits to local crime prevention.

Photo: Munem Wasif/ DRIKNEWS

Human rights activists may wish to skip this section, because how communities cope with crime does not exactly feed fantasies about community participation: there is an enormous appetite for punishment and retribution that is as colourful and striking as it is abusive. Stories about beating back the dacoits in north Bengal that Tariq Ali uncovered resonate with the wonderfully undemure stories my 92 year old grandmother tells me of doing lorai with dacoits who broke her front teeth and stole her jewellery in the wilderness of post-independence rural Rangpur. 1970s north Bengal was no haven of law and order, no doubt fuelled by poverty and remoteness, pre-Jamuna bridge. And no holds were barred in the stories Tariq brought back about how particularly "effective" chairmen dealt with repeat offenders, including the jat chor, a pale imitation of the criminal castes created by Raj statutes, the thugees of Indiana Jones and the Temple of Doom. The occasional permanent disability or death was a price widely felt justified in pursuit of crime-free localities.

But that was then. Whatever we may wish to believe about the deficits of the post-independence period, Bangladesh in the 2000s has not been a post-war frontier zone of armed gangs. Yet community strategies for coping with crime and its effects are still not pretty: in one recent account our research uncovered, a family-sponsored effort to wean a young man off heroin went wrong, and the young addict died. This was considered an understandable price that had to be paid in the fight against drug abuse, and no action was taken. Mapping community prevention initiatives in 2007, colleagues returned with numerous accounts of violence and humiliation -- most memorably the practice of "cross-cut" -- of shaving shapes into the skulls of apprehended thieves -- which are used, apparently to great effect, against trouble-makers and thieves. Beating is of course routine, and a first-time offender may be lucky to escape with his life, so severe is the first beating.

The news is not all bad from the frontline. Politically-backed gangsters no longer carry on their illegal business with impunity. In one Dhaka slum, a fully-functioning community policing initiative has secured the area against drug peddlers and thieves. No NGOs were involved -- indeed NGOs appear to have largely neglected community safety issues to date -- but ward and community political leaders worked with the police and community members to make it happen. Despite the misgivings of the human rights lobby, the Rapid Action Battalion has been and remains widely popular, and with good reason: many communities are felt to be safer than they once were. But there are questions about the extent to which the rule of law can be strengthened in a society which is hard-wired to protect itself in ways that themselves flout the law. Human security is the most basic of public priorities and quite as vital as national security. But the frontline is not the border -- it is in each community and every home.

Naomi Hossain coordinates governance research for Brac's Research and Evaluation Division. A full account of the research mentioned here will be published in the forthcoming State of Governance in Bangladesh 2007 Report by the Institute of Governance Studies at Brac University. Any views expressed in this article are the author's alone.