A sex worker in Kolkata, India in 2012. Arindam Mukherjee/Demotix. All Rights Reserved.

Horror stories of innocent young women tricked or sold into prostitution by unfeeling parents, carted across international borders, and thrown into slave-like conditions in brothels have made human trafficking one the hottest topics of the decade. This conception of trafficking has all the trappings of a popular issue: the innocent girl, sexually abused, and the villainous perpetrator, an organized crime boss. It invites a saviour mentality and acts of rescue. The rhetoric of trafficking has been augmented as of late by the concept of slavery, a hot-button idea that generates even more public outrage and donor support. But how accurate is this picture and how widespread is the abuse?

Ethnographic studies of sex workers and trafficked victims show a far more complex picture, one in which migrant smuggling and labour migration blur with what is labelled as trafficking. A young woman may leave her village in search of a job that will support her family or her children, expecting to work in a factory but discovering that sex pays better. Alternatively, she may take a job in a bar only to learn she is expected to do sex work as well. Women may be trafficked by neighbours or relatives as well as by organised crime bosses. Moreover, it appears that the majority of exploited labour do jobs other than sex work, and that they are coerced by a range of factors including poverty, kinship obligations, fear of violence, debt, and even the desire for the trappings of modernity.

The movement into victim status is often a complicated process. Some steps might be made on the basis of consent while others are relatively less free. What makes labour exploitative is similarly diverse and hard to specify. To add to the definitional morass, there are currently efforts to re-frame trafficking victims under new labels such as ‘modern-day slavery.’ The US State Department, for its part, no longer includes the criterion of cross-border mobility in its definition, even though this has long been a core principle of trafficking.

Not only are these definitions vague, overlapping, and even contradictory, but they are changing over time. This creates clear difficulties for determining who should be considered trafficked, yet agencies and advocates continue their attempts to tabulate numbers of victims and traffickers. Some count forced labourers, some sex workers, some cross-border labour migrants, and some a combination of these and other statuses such as involuntary domestic servitude and child marriage. Practical obstacles to finding people in the shadowy, secretive conditions in which such workers exist only exacerbate these problems of definition. Big numbers are necessary to draw attention to the problem, even though they are acknowledged as guesses, and they vary wildly.

For example, the U.S. State Department estimated in its 2005 Trafficking in Persons Report that 600,000 to 800,000 people are trafficked across borders every year, of whom 14,500 to 17,500 are trafficked into the US. Approximately 80 percent are women and girls and up to 50 percent are minors. Of these, there are “hundreds of thousands used in prostitution.” In the same year, the International Labour Organization estimated that 2.45 million people are trafficked and 1.05 million are trafficked into sex work from a global population of 12.3 million forced labourers. Meanwhile, Kevin Bales estimated in his 2005 book Disposable People that there are 27 million modern-day slaves.

However, the number of identified victims is a great deal smaller. For example, according to the State Department TIP Reports there were only 30,961 identified victims in 2008, the first year for which this information is provided. By 2012, there were 46,570. As of 2009, the International Organization for Migration (IOM) had about 13,500 trafficked persons representing more than 90 countries in its database of registered victims.

Despite this diversity of these numbers, a few of these estimates of trafficking victims circulate widely, gaining credibility through repetition. Some are repeated over and over in various documents until they acquire an aura of truth and are commonly cited simply as, for example, ‘US government’ data. Clearly, this is a hidden population and hard to count. Yet the vast disparity between estimates and counts of actual victims raises questions about how these numbers are produced and how big the problem is. There are undoubtedly some people victimised by the processes defined as trafficking and slavery, but how many is still unknown. The proliferation of large numbers does little to clarify the picture.

Any system of measurement confronts problems in determining how to reduce the buzzing confusion of social life to categories amenable to counting. The many systems employed today to count trafficking victims use different conceptions of trafficking and different measurement protocols. Each has an underlying theory about what the problem is and how it should be solved. Under conditions of uncertainty of this kind, a social phenomenon will ultimately come to be defined by whatever system of measurement prevails. In other words, the act of measurement creates the object of measurement. We may not know what intelligence is, but we do know that there is something that IQ tests measure that we call intelligence. Similarly, concepts such as the rule of law or failed states are broad and multi-faceted, yet are given more specific content by projects that claim to measure them in ways that permit comparisons across countries.

Thus, as scholars, international and national governments, and non-governmental organisations measure trafficking, they define it. The definition of the problem implicitly determines which policy should address it, whether rescues, labour regulations, migrant visas, information flyers at airports, or poverty reduction programmes. How things are counted has clear consequences for understanding what the problem is and what should be done about it.