Analysis of data reveals no correlation between the level of urbanisation and the rate of reported rape. Law and policymaking must engage with evidence on incidence and distribution of crime.
The gangrape of the Delhi physiotherapy student in December last year sparked a national debate on the incidence and causes of rape and other sexual offences in India. One strand of this debate has been the assertion that rapes are a product of “urbanisation” and “Western values”, and occur with greater frequency and intensity in urban rather than rural India. RSS chief Mohan Bhagwat’s observation that rapes occur in India but not Bharat is representative of this view.
Arguments of this nature make two distinct claims: first, that rapes occur more frequently in urban areas; and second, that the best explanation for this variation in the incidence of rape is a difference in cultural and social factors between urban and rural areas. Before we inquire into what cultural and social factors are at work, we must first establish that there in fact is a higher rate of rape incidence in urban areas. Here, we summarise the findings of a working paper recently published by the law, governance and development initiative at the Azim Premji University (available in full at http:// goo.gl/tXVfx), which shows that NCRB data does not support the claim that urban rates of rape are higher than rural rates; appellate court outcomes offer no reliable guide to the rates of rape incidence; and third, unless we conduct victim surveys, we cannot develop a complete picture of crime.
The National Crime Records Bureau (NCRB) is the only provider of aggregate crime data in the country. The perception that rates of rape are higher in urban areas arose partly due to NCRB reports showing rape rates to be higher in mega-cities. The NCRB acknowledges that this is not an adequate method to classify crime data on the basis of its urban or rural nature, as there are numerous urban areas in the country other than the 53 mega-cities. Nevertheless, the NCRB has not developed an alternate framework to explain the geographical distribution of crime leading to the widespread perception that rape is predominantly an urban crime. Our analysis of district-level data on reported instances of rape provided by the NCRB, collated with census data on the level of urbanisation of these districts, reveals no discernible correlation between the level of urbanisation and the rate of reported rape incidence in the districts across the country, as seen from the scatter plot.
On the contrary, many states appear to show an inverse relationship between these two variables, such that in these states, districts with higher urbanisation report lower rates of rape incidence. In several states, districts with higher urbanisation report higher rates of rape until they reach levels of 40 per cent urbanisation, after which districts at still higher levels of urbanisation report lower rates of rape. Hence, the claim that higher levels of urbanisation result in an increase in rape crimes is not sustainable by available evidence on such crimes. Instead, we observe a regional pattern where districts with the highest rates of reported rape incidence appear to be clustered in central India and the northeastern states. The question of why this pattern exists is a serious one that deserves careful analysis.
Recent news reports have gone beyond the NCRB and used 25 years of data on convictions reported in the Criminal Law Journal to show that rape occurs more frequently in rural rather than urban areas. This data source cannot support any claim about the incidence of rape for two reasons. First, cases in which convictions are secured and reported in the Criminal Law Journal form a small fraction of the cases entering the criminal justice system, while completely ignoring those that are unreported to the police. In 2011, 36,156 cases of rape were reported to the police, while conviction was secured in only around 4,072 trial and appellate cases. Even fewer than these would be reported in the Criminal Law Journal, as it publishes selected decisions of the high courts and the Supreme Court. Second, the nature of the different stages in the criminal process ensures that the cases emerging at the end of the process, such as those reported in the Criminal Law Journal are not necessarily representative of those being registered by the police at the beginning of the criminal process. The availability of competent legal representation, the choices of the victim, and the proficiency of the prosecution, are all factors among others that have a strong role to play in determining which cases pass through the various stages of the criminal process. For these reasons, it is both misleading and inaccurate to make empirical claims as to the greater incidence of rape in rural or urban areas on the basis of appellate court data.
A major drawback of existing crime data is that it only takes into account police recorded crime. In order to develop an accurate sociological understanding of the incidence and distribution of crime in India, it is critical to go beyond this data. The under-reporting of crimes against women in India specifically has been noted by various observers. In a number of other countries, this data problem has been substantially remedied by conducting crime victimisation surveys. These are carried out at the household level, by asking respondents about their experience of crime during the relevant period. In the UK it is estimated that only around 15 per cent of cases of serious sexual assault are reported to the police, while in the US, only around 46 per cent of instances of rape are reported. A crime victimisation survey in India will reveal the extent of under-reporting in the country, and it is possible that there will be a significant variation in under-reporting rates in urban and rural India.
The women’s movement in India has generated the most nuanced perspectives on the relationship between law and social change. However, the popular outrage in response to the Delhi rape case has resulted in a quick legislative response. The otherwise extensive Justice J.S. Verma Committee report and the rather brief legislative debates have no meaningful engagement with the empirical data or other sociological analyses of rape crimes in India. We show that a sound empirical understanding of incidence and distribution of crime in India can provide valuable guidance for effective and meaningful law reform. Without this foundation, our law and policymaking will rest on inaccurate and unsustainable popular myths.
This article was originally published in The Indian Express on April 19th, 2013.