Does Torture Prevention Work?, by Dr Richard Carver from Oxford Brookes University and Dr Lisa Handley, an independent scholar from the United States, is the first independent and global study of the impact of torture prevention measures. The research provides important new insights into the most effective ways to reduce incidences of torture. Carver and Handley led a team of researchers in fourteen countries, and asked them to look at torture and prevention mechanisms over a thirty-year period. Their research demonstrates that torture can be prevented.

This book is important not only because of its key messages and findings but also because it fills a significant gap in the research on torture, a practice that sadly remains a prevalent concern worldwide. There is indeed little extant scholarly analysis of the impact of preventive measures on torture. The legal literature tends to be normative and usually limits itself to what States are required to do, rather than what actually works in practice. Most of what is written about torture is found in country-specific studies, drafted by non-governmental organizations, with alerts that torture is endemic, but not generally explaining why preventive measures succeed or fail.

The book is divided into eighteen chapters, with two discernable sections. The first section is composed of the first three chapters, which introduce and explain the methodology set up for the research as well as the key findings. The second section includes chapters that look at the fourteen country-specific studies: the United Kingdom, Chile, Hungary, Indonesia, Israel, Peru, South Africa, Georgia, Tunisia, Turkey, Ethiopia, India, Kyrgyzstan and the Philippines. The case studies are grouped into four main categories: countries that have seen consistent and sustained improvement (UK, Chile), countries where positive developments have stalled after initial success (Hungary, Indonesia, Israel, Peru, South Africa), countries that have made recent dramatic improvements which may or may not be sustained (Georgia, Tunisia, Turkey), and finally, countries where there has been no discernable sustained improvement over the years (Ethiopia, India, Kyrgyzstan and the Philippines). In the final chapter, the authors provide their conclusions.

This review will focus on the first section of the book, due to the importance and transversal nature of the research methodology and key findings presented therein. The methodology of the research combines quantitative and qualitative analysis to examine which torture prevention methods were the most effective during the period under review (1984–2014). The researchers identified sixty independent variables, which they divided into four main categories: detention, prosecution, monitoring and complaints mechanisms. These correspond to the main legal obligations contained in the main relevant international and regional treaties, as well as the key recommendations from human rights mechanisms.Footnote ¹ In addition, the researchers added a variable about training in each of the categories, training being recognized as an important feature of torture prevention.

The authors came up with an index to measure the incidence of torture: the Carver-Handley Torture Score (CHATS), focusing on the frequency, geographical spread and severity of torture.Footnote ² They used the torture definition contained in Article 1 of the United Nations Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or PunishmentFootnote ³  – thus excluding ill-treatment and cruel, inhuman and degrading treatment (for example, poor material conditions of detention or the use of force during demonstrations). The authors point out the “human rights information paradox”: the significant rise in incidences of torture, which could index a rise in the phenomenon, is often connected to a better documentation and reporting of those incidences, and therefore to a shift by the government towards eliminating torture.

The key finding from the research is that torture prevention works. Detention safeguards have the highest torture prevention impact, followed by prosecution and monitoring mechanisms. The interesting finding is that complaints mechanisms had no measurable impact on torture prevention.Footnote ⁴ The study found a significant gap between law and practice – particularly with respect to detention safeguards and investigation and prosecution of torturers, often determined by the political environment. The third main finding is that training has a positive impact in all areas: one of the chief recommendations is that training should be targeted at improving professional skills, particularly at institutions like police academies, and that it should not be limited to information and dissemination about human rights norms, which is often the case in practice today. Importantly, no single measure alone is sufficient to prevent torture: a holistic approach to torture prevention is therefore necessary in order to create a conducive environment where torture is less likely to occur.

When detention safeguards are applied in practice, this has the highest correlation in reducing incidences of torture. Amongst all of them, abstaining from unofficial detention and the implementation of safeguards in the first hours and days after arrest are the most important means for preventing torture. In particular, notification of relatives or friends and access to a lawyer have the greatest effect in reducing torture, closely followed by access to an independent medical examination. The study also highlights the positive impact of reducing reliance on confessions, which for obvious reasons often leads to coercion and torture. Audio and video recording during interrogation are important but are not much used in practice.

When perpetrators are consistently prosecuted, the risk of torture diminishes. However, there is a huge gap between law and practice: while most countries criminalize torture, prosecutions are rare. One of the factors that came out of the study was whether complaints were lodged with prosecuting authorities. Other important factors are disciplinary sanctions against perpetrators and the absence of amnesty laws for perpetrators.

Monitoring bodies – understood broadly to include National Preventive Mechanisms under the Optional Protocol to the Convention against Torture and other Cruel, Inhuman and Degrading Treatment or Punishment,Footnote ⁵ but also civil society organizations and international bodies – have a direct effect in reducing torture. Additionally, the ability to carry out unannounced visits and to interview detainees in private are considered key elements. The impact of monitoring bodies in detention might, however, be less significant than expected, and the focus should be on police custody rather than on prisons, which are the focus of most monitoring bodies.

Complaints mechanisms do not have a measurable impact on the prevention of torture, unless there is a specific mandate to carry out investigations and refer cases to a prosecutorial authority.

In addition to the general conclusions mentioned above, there are some interesting yet – for some – unsurprising findings. These include poor quality of data concerning incidences of torture and related issues: poor data handling, shifting definitions of torture which make it difficult to compare data from one year to the next, and torture indicators that do not necessarily indicate much about the incidence of torture. The authors recommend that relevant treaty bodies be supplemented by a format for State recording of data on torture and ill-treatment. This would certainly be helpful.

This book is important because it is the first evidence-based feedback on the many efforts of national and international actors in the field of torture prevention. The findings of the research can provide concrete and actionable material for those who work in the field of torture prevention, enabling them to focus on and invest in what works rather than only going along with a formatted response – which sadly is often the norm. It paves the way to a hopefully more tailored and efficient response. Careful attention should nevertheless be paid to context and environment analysis – the case studies highlight the discrepancies between contexts and the need to keep away from a one-size-fits-all approach. The book's message of cautious optimism will most certainly appeal to and comfort those who are actively involved in this delicate and sometimes frustrating work, where low-hanging fruits are few and far between.