In a recent article published at Allegra, Ferruccio Pastore addresses some of the problems related to counting migrant deaths in the Mediterranean. Furthermore, while stressing the arduousness of such analyses, he presents recent trends and suggests possible causes of diverging mortality rates. Here I propose some additional reflections.
Data on migrant deaths at the EU sea borders
Data on border deaths – e.g. the different datasets that have long been compiled by “United for Intercultural Action” and “Fortress Europe”, as well as the data that IOM started collecting more recently – are mostly based on press reviews. These are never fully reliable. First, because news can be overlooked; second, because different sources may provide different data regarding the same event; third, because an event can be counted more than once if reported at different times, or not be counted if not reported at all. Moreover, while the number of recovered bodies often differs according to different sources, the number of the missing is even more difficult to ascertain. This is a crucial issue given the high share of the missing on the total number of casualties (75% according to a study I carried out with regard to the Strait of Sicily for the three-year-period 2003-2005).
A research by the Vrije Universiteit Amsterdam has found that 3.188 border deaths were recorded at public registry offices in Spain, Gibraltar, Italy, Malta and Greece from 1990 to 2013. While the results provide only a small part of the picture (neither the bodies recovered by the authorities of countries of departure nor the missing persons are included), it demonstrates that some kind of official counting is possible. Therefore the authors propose the establishment of an independent authority within the Council of Europe: a European Migrant Death Observatory tasked with assembling and analysing the data collected by states under its supervision.
Collecting all information available on casualties would make the identification of bodies easier and provide relatives of dead or missing migrants with a reliable institution they could turn to when looking for information about their loved ones.
Mortality rates and possible causes
While IOM data limits the analysis to the last two years, it would make sense to include also previous years, by using homogeneous data sources for all years. However, a look at the most recent data provided by Pastore suggests that clear causal relations can hardly be established, if at all.
The author first points out that the effectiveness of search and rescue activities is a crucial determinant for mortality trends. Although statistics over short periods of time are more subjected to random fluctuations, there might well be a causal link between the end of Mare Nostrum and the strong increase in mortality in the Central Mediterranean in the first three months of 2015. However, things become more confusing if we consider that the total survival rates of 2015 and 2014 (the year of Mare Nostrum) are almost exactly the same (53.2 and 53.6 respectively). Furthermore, if we calculate the survival rate of the period April-December 2015, we find that it is clearly higher (59.0) than the 2014 rate. Pastore notes that Frontex was upgraded in Spring 2015. Moreover, NGOs (MOAS, MSF and Sea-Watch) deployed their search and rescue vessels close to Libyan waters from May, while the military operation Eunavfor Med was launched at the end of June and entered its second phase in October. Should we therefore conclude that Mare Nostrum was less effective, and that the joint action of NGOs, Frontex and Eunavfor Med should be seen as a progress? If so, then it’s striking to see the high survival rate in the Central Mediterranean during the first three months of 2016, if compared to 2015. How can this trend be explained, considering that MOAS, MSF and Sea-Watch were not operating in this period? Maybe the smugglers were more concerned about weather conditions? Maybe it was just luck? And again: unfortunately, statistics over short periods of time are more easily affected by random fluctuations.
Pastore then analyses possible causes of the mortality increase in the Eastern Mediterranean, and of the resulting decrease of the gap between this region and the Central Mediterranean, in the first three months of this year. One cause could be the “spreading habit of transporting migrants at night as a countermeasure against enhanced patrolling efforts in the Eastern Mediterranean”. Indeed, setting off in the dark increases not only the chances to escape the controls of Turkish authorities (which, I would like to stress, have been strengthened following European requests) but also the risk of accidents. According to Pastore, this is a new trend for boats leaving from Turkey, while – I would add – it is a long established tradition for boats leaving from Libya.
A further factor mentioned by Pastore is the increasing use of “low-quality, inflatable rubber dinghies” instead of more resistant wooden vessels. This is presented as a recent development in the Aegean Sea. This argument raises more than one doubt, because the Strait of Sicily has shown the same trend since last year: the smugglers resorted to cheaper vessels when Eunavfor Med started seizing the boats in order to prevent them from being re-used. If such a trend increases mortality in the Aegean, why shouldn’t it do so in the Strait of Sicily, too?
Finally, the increase in mortality in the Aegean Sea is surprising because it happens simultaneously with an increased presence of both navy ships and private rescue vessels in this region.
All this confirms that mortality rates should be handled with great care because of the too many variables contributing to determine them.
Also the recent mortality increase in the Aegean is affected by the above-mentioned problem of the low reliability of data calculated over short periods of time. On the other hand, over long periods of time the possible causes tend to increase and overlap even more, thus making the establishment of clear causal links impossible.
To conclude, I would like to stress that most of the actors involved in search and rescue at sea (Italian Navy, Frontex, Eunavfor Med, Nato) operate in the framework of policies whose aim is to prevent people from leaving for Europe. Focussing on deaths at sea – also through the analysis of mortality rates – may let us forget that preventing people from risking their lives in the Mediterranean could result in letting them die, or “simply” exposing them to gross human rights violations, elsewhere.
One thing is for sure: when Europe had not locked its doors yet, there were no dead bodies to be counted, and there would be no longer any if these doors were unlocked again.
