Sahara desert vs Mediterranean - where do more migrants actually die?

About 10 years ago, I noticed how international organisations began to suggest that more migrants would die when trying to cross the Saharan desert than the Mediterranean Sea.

For example, members of the International Organization for Migration (IOM) shared their assumption that the death toll in the Sahara “has to be at least double those who die in the Mediterranean.”

Representatives of the UN refugee agency would also voice such a claim: “We, the international community, the UNHCR, say that for every death in the Mediterranean there are at least two in the Sahara, unknown and anonymous.”

My research has long focussed on border violence and the deadly effects of restrictive migration policies, including in the Mediterranean context, and so I was struck by these estimates, which seemed both extremely high and ambiguous.

However, that statistics on deaths at sea vary considerably and are themselves deeply contested, and that geographical designations of the Sahara or the Mediterranean are anything but specific, has not prevented the circulation of dubious estimates on ‘desert deaths’.

Given the ambiguity of the claim, media outlets reporting on migration across the Sahara would come to very different mortality figures, at times the result of simply multiplying the estimated death toll for the Mediterranean by two, even three. 

Trying to understand how this ‘Saharan knowledge claim’ first emerged and why it continues to be spread by international organisations, I began to look into statistics production on deaths in the Sahara in 2022 and my findings were published last week in the journal Social and Cultural Geography. 

When I started my research, I quickly came across a report called “Forgotten Fatalities: The number of migrant deaths before reaching the Mediterranean”, published by the Mixed Migration Centre in 2016.

This two-page-long report, credited by Reuters at the time as “probably the first attempt to gauge the size of a largely-unrecorded death toll among migrants in the Sahara”, shared findings of a survey carried out between 2014 and 2016.

Higher than at sea?

It suggested that 1,245 people had lost their lives in Libya, Sudan, and Egypt and concluded that “it would be safe to assume the number of migrants and refugees dying before reaching the shores of Egypt and Libya is even higher than the number of deaths at sea.” 

When interviewing Bram Frouws, the author of this report and director of the Mixed Migration Centre, he explained that this survey was meant to counter the void of information about deaths in the Sahara and to draw attention of policymakers and the media to the reality of suffering in the desert.

At the same time, he stressed that this survey was “not pretending to be very scientific in nature” and admitted methodological challenges and shortcomings.

Among them the difficulty of accessing people in transit, the risk of including ‘hearsay’ and double-counting.

According to Frouws, these issues, parred with growing ethical concerns when interviewing traumatised survivors ultimately led to the end of data collection in early 2020.

In the meantime, however, the claim that the report helped birthed had begun to circulate widely.

Chinese whispers and 'circular referentiality'

I noticed how the suggestion that the desert was deadlier for migrants than the sea emerged in reports and press releases of international organisations, which often referred to each other as the source. In my study, I refer to this echoing of the claim as ‘circular referentiality’, in the process of which the claim appears to have grown in authority, gaining much media attention. 

This circular referentiality was also the result of data-sharing practices between organisations.

For example, the Mixed Migration Centre would share its mortality data with the IOM, which meant that the IOM’s death toll increased significantly for a few years. For example, the figure of 54 deaths documented by the IOM alone for 2018, rose to more than 800.

In turn, when the Mixed Migration Centre ended its data practice, the IOM’s recorded death toll decreased dramatically, from 913 fatalities in 2019 to 149 a year later. While increases in the documented death toll prompted press releases and media reporting, decreases in statistics on Saharan deaths did not. 

Strikingly, even with the — deeply problematic — data of the Mixed Migration Centre included in the IOM’s mortality statistics, the overall death toll would never exceed the recoded toll for the Mediterranean Sea.

Indeed, even the deadliest year on record in the IOM’s Missing Migrants Project for the Sahara — 2017 with 1,328 deaths — has a lower toll than any of the recorded annual tolls for the Mediterranean Sea between 2014 and 2024. In fact, in my interviews, members of Missing Migrants Project were equally perplexed about the repetition of the Saharan knowledge claim by members of their own organisation.

Given that there simply is no convincing data that could back up the suggestion that more migrants die in the desert than the sea, I examined the context in which members of international organisations had voiced such claim, considering that this may shed some light on the reason why it circulated so widely.

Peak anxiety

The Saharan knowledge claim proliferated during a time of peaking anxieties over migration in Europe.

Following the problematically-termed ‘migration crisis’ of 2015, EU member states and institutions focused their energies, and resources, on preventing further migrant arrivals. The Sahara, considered a transit zone for those seeking to subsequently cross the Mediterranean to Europe, became of specific interest. 

From 2015 on, European actors implemented a range of border control and anti-smuggling measures in the Sahel region.

For human rights groups, this ‘securitisation’ of the desert meant that the once legal and much safer transport economy was pushed ‘underground’, with migration routes becoming increasingly dangerous. Indeed, critics consider Europe’s externalised border control as one of the primary reasons why the desert has become “an open-air tomb”.

It is at this time and in this context that the Saharan knowledge claim came to be spread by members of the IOM and the UNHCR.

In the context of the ‘migration crisis’ in Europe, European funding windows opened widely and the IOM experienced “an exponential growth of operational activity” in the Sahara. Though framing these activities predominantly as humanitarian responses, it has long been shown that the organisation has become deeply implicated in border control activities, including in the training of local security forces or even the construction of new border posts. 

This can be considered as forms of ‘deterrence humanitarianism’: practices that whilst underpinned by humanitarian rationales, primarily the saving of migrant lives at risk, aim to hamper migrant mobilities.

The vague but powerful claim that twice or possibly even three times as many migrants die in the desert than the sea can be considered as an ideal discursive vehicle to emphasise the urgent need for the increased presence and role of international organisations in the Sahel region – both to better ‘manage’ migration and to prevent deaths in the desert. 

Of course, all of this is not to question a devastatingly deadly reality in the desert. Testimonies of migrant survivors provide glimpses into the horrors that occur in this space. Still, we will never know how many lives have been lost in the attempt to migrate across the Sahara.

When we look at the career of the Saharan knowledge claim, we need to remind ourselves that data on migrant death is not ‘innocent’.

That the suggested death toll has received no critical scrutiny and continues to be spread in the reports of researchers, journalists and international organisations should worry us.

This should worry us especially in times when deaths at borders are politically instrumentalised to legitimise increased border controls – controls that have been at the heart of producing the deadly conditions of migration in the first place. 

This year, we turn 25 and are looking for 2,500 new supporting members to take their stake in EU democracy. A functioning EU relies on a well-informed public – you.