Refugees and the definition of Syria, 1920-1939

The wheels of academic writing turn slowly.

It’s seven years since I first gave a talk at a workshop in Princeton outlining some ideas about how the arrival and settlement of refugees in Syria helped to define the modern state’s territory, institutions, and national identity. It’s six years since I developed them more fully in a seminar at the Refugee Studies Centre in Oxford, which I entitled ‘Refugees and the definition of Syria, 1920-1939’. (The name stuck.) Over the next year or two I did some further archival research to test the ideas out, and was pleased to find that rather than contradicting my argument, this extra work allowed me to nuance and extend it. Meanwhile, just as I was learning more about Syria’s history as a destination for refugees in earlier generations, the civil war there broke out, and turned the country into the world’s largest producer of refugees.

Three years ago, while I was on research leave for a semester after changing jobs, I worked these presentations up into a full article. That turned into a bit of a monster (especially when the footnotes were included: Lordy!) but I was quite happy with it, and a couple of academic friends read it and gave me some positive feedback—as well as some advice on points that needed improving, of course. So I made some minor revisions, then sent the draft to a contact who was preparing a special issue of a historical journal, on refugees and statelessness.

And then nothing happened. For nearly a year. The person I’d been in touch with had gone on parental leave, her co-editor didn’t reply to my emails, and when I eventually contacted the journal, they couldn’t help—they’d never heard of the special issue. So I withdrew the article. By that point, two years ago, I was back to a full teaching load with plenty of other responsibilities. I didn’t know quite what to do next.

Somewhere over this horizon, your article will be published

Eventually, though, I asked another couple of (senior) colleagues to read over the article and tell me if they thought it would be worth submitting it to Past & Present—a very good journal, but one with a famously intimidating review process. Both of them thought that with a bit of reframing to make it suitable for a non-specialist audience (ie, historians who don’t specifically work on refugees and statelessness), the article would make a plausible submission. In November 2015, with a bit of free time, I gritted my teeth and made what turned out to be some fairly minor amendments to reframe the article—and, a bigger job, reformatted the footnotes in line with the requirements of a different journal. And so, in some trepidation, I was able to send it off.

This was quite a big deal for me, because by that point it had been four years since my book came out, and in the meantime I’d published nothing but book reviews. I’d started my first permanent job at an institution where I didn’t feel at home, then moved to my second—which meant two rounds of settling in to a new city, getting to grips with a new institutional culture (and new administrative responsibilities), and preparing a lot of new teaching. Finding time to research and write had been difficult, and I’d also had to change what I was working on: the war in Syria had made it impossible for me to continue a project I’d begun. I knew what I wanted to do instead, and I’d started making connections here in Glasgow (thanks to GRAMNet) that would help me develop it—but I was grimly aware that the gap opening up in my publications record was like an ever-growing question mark over my future as a researcher. Anyone working in British academia will know what I mean.

Last March, I got the reply from Past & Present: to my delight, they wanted to publish it. During the double-blind peer review process, five (!) reviewers had read the article. One of them was lukewarm, the other four were positive or very positive. They all had suggestions for minor revisions, and a kindly-worded email from the editor suggested how I might approach them. I submitted the revised final version in early July, after I’d made some amendments and got a friend who’s an academic copyeditor to check the footnotes. (I told him to charge me the full rate, of course.) Proofs came my way for checking in the autumn, along with a publication date: May 2017, seven years to the month after the workshop where I first presented the argument, with online access a bit earlier. I didn’t imagine, when I started on this work, that it would take so long to see it to completion—or that the country whose history I’d been writing would experience such catastrophe in the meantime.


All of which is by way of announcing that my article ‘Refugees and the Definition of Syria, 1920-1939’ is now available online, and will be out in print soon. For anyone who wants a shorter version without footnotes, a post will be going up on shortly. The next article I publish should be out rather quicker—though the gestation time has been almost as long.

Many thanks once again to everyone mentioned in this post
who read the article in draft form and helped improve it.

Can a refugee carry a gun?


Can a refugee carry a gun? Or, to put it another way, can someone who is armed still be considered a refugee?

The answer seems to be no. When Kurds fled Turkey in the 1920s and 30s and entered the French-controlled territory of Syria, they were one group of refugees among several: in particular, Armenians and other Anatolian Christians, including survivors of the 1915 genocide as well as people pushed out of the new Turkish Republic in the 1920s, and after 1933, Assyrians coming from Iraq. But the French authorities in Syria only referred to Christian refugees as ‘refugees’. Regardless of the circumstances of their departure from Turkey (usually fleeing military repression) and arrival in Syria (often accompanied by their flocks), Kurds were much more likely to appear in official correspondence as Kurdes réfugiés en Syrie—‘Kurds who have taken refuge in Syria’—than réfugiés kurdes, ‘Kurdish refugees’. In the lengthy reports to the League of Nations that the French foreign ministry produced each year, there is always a section about assistance to refugees: it fit the image of a benevolent mandatory power to help needy refugees. But not Kurds: they sometimes appear in summaries of the political situation, but never in the section about refugees. The League of Nations itself took action on behalf of both Armenian and Assyrian refugees in the 1920s and 30s, but not Kurds.

There are several reasons for this. But one of them is that the Kurds were usually armed.

The Baquba refugee camp, from the front cover of Austin’s book

Here’s another example. For a few years after 1918, the British military occupation forces in Mesopotamia ran a large refugee camp at Baquba, about 33 miles north-northeast of Baghdad. The people who lived in it, nearly fifty thousand of them, were Armenians and Assyrians who had been displaced from eastern Anatolia and the Caucasus during the first world war. From among the refugee population the British had formed a contingent of irregular troops, four battalions of them. But at the very start of 1920, a British attempt to make the arrangement less irregular by formally enrolling the troops created unexpected tensions. For reasons the British didn’t fully understand, troops in two of the battalions refused to sign the enrolment forms (they appeared to suspect that Britain would ship them off to serve the empire in India). One of the battalions was Armenian, and its opposition was particularly strong: when the battalion’s officers agreed, under heavy British pressure, to sign the forms, the rank and file threatened to shoot them.

This is what happened next, in the words of the camp commandant, Brigadier-General H.H. Austin (the emphasis is mine):

I decided thereupon to disarm and disband the Armenian battalion; and issued orders that the battalion should be paraded fully equipped and marched up to my Headquarters, alongside which the 1/4th Battalion Devons were camped.

Two companies of the Devons were told to hold themselves in readiness about their camp; but not to show up, as though trouble was anticipated, unless I blew a whistle to signify that my order to “ground arms” was disobeyed by the Armenians. In due course the battalion arrived, and was formed up with its rear to the river bank—here 40 or 50 feet high and a sheer drop to the water below. After addressing the men for some time through an English-speaking Armenian official of the orphanage, I informed them it was my painful duty, as a result of their distrust of the British Government, no longer to regard them as soldiers, but as refugees pure and simple. They would, accordingly, hand over their arms, accoutrements, and equipment now; and on return to their camp make over their uniform to their respective company commanders. The order to “ground arms” was obeyed without any sign of hesitation: a company of the Devons emerged from their camp to take over rifles, bandoliers, etc; and every Armenian of the battalion was searched over to see that he had no revolver or ammunition concealed about his person. They were then marched back through the Armenian sections of the camp, to their own on the other side of the river, and a few days later transferred and distributed among the Armenian population in “A” area.

H.H. Austin, The Baqubah Refugee Camp
(London and Manchester: The Faith Press, 1920), pp. 47-48

No longer soldiers, but refugees pure and simple: when refugees are disarmed, they become ‘just’ refugees.[1]

The photograph at the top of this post is evidence of this happening. In early 1939, the final territory under the control of the elected Republican government in Spain fell to the fascist military rebellion led by Franco. Over three hundred thousand people—young, old; children, women, men—fled north into France. The rifles in this pile were taken from Republican refugees as they entered the country.

There was precedent for this kind of thing. When the (substantial) remnants of the White Russian army were evacuated from the Crimea in 1921 at the end of the Russian civil war, their commander General Wrangel wanted to maintain them as a military formation to continue the fight. The Allies, hosting 120,000 Russians in the Straits Zone, disarmed and disbanded them instead. Spanish Republicans, too, hoped to continue the fight against Franco—but the French government, terrified of a war with the fascist powers, had no intention of permitting that.

Being defined as a refugee is itself a loss of control. This is one reason why many refugees reject the term: Spanish Republicans called themselves exiliados, ‘exiles’, while Russian refugees preferred to be ‘émigrés’. Here’s a Kosovo Albanian woman discussing the term, closer to our own time:

Well, you cannot describe it. It is awful, very hard to be like that. The name can show you, you know, R.E.F.U.G.E.E. is like the worst thing in the world, so it is something that you cannot describe. You don’t have any power and you don’t have anything but your soul, your body and nothing else. This is very difficult and hard for everybody. Even for the people who accepted refugees it was very hard, every time you feel like you are not you. So, every day you feel empty, you feel… I mean it’s just very hard, without any power, with nothing.

Quoted—from a book by Losi, Passerini, and Salvatici—in Peter Gatrell,
The Making of the Modern Refugee (Oxford, 2013), p. 265

The sense of powerlessness that the term R.E.F.U.G.E.E. brings is so strong that she is unwilling to speak it, and spells it out instead. But it would be a mistake to think that a person loses control at the moment when they flee their home, or cross the border into a neighbouring country.

The term ‘refugee’ has had a distinct meaning in international law for nearly a century. It has evolved in that time, and expanded from a very narrow range of applicability covering two specific groups of displaced people (Russian and Armenian refugees in the 1920s) to a theoretically universal one (anyone displaced over a border by a legitimate fear of persecution). The point of the legal definition, which is the basis of national refugee law in most places where such legislation exists, is to make protection available to refugees, in the shape of rights—eg, the right to asylum; the right not to be pushed back into the country they are fleeing; the right to work—and humanitarian assistance. In practice, though, to access that protection as ‘refugees’, people are expected to give up any control over their own destinies, and become as passive and needy as the term requires them to be. This is not always voluntary: consider Denmark’s recent decision to strip asylum-seekers of their money and belongings as a precondition of being considered for refugee status.

When refugees are armed, they have altogether too much control over their own destinies to be considered ‘refugees’. The British themselves had formed the Armenian battalion at Baquba, but when the troops showed their autonomy they were swiftly disarmed, in a setting that was designed to expose them: backs to a forty-foot drop, British soldiers waiting nearby to intervene if they failed to ground their weapons on command. They were literally ‘marched back’ into civilian life, no longer soldiers, but refugees pure and simple. For the Spanish Republicans, disarming was only the first step. They were then—as refugees often are—split up, transported long distances, and interned: women, children, and the elderly in rough accommodation rapidly converted from forestry camps or army camps, the men of military age in barbed-wire pens on the beaches of Roussillon. Many of them did not survive this exposure. The author of this loss of control was not Franco but the French state.

So: can a refugee carry a gun? Probably not. But to understand why, we need to understand the quid pro quo that states expect when they give asylum to refugees—when they define people as ‘refugees’. The point isn’t that refugees should be given guns, but what happens when they arrive with guns is a particularly clear illustration that protection, however flimsy, is conditional on loss of control.


Click images for source if not indicated

[1] Austin’s book gives an account of the events that had pushed these Assyrians and Armenians from their respective homes and brought them into contact with British troops advancing into northwestern Persia (Iran) in 1918 (pp. 3-14). Although by then some of them had been displaced over hundreds of miles, he doesn’t use the term ‘refugees’ to describe them until he reaches the point, on p.14, when they were brought under British protection.

Twilight of the saints

Vale of Nablus 1890s

This is a book review I wrote for the American Historical Review, which I’m republishing here with permission (and some pictures). A full citation for the published review follows at the end. As ever, I’m struck by how the formal tone of an academic review jars in the context of a blog, but there it is.

JAMES GREHAN. Twilight of the Saints: Everyday Religion in Ottoman Syria and Palestine. New York: Oxford University Press, 2014. Pp. x, 341. $74.00.

In 1747, when a plague of locusts threatened the harvest, the Ottoman governor of Damascus dispatched a delegation of Sufis to an enchanted spring in Persia. The water they drew there, carried carefully back to Syria, would lure a magical black bird – the samarmar – to consume the locusts. Their return was greeted with parades and popular celebration.

Well of the Samaritan 1890s
The Samaritan’s Well, Nablus, 1890s

The bird never appeared. But the fact that everyone thought it would, including Ottoman state officials and urban religious elites, is the starting point for James Grehan’s richly detailed historical ethnography of everyday religion in Ottoman Syria and Palestine. Grehan argues that histories of religion, especially in the Middle East, have focused excessively on textual traditions. They have overemphasized the salience of religious difference in everyday life, and the ability of religious institutions (the main generators and guardians of textual sources) to determine everyday religious practice. Attempts to go beyond this by studying “popular religion” have only helped up to a point: the dichotomy between “popular” and official religion still grants normative status to text-based orthodoxies, and cannot account for the prevalence of “popular” practices among educated urban elites.

Grehan sets out to offer a more nuanced account of what he terms “agrarian religion”: everyday religious practice in a predominantly rural and illiterate society, where “even the towns” – and their literate elites – “were sunk in an essentially agrarian milieu” (15). His local and western sources include topographies, travel narratives, memoirs, and (for the later part of the period) Ottoman statistical surveys. The scholar and Sufi Abd al-Ghani al-Nabulsi (1641–1731), whose writings figure often, is a particularly genial guide.

Nablus, new mosque 1940
The new mosque at Nablus, 1940

Common to all religious traditions in Ottoman Syria and Palestine was a weak infrastructure of sacred buildings and educated personnel outside the towns. Ottoman state surveys from the late nineteenth century show that mosques, churches, and synagogues, ulama, priests, and rabbis were all concentrated in towns; where villages had them, they were large ones like Jenin or were close to larger towns. Having established the weakness of institutional religion, Grehan explores the everyday religious life of the population through five thematic chapters looking at saints, tombs, sacred landscapes, the spirits that haunted the land, and the magic of blood and prayer. The chapters focus on the countryside, but return often to the towns and cities whose own religious culture was profoundly connected to that of the rural hinterland. Sunni Islam provides the richest body of evidence for Grehan’s account, but there are frequent references to other Muslim, Christian, and Jewish communities. These furnish ample material to support his argument that the lines of sectarian difference, however sharply defined in normative religious texts, were blurred to the point of indistinctness in daily life.

Saints, living and dead, were venerated by everyone. It was not uncommon for a holy man to be revered beyond his own faith community: Christians as well as Muslims would stop to kiss the hand of Ali al-Umari, a renowned Sufi in nineteenth-century Tripoli (63). In a landscape where religious buildings were rare outside towns, the tombs of saints provided a focus for religious practice, both as social institutions – places of sanctuary or mediation – and sites for worship. Different religious traditions often shared the same sites, though they sometimes disagreed over the attribution of the tomb, and even educated townsmen like al-Nabulsi saw no contradiction in reporting uncertainty over the identity of a tomb’s resident saint while praying at the site. Tombs were important in towns, too, like the shrine of Ibn al-Arabi in Damascus: there was no doubt about the identity of the person venerated there, though the actual site shifted over time (113).

Joseph's tomb 1930s
Joseph’s Tomb, Nablus, in the 1930s

Tombs belonged to a sacred landscape where stones, caves, springs, and trees were also imbued with religious meaning. Caves often became the nucleus of a church or mosque; saints’ shrines often featured holy trees, but whether the tomb or the tree was the original focus of veneration remains moot. Sacred sites generated scriptural justifications to domesticate them within one tradition or another, but nature itself was “more compelling than scripture” (116). The spirits that haunted these landscapes were familiar to all: talismans, charms, or icons could mediate human interactions with them, and dreams and visions grant more direct access to a spirit realm. Blood sacrifice and prayer offered ways of gaining saintly intercession, and not just for peasants at the limits of the state’s reach: when the Beirut–Damascus railroad was opened in 1895, “religious officials presided . . . with the usual sacrifices” (174).

Agrarian religion “pervaded everyday piety, paid only lip service to orthodoxy, and casually embraced customs and beliefs that had no warrant in scripture or law” (165). Grehan’s argument for dispensing with notions of “popular” religion is persuasive; his argument against the salience of sectarian divisions deserves to be taken seriously, too, particularly in public rather than historiographical debate, though in regard to the latter, more explicit engagement with recent scholarship on sectarianism (189 n. 126) would have been welcome. There are other points of criticism: Grehan argues that agrarian religion’s “immense stability” also permitted “discreet adaptation and invention” (16), but – because he explicitly decides not to reconstruct these patterns of change – the picture presented here is one of timelessness, though it covers two and a half centuries. Gender is not considered in any depth, nor is the survival into the present (as I have witnessed myself) of many of the beliefs and practices Grehan describes. On the editorial side, a list of images would have made the fine illustrations more accessible.

Nonetheless, this is an evocative, thought-provoking, and richly textured work. Grounded in the comparative history of religion as well as the history of the Middle East, it deserves a place on a wide range of postgraduate and advanced undergraduate reading lists.

Joseph's tomb pre-1914
Joseph’s Tomb, this time in the 1890s

Click images for links to originals,
which are all from the Library of Congress

JAMES GREHAN. Twilight of the Saints: Everyday Religion in Ottoman Syria and Palestine. 
Benjamin Thomas White
The American Historical Review 2015 120 (5): 1996-1997
doi: 10.1093/ahr/120.5.1996a

Images of refugees, part 3: refugees at sea

This is the third in a series of posts about images of refugees. For the first post, click here. For the second, click here.

Photographs of refugees on land often work to make both the refugees themselves and the landscapes they’re walking on interchangeable—so many huddled figures trudging across so many featureless bits of countryside. My last post explored some of the reasons for this: they’re partly to do with the choices that picture editors make, and partly to do with the standard formats of newspapers or news magazines and the cameras, lenses, and film that were typically used to take the photos that appeared in them. (Only rarely do refugees’ own views or choices come into it.) And in the post before that I wrote, more briefly, about the typical news photograph of a group of refugees in flight, burdened with their possessions. The aesthetic roots of that very standardized image go back much further than mid twentieth-century: I traced them back to 19th-century narrative painting, and earlier standard subjects in the European Christian tradition of painting.

But what about the other standard image of refugees, which has been just as common on news websites recently as the ‘overland trudge’—that is, the image of refugees at sea?

vlora albanians bari 1991 BARI - 1991 agosto 1991 LO SBARCO DELLA MOTONAVE VLORA CARICA DI CLANDESTINI ALBANESI - La nave vlora in porto - foto Arcieri - Quaranta
2015, 1991, 1939?

This picture shows the Albanian ship Vlora, in 1991. A cargo vessel that had recently returned to Albania from Cuba laden with sugar, the Vlora found itself heading to the Italian port of Bari with many thousands of Albanians aboard, hoping to escape the chaos of the end of communist rule. You can read about it on Migrants at sea, or watch this two-minute film on YouTube, and there’s a longer a documentary about the incident, too. The story doesn’t reflect especially well on the Italian authorities.

Vlora refugee ship black and white
‘These aren’t Syrians. They’re Europeans trying to get to North Africa during World War II. So next time you think of closing the borders you might want to check with your grandparents.’

It wasn’t as an image from 1991, though, that the picture recently went double-viral. (I read about it here.) On the one hand, it did the rounds of Twitter racists amid claims that it showed thousands of ‘migrants’ in Libya or Syria preparing to invade Europe today. On the other, in black and white, it was circulated by anti-racists on Twitter—and Tumblr—claiming that the people in it were Europeans fleeing to North Africa during the second world war.

Like the Robert Capa photo I discussed in the last post, which appears on the cover of a history book about refugees in France (and on the internet as a picture of refugees in the Spanish civil war) even though it was taken in Israel just after independence, this photo shows us that refugees are interchangeable. You can pretend that a picture of people fleeing the political uncertainty and economic misery of Albania a quarter of a century ago shows Tripoli or Tartus this summer, and some people will believe you (and retweet). Or you can put the same picture in black and white and claim it shows European refugees in the 1940s, and other people will believe you (and repost). One of those claims is intended to provoke hostility toward refugees, and the other is intended to elicit sympathy—but it’s striking that both of them reduce the refugees themselves to silence in precisely the same way. The refugees become ‘speechless emissaries’, to borrow a term from the anthropologist Liisa Malkki.* In one claim, they bear mute witness to the threat of further swarms overrunning Europe; in another, they silently represent the shared human need, and right, to flee from danger. But they never get to speak for themselves. (It’s probably fair to guess that in neither case are the people behind the claim refugees.)

One of the reasons why a single image can be used in these different ways is because the ‘refugee boat’, just like the ‘overland trudge’, is already so well-established as a visual trope. The Vlora of 1991 can stand in for boats in 2015 or 1939 because we’ve already seen refugee boats in 2015 or 1939, and every decade in between.  Look:

Rohingya refugees at sea
Could be anywhere
Vietnamese refugees Hong Kong
Could be anywhere
Exodus 1947 ship
Could be anywhere
SS Habana, Basque children
Could be anywhere

Let me restate something I wrote about images of refugees on land. The limitations of cameras, lenses, and film, the constraints of publication format, and the aesthetic (and moral) choices of photographers and picture editors all work together to mean that when you see a group of refugees in a photograph, you usually can’t see many identifying features of the landscape they’re walking across. This is even more true for images of refugees at sea: a patch of sea has even fewer identifying features than a patch of desert or hillside–if it is marked by distinctive shapes or colours, they’re changing all the time. The photos above were taken in the Andaman Sea in 2015, Hong Kong harbour in the late 1970s, Haifa in 1947, and (I think) Southampton in 1937. A dockside, if you can see one, doesn’t help much: a quick switch from colour to black and white was all it took to put the Vlora back in the same period as the Exodus 1947, carrying Holocaust survivors to Palestine, or the SS Habana, bringing Basque refugees to Britain in 1937.

When the boat is small and photographed fairly close up, you can make out some distinguishing features of the refugees, but not many: see what a difference the slightly more distant perspective in the first photo makes, compared with the second. Among the Rohingya refugees from Burma (2015) you can make out individuals, and tell adults from children; among the Vietnamese boat people in Hong Kong (1970s) you can see individual expressions, distinguishing features—but that’s rare indeed in photos that follow this trope. (Perhaps less so for paintings, and we’ll come back to that in a moment.) When the ship is large, and therefore the photographer has to be further away, even basic details are lost: for example, could you tell without looking closely that almost all the figures in the last of those four photos are children? Details of the vessel itself don’t tell you anything about where it is, either, and only very rough information about when the picture was taken. Ships travel a long way, and have long service lives: the Vlora was built in 1960 and only broken up in 1996. So when you see an image of a boatload of refugees at sea or at a dockside, there’s very little to tell you when or where the image was taken.

Massimo Sestini Refugee boat image
This could be anywhere too

All this means that the image of the ‘refugee boat’ is, if anything, even more standardized, even more of a trope, than the image of a group of refugees fleeing on foot over land. Every time you look at a photo of a refugee boat, in a way you’re looking at every other photo of a refugee boat, too—certainly every other one that you’ve seen, and every other one that whoever produced the image has seen. And every time a photojournalist frames an image of one, he or she is in a way taking a picture of all those other pictures too.

Needless to say, it’s impossible for any of these images to tell us much about the enormous variety of different individual stories, individual lives, on a single refugee boat—let alone the range between an Albanian adult on the Vlora in 1991 and a Basque child on the Habana in 1937. The image above was taken in the Mediterranean in 2014. It won the photographer, Massimo Sestini, a World Press Photo award, and in a way it was ahead of its time, seeming to capture the spirit of this summer: that’s why you may have seen it on the Google refugee appeal or, if like me you’re based in Scotland, the new Scotland Welcomes Refugees site. But in another way it could have been taken anywhere, at any time since press photography became a thing. And, like the standardized photographic image of refugees on land, the ‘refugee boat’ picture has roots that go back much deeper than the emergence of photojournalism. Here’s one very influential predecessor:

Géricault, raft of the wreck of Medusa
Where are we?

Jonathan Jones wrote about Géricault’s Raft of the Medusa (1818-19) earlier this summer, explicitly making the connection with the ‘refugee crisis’. The painting was a media sensation in its time, viewed by 40,000 people when it was exhibited at Egyptian Hall in London in 1820. (I learned about it when I read A History of the World in 10½ Chapters by Julian Barnes, twenty-odd years ago; it’s parodied in one of the Asterix books too.) A detailed exploration of the genealogy of the ‘refugee boat’ image would need an art historian, not me, but I’d suggest that this and other paintings of shipwrecks and their survivors, and the long tradition of paintings of Noah’s Ark at sea in the Flood, would be the place to start looking. Here, I’ll just point out once again that a painting can combine individual detail and panoramic sweep more easily than a press photo: Géricault’s painting is a monstrous seven metres by five (!), so the dead and dying figures are pretty much life size, if seen at a short distance.

But they are seen from a short distance, not from the raft itself–which leads to my final point. Even more than images of refugees on land (or of refugee camps), the viewpoint that pictures of refugees at sea adopt is, almost by definition, not that of the refugees. The viewer, like the photographer, is looking at the boat and the refugees from a different and usually safer perspective. The Sestini photograph is a paradigmatic case, taken from an Italian navy helicopter: not so much a bird’s-eye as a God’s-eye view.

I think it’s important to find ways to go beyond this visual trope: it objectifies the ‘refugee boat’ and it objectifies refugees (and I say that without intending to denigrate the photographers or the worthwhile ends to which such photos are often put). These images shape the meaning of ‘refugee’ before we even articulate it in words, and if that means that when we talk about refugees we immediately think of an indistinguishable mass of more or less interchangeable people, there’s a problem. Massimo Sestini seems to recognize this: in the other pictures that form part of the same reportage–here on the Time website–there are photographs of individuals, taken much closer up. But they’re all taken on navy rescue vessels. The refugees have entered the photographer’s world: he hasn’t entered theirs.

For the photographer, then, the challenge is to change their perspective, and to look at things from the refugee’s point of view. (Over a year after he took this award-winning set of photos, Sestini has started trying to locate some of the individuals pictured in the boat, so he may be doing that.) It’s a challenge for editors, too: the choice of the representative image, the one that’s at the top of the story or on the front page of the website, is the one that matters most, whether it’s a news website or a charity appeal.

But the really great challenge to this objectification of the refugee boat will come from refugees themselves. Refugees are more likely now than ever before to have the means of making their own record of their journey, and swiftly making it publicly available. We’ve heard quite a bit, in recent years, about ‘citizen journalists’ using smartphones and social media to create their own record of events. Perhaps we’ll learn to see refugee boats and their passengers differently when the photos we’re looking at are taken from aboard the boat itself, by refugees.

Next post in this series: the image of the refugee camp.


*I’d had this article on my laptop for a while but not got round to reading beyond the first page or two—my friend David Farrier emailed it to me after he’d read my last post, reminding me that I need to go back to it.


Images of refugees, part 2: refugees on land

My last post was meant to be free-standing, a quick riff off something I’d been teaching that day. But it sparked a couple of conversations on Twitter that have prompted a few more thoughts.

Gérard Noiriel, Réfugiés et sans-papiers, paperback of the second edition

‘Refugees’ appear to be interchangeable, visually speaking, provided that you—the photographer, the picture editor—present them in a particular way. In my next post I’ll talk about one example that’s recently been doing (and re-doing) the rounds on the internet. But this post will focus on an example that’s been on my mind since I noticed it a while ago, when I bought a copy of this book.

Gérard Noiriel is one of France’s foremost historians of immigration, racism, and national identity. Réfugiés et sans-papiers, originally published under a different title in 1991, is an important study of how modern France has dealt—or failed to deal—with refugees and clandestine migrants, in law, politics, and other areas. I’m not taking issue with Noiriel, here, but with the design team at his publisher, Fayard. Have a closer look at the photo on the cover. More cheerful than many images of refugees, it nonetheless has much in common with the pictures in my last post: a road, a pair of refugees trudging down it, one of them carrying a heavy burden (which is, perhaps, metaphorical as much as literal in images like this).

Gérard Noiriel, Réfugiés et sans-papiers, cover, close-up
A closer look

The problem is that this photo has nothing to do with refugees in France. Taken near Haifa in 1949-50, it shows two recent arrivals in Israel: they had probably come from one of the Displaced Persons camps that continued to dot Europe for a decade after the end of the second world war. It’s by Robert Capa, who made three trips to Israel between 1948 and 1950—the first to cover the war of independence (or first Arab-Israeli war), the second and third concentrating on how the new state incorporated the large number of Holocaust survivors flowing into it. You can find it on the Magnum website.*

You see the problem. The image of ‘the’ refugee is so generic, such a standardized trope, that one set of refugees can stand for another, regardless of time and place—even when the specific picture is taken by one of the most renowned photographers of the twentieth century, and is famous enough for photography websites to include it in sets of images intended to inspire or instruct would-be emulators. This picture is all over the internet; when I searched Google Images for ‘Robert Capa refugees haifa’ one of the first (of many) results was a Pinterest board that lists this photo as showing refugees during the Spanish civil war.

Kosovar refugees
Could be anywhere

In my last post I only discussed the appearance of refugees themselves in these generic images: trudging masses, rarely distinguishable as human figures. But on further reflection I realize that it’s worth thinking about the landscape they’re set in, too. This is another point of difference between twentieth-century news photographs of refugees and the nineteenth-century painting I included there, despite their clear family resemblance. The size of Gruzinsky’s painting not only allows individual figures to emerge clearly: it also permits the painter to include a great swathe of mountain landscape behind them. There’s a balance, in other words, between the panorama and the close-up, which results partly from scale and partly from the (slow) speed of composition. Fine art photographs might manage this, but in news photography it’s unlikely.**

Mexican refugees going to Marfa, 1910, Library of Congress
Could be anywhere

News photographs were and are taken to be reproduced at a relatively modest size: even a double-page spread in a 1950s issue of Life would make a pretty small poster. The equipment used to take them reflected this intended result. For most of the twentieth century, this would be a relatively small-format film camera, with a lens whose angle of vision may have been a bit wider than normal—that is, non-peripheral—human vision, but not a lot: the classic lens for Magnum-style photoreportage had a focal length of 35mm.✝ That gives an angle of vision a little wider than humans’ non-peripheral vision, which helps account for the sense of heightened realism that the Magnum house style conveys: you see a bit more of a scene in a photo taken through a 35mm lens than you would if you were just looking at it with your own eyes, but not so much that it’s obviously unrealistic. But this equipment, and this size of reproduction, place constraints on what one photo can achieve. For human figures in an image like this to be clearly individualized, like the woman and child in the Capa photo, the photographer has to be so close to them that the landscape disappears: in this picture we just get a bit of road and the edge of a field, with a nicely dark barn to contrast sharply with the woman’s sunlit dress.

5 Refugees leaving Libya
Could be anywhere

Look again at the photos of refugees on foot in my last post and you’ll see that in all of them, including the selection returned by a Google image search, the landscape is too sharply cropped for any significant features to be visible. Give or take a tree, East Prussia in 1945 could be Stalingrad in 1942 or Palestine in 1948—or, if that image were in black and white, DRC in 2008. Even when the photographer’s perspective is far enough removed from the group of refugees to reduce them to anonymous figures, too little of the landscape comes into view to be identifiable beyond the most basic distinction (arid desert; grassy hillside). The same goes for the images I’ve interspersed here, which are from Macedonia in 1991, Mexico in 1910, and Libya in 2015, and come from sources as varied as the UN website, the Bain photographic agency archive at the Library of Congress, and a Counterpunch article.

The balance of panoramic landscape and individual human figures is very hard to strike in photography if you don’t have a large-format camera and a lot of time to compose the picture. One of the images I found for my most recent lecture does actually come close, but it’s an exception to prove the rule:

Everything destroyed and burnt. Men only remained = Tout était déruit et incendié. Les hommes seuls restaientThe panoramic format here allows the photo to take in a sweeping view of the landscape, while the very deep depth-of-field means that one of the refugees (they’re—probably—Serbian refugees in Albania, in 1915) is close enough to the camera to be clearly distinguished as an individual figure while remaining in focus. But this is a little while before the invention and popularization of 35mm film cameras in the 1920s. News photography as it developed later in the century wouldn’t generally use this letterbox format: this picture—it’s in the Library of Congress—is mounted on an awkwardly long postcard, and would fit even more awkwardly in a magazine or newspaper format. And the one figure who stands out from the huddled mass remains huddled and anonymous: hooded, and too dark against the pale background for any individual detail to appear in his clothes.

This is what refugees are reduced to by the highly standardized visual tropes that are used to depict them: essentially interchangeable people trudging with their possessions across what are—in photos, at least—essentially interchangeable landscapes. When it becomes so generic, the image of the refugee is doing part of the job of making refugees something less than fully human. Some of my students worked this out in class today, when they reflected on their own surprise at finding images of visibly prosperous Belgian refugees fleeing the German invasion in 1914: these stolid bourgeois in frock-coats or dresses looked like people, not ‘refugees’. Thinking through their own reaction, they understood for themselves what that says about how strongly stock images of refugees shape our understanding of what a ‘refugee’ is. (Racists on Twitter, arguing that Syrian refugees can’t be proper refugees if they’re carrying smartphones, share the surprise—but none of the critical self-reflection.)

It is possible, I think, to break out of this dehumanizing trope. Photographers don’t always stand well back from ‘refugees’ and visualize them as an amorphous mass in an unspecific landscape: they may work with individual refugees to document their experiences, close-up; they may situate them in specific rural or urban landscapes. (Some refugees are themselves photographers, too, professional or amateur—and in the age of the smartphone that can be a lot of people.) The problem is that when there’s only room for one image—on the newspaper homepage, on the cover of the book, on the NGO website—picture editors and publicity departments reach for a generic one that really obviously says ‘refugees’. That decision isn’t usually taken by a refugee, and for that matter neither is the photo. But it doesn’t have to be this way.

I’ll end this post with an example of a visual depiction of refugees that starts with a panoramic (indeed, cinematic) view of a landscape but then gradually shifts its focus through different scales, so to speak, to concentrate on individuals—who, by the end, are no longer refugees.

Poles in Persia, British Pathé newsreel
Poles in Persia

Poles in Persia is a British Pathé newsreel from 1943. I’d read about the group of Polish refugees who trekked overland from Poland to Persia during the second world war, but I only saw this newsreel when a friend tweeted me a link after reading my last post. It would be a mistake to view this as straightforward reportage: it’s very much a staged piece of film-making, produced as Allied war propaganda. But visually it is very interesting, and powerful.

It begins with distant figures approaching across a parched mountain landscape, Lawrence of Arabia-style—though more than fifteen years before that film was made. They come closer, still as a typical trudging mass, burdened with infants and baggage. But then the camera is in among them, and the sonorous voiceover introduces us to a family, the Kowalskis. (Whether they were a real family or not, I have no idea.) Arrived in Persia, which was effectively under Allied occupation in 1943, the refugees are settled, cared for, clothed and fed. But more than that, they stop being refugees: the Kowalskis join the Allied war effort—’The Poles know where their duty lies… they’re not people to hide behind the efforts of others’. Father and grown-up son and daughter all volunteer for military service; mother busies herself with agricultural labour around the refugee camp, and looks after the two younger children, who go back to school.

It’s propaganda, to be sure, and it’s striking that no Persians cast so much as a shadow in the film—the reintegration that matters here is into the Allied war effort, not into the more or less unwillingly occupied host society. Still, it’s a demonstration of how a set of visual tropes that had already been well established for decades could be first adopted and then transcended in a five-minute newsreel, to turn a group of refugees from a destitute mass into individual human beings with lives and a future.

Next post: refugees at sea.

Stacy Fahrenthold asked me to think about what scale
is doing in images like these and Michaël Neuman told me
about the Poles in Persia newsreel—thanks to both.

I scanned the Noiriel cover from my own copy (fair use, I think?);
for all other images, click for source.

*As you’ll see if you click through either of these cover images, this photo is not used on the book’s current edition—another quite generic ‘refugee’ photo is, instead. I wonder if the new one actually shows refugees in France, and whether the change of image resulted from any pushback from readers or author, or just from Magnum upping their rates or something.

**Compare Gregory Crewdson’s giant composite photographs: on a gallery wall, they probably would let you step out to take in the panorama of an entire street, or in to peer through one shop window. But these are produced by a team resembling a film crew, over several days, and printed in a format that’s more than one metre by two. The reproductions here—it was the second of these three photos that I particularly had in mind—are also reductions to almost miniature scale.

If anyone’s interested in the technical details of this I can discuss them in the comments…

Images of refugees

Gruzinsky, The highlanders leave the village; Пётр Николаевич Грузинский, Оставление горцами аула при приближении русских войск
Pyotr Nikolayevich Gruzinsky, Оставление горцами аула при приближении русских войск (The highlanders leave the village as Russian troops approach)

I came across this painting last week, when I was searching for images to illustrate a lecture on the late Ottoman refugee crises. It’s the first proper lecture in an honours module I’m teaching on refugees and statelessness in world history, c.1900–1951. That ‘c.’ allows a lot of wiggle room: in this lecture I briefly go back as far as the Russian annexation of the Crimea—the first time round, that is—in 1783. But most of the lecture treats the fifty years or so from the consolidation of Russian rule in the Caucasus in the 1860s to the Balkan wars of 1912 and 1913: a half-century when millions of Muslims left the Russian imperial borderlands, and the new Christian nation-states that had broken away from the Ottoman empire, and sought refuge in the empire’s truncated (but still extensive) territories. This painting is by Pyotr Nikolayevich Gruzinsky, a prince of the Georgian royal family, and therefore a member of the Russian imperial aristocracy, in the mid-nineteenth century. It surprised me somewhat for its sympathetic depiction of Muslim refugees being forced out of the Caucasus in the decades when Russia’s grip on the mountains was consolidated.

It’s striking how this painting prefigures the stereotypical image of forced migrants that appears in the print and then audiovisual media through the twentieth century and up to the present. I did a Google image search for ‘refugees’, and one of the suggested subcategories that came up was ‘refugees fleeing’—here’s what that click looked like:

Google image search for refugees
Google Images suggestions for ‘refugees fleeing’

There are a number of similar, and similarly ‘stock’, images on the Wikipedia page for ‘refugee‘. Here are the ones that show people trekking overland, on foot or on a cart:

Refugees fleeing from Kibati refugee camp to Goma refugee camp, eastern Democratic Republic of Congo, 2008
Ostpreussischer Flüchtlingstreck 1945
East Prussian refugees in 1945
Russian refugees near Stalingrad, 1942
Russian refugees near Stalingrad, 1942
Palestinian refugees in 1948

Gruzinsky’s painting is a reminder that some such visual tropes have roots that long predate photoreportage and newsreels. It’s a nineteenth-century narrative painting, and if I was put on the spot and asked to trace its antecedents my first guesses would be artistic depictions, in the European tradition going back to the Renaissance at least, of the biblical exodus and the holy family’s flight into Egypt. There’s a contrast with most similar scenes in news photography, though, which is that the scale of the painting allows the figures to be depicted as individuals, clearly differentiated rather than trudging huddled masses. (The close-up image of refugees from Kibati, taken by a medical worker, is something of an exception: it was taken in a hurry, with the sound of gunfire not far away, and the person who took it was probably running too.) I wonder if Gruzinsky actually witnessed any of these scenes.

Click images for source
Apologies for any mad formatting, my laptop is playing up

Women, children, and baby dragons first

All roads lead back to King’s Landing.

I was pursuing a bit of secondary reading about humanitarian evacuations—well established as a practice now, but not in 1921, when the one I’m writing about took place. There are various examples from later in the twentieth century, especially concerning children: the children brought to France and Britain (and the Soviet Union) during the Spanish civil war, the Kindertransports, or somewhat later the contentious ‘Operation Babylift‘ that removed the children of US servicemen from Vietnam.

The term we use today, though, was slower to emerge. The rough measure of a Google ngram shows that ‘humanitarian evacuation’ is absent from the corpus of books in English before 1968. Rising from that point, there’s a low peak of frequency around 1980, but it’s in the late 1990s that the phrase really takes off, especially with the humanitarian evacuation of Kosovo Albanian refugees who’d fled to Macedonia in 1999. (The UN High Commissioner for Refugees at the time, Sadako Ogata, told the security council that this evacuation had ‘no precedent’ in UNHCR’s history.)

Couldn't get the chart to embed properly, but click through and it should work
Couldn’t get the chart to embed properly, but click through and it should work
Look for scholarly literature on humanitarian evacuation and that’s where most of it is focused, certainly the earliest substantial body of work on a particular case. A more recent example that’s caught some attention, involving an ad hoc collaboration between UNHCR and the International Office for Migration, was the evacuation of ‘third-country nationals’—mostly migrant workers—from Libya in early 2011.


I did come across an article, though, from 2003, that used the term to refer to an earlier period in the Balkan wars of the 1990s. Noticing that the author was a PhD student at the time that article came out, I googled her to see if she’d produced any further stuff on humanitarian evacuations. Now a professor at U Mass Amherst and a very prolific human security analyst, Charli Carpenter is also a total sf geek—author of a fine Foreign Affairs article on ‘Game of Thrones as theory‘ and producer of startlingly epic trailers for panels at the International Studies Association.

The one from last year is about killer robots.


When am I next going to a conference?

I can’t express how irritatingly capricious WordPress is being today
about frames, paragraph breaks, and smart quotes.
Believe me, I’ve tried my best.