Introduction

When looking at the history of visual humanitarianism, one surprisingly realizes that film history has only scarcely been covered, while scholarly interest has increased in humanitarian campaigns on digital media (Cottle and Cooper, 2015). Yet, debates that emerged in the 1980s about the paradigm of distant suffering, immersion and chronotopic engagement by means of communication technologies, such as virtual reality, remain to be examined through historical patterns. In the age of mass communication, aid agencies turned very early to motion pictures to produce an immersive spectacle, relying on the cinematic realism of non-fiction movies to increase the ‘perceptual experience’ and the ‘aesthetics of astonishment’ of the viewers (Crawford-Holland, 2018). Back in the 1920s, ‘cinema … “virtually” extended human perceptions to events and locations beyond their physical and temporal bounds’ (Uricchio, 1997: 119). Humanitarian cinema thus participated in transnational campaigns aiming to mobilize and sensitize national audiences. More specifically, these movies also advocated on behalf of distant beneficiaries, and shed light on ‘hidden’ realities through the mechanical image.

Yet, historians have only started to unravel the rich diversity of humanitarian films of the 1920s, even less from the perspective of visual advocacy. Therefore, the purpose of this paper is to question the regimes of visibility through which early humanitarian cinema operated and the parallels raised with modern advocacy strategies. While previous scholarship has focused mainly on humanitarian representations and visual strategies (Kurasawa, 2012), this paper opens new lines of inquiry throughout movies produced by the International Committee of the Red Cross (ICRC), private charities and state-funded agencies during humanitarian operations launched in Eastern Europe after World War I. More specifically, it examines the performativity of moving images in making public claims, forging and channeling specific sensitivities among ephemeral audiences who gathered to watch these films. The ‘technologies of witnessing’ (McLagan, 2006: 191) offered by cinema not only allowed audiences to delve into the testimonial function of such images, but also to question the ‘proposal of commitment’ (Boltanski, 1999: 149) that was made to the spectators to feel and act in a particular way.

Examining the performativity of images involves the relation between visual forms and non-visual forms, because a medium cannot be isolated from its circuits of dissemination and its contexts of exhibition. The transnational networks within which humanitarian cinema operated, the messages embodied in these films, the multimodal experience of watching them in a designated space, but also the new relation with the moving image at the beginning of the twentieth century, all participated in forming injunctions to care about and react to the injustice exposed. Drawing from recent scholarship building on emotions and humanitarianism, this paper thus considers early humanitarian films as a form of ‘mediated humanitarian affect’; by the 1920s, this media technology offered a new ‘scale of mediated communication, sensorial range of human experience, and capaciousness of moral attention’ (Ross, 2020: 169). The movies not only proposed ‘inducements to affective expression’ (175) but were the key component of humanitarian campaigns that all mixed fundraising, awareness and education. This period was indeed a laboratory for aid agencies to develop and adapt their communication practices, with blurred lines between publicity and propaganda, promotion, identity, and reputation.

The paper first examines the creation of humanitarian films in the 1920s that resulted from competing communication strategies among organizations. It then reflects on the use of humanitarian cinema, both as a mean to advertise, as well as to make public claims. The paper continues by exploring the testimonial function of the films in humanitarian publications and promotional material and discusses the idea that ‘seeing is believing’. Following on that, the link between visual evidence and affects is addressed, as humanitarian cinema allowed contact with suffering that was more intimate. Finally, the immediacy of the cinema technology and its induced immersive spectacle is analysed, to question the perceptual experience of the films’ settings with the production of eyewitness images and first-hand accounts during the screenings. The paper concludes by highlighting the visual politics at work in these movies and their paradoxical use, as they combine the needs-based and rights-based paradigms in the humanitarian response.

Humanitarian Cinema at the End of World War I

Interest in humanitarian cinema of the interwar period has grown since the historian Roland Cosandey discovered the whole collection on the Russian famine in the Swiss film library in 1998 (Cosandey, 1998). Scholarship in humanitarian cinema has emerged in the last decade, after aid agencies sought to preserve and promote their film archives (Junod, 2005) that were completely ignored in film history. In addition, some historians active in film preservation realized the extreme frailness of the material and the versatility of archival policies; many remaining film collections are rather fragmented, with many original films lost to history or cut, because it was common to edit and modify original footage in the 1920s. With new focus on past and present media innovations by aid agencies, historical investigations have brought to light the 1919–23 momentum of humanitarian cinema.

An intense production of silent movies followed relief operations for starving populations, refugees and genocide survivors in Central and Eastern European countries. The defeat of Germany and the partitioning of multinational empires led to the creation of new states, thus sending millions of displaced persons on the road, which – together with the war – provoked unprecedented deprivations throughout Europe. The 1917 Bolshevik Revolution and the civil war in Russia also threatened Central Europe to fall under Soviet influence. The 1921–22 Russian famine thus triggered a large-scale international response that intertwined with anti-Russian sentiments in the West (Sasson, 2016). On the Eastern front, the emergence of the Turkish National Movement and territorial disputes in the Ottoman Empire resulted in the first genocide of the 20th century; humanitarian operations to assist Christian minorities mobilized Western countries that remained active in the region until after the 1919–22 Greco-Turkish war (Laycock and Piana, 2020). The Near East became a strategic issue for Western powers in the new era of multilateral diplomacy sparked by the creation of the League of Nations in 1920 (Cabanes, 2014).

Consequently, the interwar period was a moment when ‘internationally coordinated responses were replacing nationally driven action’ (Barnett, 2011: 83). In addition to the ICRC and national Red Cross societies already operational abroad (United States, France, and Sweden), the Save the Children Fund (SCF) was a private, voluntary action-based organization founded in 1919, the same year that the government sponsored American Relief Administration (ARA). The SCF was preceded by the Near East Relief (NER), founded in 1915 to help Assyrian and Armenian populations in the Ottoman Empire. Many worked in collaboration with faith-based charity groups, such as the British and American Quakers (Muckle, 1990). The coordinated response in Europe in the 1920s increased ‘the institutionalization and professionalization of humanitarian practice, including business-like fundraising, purchasing, and accounting procedures’ and was shaped through mediatized operations that included ‘the systematic use of photographs and motion pictures’ (Götz et al., 2020: 46).

The ICRC films have received substantial attention from historians since the institution digitized its audiovisual collection with the help of the Swiss association Memoriav (Natale, 2004). The ICRC created a Propaganda Commission in 1919, but realized in 1920 that it did not have any movies to show at the 10th International Conference of the Red Cross to be held in 1921 (Piana, 2015). This situation spurred the organization to commission sixteen films from its delegations in the field between 1920 and 1923, ranging from the fight against typhus in Poland, the exchange and repatriation of war prisoners in Germany, Estonia and Greece, the rescue of Russian refugees in Constantinople, the feeding operations and educational activities for children in Hungary, and other relief activities during war and peace. Unlike the ICRC, the American Red Cross had a tremendous production of a hundred films through its Bureau of Pictures (the first film unit to open in 1917 until 1922), recording Red Cross nurses attending to war wounded, the improvement of hygiene and health practices, as well as the assistance provided to civilians from France to the Balkans and North Africa (Horne, 2012; Veeder, 1990). The newly founded League of Red Cross had sixty films available in 1921, as well as Italian, Swedish, and English Red Cross societies. The ARA (1919–23) was also keen on using cinema to raise awareness, with films such as Starvation produced as early as 1919, to America’s Gift to Famine Stricken Russia released after the end of famine in 1923. The former followed American food relief operations in Central and Southern Europe after World War I amid persistent fighting, leading the film crew to witness war crimes in the Baltics. The latter provided visual accountability of the success of ARA’s food distribution programs, tracking the tons of American products (including corn, milk, and cocoa) distributed under the supervision of ARA’s staff to eleven million starving Russians. During his early work in Belgium, ARA’s president Herbert Hoover, had already commissioned the documentary film Not Bread Alone to his publicity chief George Barr Baker, a former magazine editor.¹

The SCF also embraced the new tools of propaganda very quickly. Similar to the ARA, the SCF hired a publicist in 1920 and other public relations experts contributed to campaigning. It also commissioned G. H. Mewes, a Danish filmmaker and former photographic correspondent for the Daily Mirror in Russia, to make the film Famine: A Glimpse of the Misery in the Province of Saratov.² Mewes was sent there in winter 1921 and recorded several sequences showing the extreme misery in Saratov, with abandoned shanty towns and the exodus on wagons pulled by camels, starving children in rags, distribution of food, and dead corpses in the Buzuluk cemetery. Although the most notorious SCF film, it was preceded by Soviet Famine Appeal earlier in 1921. The SCF replicated the strategy on the eastern frontier, with Salvage in Austria and Salvage in Hungary (1922), also from Mewes, but less graphic than the scenes in Saratov, whereas The Tragedy of the Near East (1923) focused on the million Ottoman Greeks who were forced to move to Greece after the burning of Smyrna by Turkish troops in September 1922. The British Quakers also recruited a professional filmmaker, the Australian naturalist George Hubert Wilkins, to film hunger in Russia in November 1922. The film was released in 1923 as New Worlds for the Old: Quaker Relief in Stricken Europe and highlighted the Quakers’ distribution of food, medicine and clothes in Central Europe and Russia, as well as the means of survival for the local population. The same year, the Friends of Soviet Russia (the American branch of the Workers International Relief) released Russia through the Shadows. The High Commissioner for Refugees, Fridtjof Nansen, appointed by the League of Nations and the ICRC to coordinate relief in cooperation with the Soviets during the Russian famine, also brought back some Soviet footage after a visit to the Volga region in November 1921.³

These movies demonstrate a growing competition to use visual media to promote relief operations in Russia, Armenia, or Greece. Although, nothing competed with the massive commercial success of Auction of Souls, a movie adapted from the book Ravished Armenia, written by a survivor of the genocide, Aurora Mardiganian, and released in 1919 by NER. The plot depicted the deportation of entire Armenian families and religious leaders in April 1915, killed along the forced marches or sold in the slave markets. Often presented as one of the first advocacy films, its vivid representations of atrocities made it an incredible success. Raising millions of dollars, it was a masterpiece in the NER campaign to help Christian minorities. Alice in Hungerland followed in 1921; replicating Lewis Carroll’s character, Alice sets off on her own adventure, traveling to the Near East to join her father, a humanitarian worker. However, the film was much less controversial than Auction of Souls because it focused more on humanity, highlighting the improvements of childcare in NER orphanages.

Between Publicity and Propaganda: Learning Visual Advocacy

Indeed, Werner’s article featured five reproductions of SCF’s illustrated advertisements, emphasising the collaboration of the press for the circulation of humanitarian appeals in the national arena.⁸ Additionally, humanitarian cinema helped increase the visibility of international campaigns. They showed a strong combination of publicity (i.e. ‘the dissemination of light on what remains in the shadow’) with propaganda – the propagation of a particular doctrine, practice, or ideology, such as the humanitarian movement and its principles. These movies sometimes led to substantial financial outcomes, with more than £60,000 raised for Famine in Russia after two years of exploitation, and £500,000 raised during the campaign that followed New Worlds for Old (Tusan, 2017: 223, 227). But rather than effective fundraising tools, these films more often seemed to fit ‘political motives’ (Palmieri, 2019: 94). They operated in a moral economy in which images served several purposes:

• Seeing to understand: This purpose mainly arises in educational and scientific cinema. One can see it employed in the numerous films about hygiene, public health and epidemics produced by the ICRC, the League of Red Cross, and the Red Cross societies.

• Seeing to recognize: This purpose is typical of institutional cinema, intended to promote the reputation and legitimacy of the organization, as seen in The ICRC and its Post-war Activities (1923). It also consists of self-gratifying accounts for donors who see the tangible results of their donations, such as in America’s Gift to Famine-stricken Russia. The movie included a scene in which the entire population of the Vaseliefka village was shown ‘on their knees in thanks as first food is brought by the ARA’.

• Seeing to act: This purpose is representative of mobilizing cinema, whose goal is to make claims, provide evidence, persuade and command. The majority of the movies produced in the Russian and Greco-Turkish contexts fit into this category, on which we will focus more specifically.

‘Seeing Is Believing’: Presenting Visual Evidence

Not all organizations engaged in the new technology of cinema with the same conviction. The SCF seemed to have been more enthusiastic, being a private charity founded by two activist sisters known for their criticism of government policy during the Great War (Mahood and Satzewich, 2009). Advocacy remained a core value once SCF was founded: ‘it has been our lot to champion children whose parents, country, government or religion happened to be unpopular and this has not made the work of getting funds any easier’ (Record of the Save the Children Fund [hereafter, Record], 1921a: 67). Operating in foreign countries, including some considered to have an ideology hostile to the West, humanitarian organizations strategically turned to simple narratives in their visual appeals to depoliticize the context. Typically, the child constituted a figure that was ‘ideologically neutral’ (Cosandey, 1998: 5). Similarly, the focus on relief operations and benevolent workers helped to shape moral responsibilities of Western viewers to act in solidarity ‘based on need and not on identity’ (Barnett, 2011: 84). However, humanitarian motives were not free of sectarian interests (religious or political), particularly among the American state-funded agencies that had an anti-Bolshevik agenda or an anti-Turkish, pro-Christian minorities incentive. Notwithstanding, raising awareness on the neutrality of aid through victimization of worthy beneficiaries became a policy embodied in the humanitarian filmic narrative.

With few exceptions, humanitarian movies advocated with persuasion (why should we help) and mobilization (how to help), but without denouncing the political roots or implications of the crises. However, some were more political or accusatory because of the controversies they raised. Whereas ICRC, Red Cross and Quakers movies used a humanitarian narrative highlighting ameliorative action (Laqueur, 1989), a few particularly focused on abuses or graphic details of the body in pain. NER’s Auction of Souls was the most political, denouncing the Turkish portrayed as the oppressors; many scenes of the movie visually re-enacted the atrocities described in the book, including mass burnings, rapes, impalings, and crucifixions. Famine in Russia had scenes showing piles of corpses lying in the frozen landscape at the Buzuluk cemetery, or close-ups of emaciated children’s cadavers. Although macabre, these depictions were used to brush aside any accusation of visual fallacy: ‘To suggest … that the thing might have been “produced” (i.e., acted for the film) and that even the coffins which were shown disgorging their ghastly burden into great pits each containing from seventy to a hundred dead might have been filled with “property” corpses, was puerile’ (Record, 1922b: 151).

Starvation, for its part, included controversial scenes showing a German firing squad shooting Bolshevik prisoners and hangings of Bolshevik spies. The execution was filmed in Latvia by one of the American crewmen on 26 May 1919, and remains ‘one of the most sensational film sequences ever shot’ because it is a direct account of a war crime (Patenaude, 2012). The film premiered in New York on 9 January 1920, and the brutalities shown prompted outrage in the audience. ARA’s chiefs also considered the scenes too sensitive because the film was used to support Hoover’s political prospects⁹: ‘[It’s] so filled up with Bolshevist outrages and atrocities of the Germans that we could not possibly stand for them’ (ARA’s Edgar Rickard, quoted in Patenaude, 2012). These examples thus attest to ethical dilemmas about what is representable and unrepresentable, especially when it comes to death, horrors, and executions.

Affective Performances

The analysis brings us to the question of the medium and the public, as well as what watching such films was like in the 1920s. In general, images function with more immediacy and affective persuasion than words do. The screen did not only offer a contemplative spectacle. Indeed, film-viewing was itself an immersive experience through the ‘magical immediacy’ of cinema technology (Horne, 2012: 15). Even more than today, it offered a privileged window to other parts of the world. And the publics who were attracted to the humanitarian screenings came in anticipation of seeing pain and care. Thus, animated pictures acted as performative devices by affecting the spectators and by asking the audience to do something about the suffering exposed. Moviegoers were put in ‘specific dispositions to feel, think, and act toward each instance of suffering’ (Chouliaraki, 2008: 372, emphasis in original). Similarly to photography being more than the photograph, cinema was more than the animated picture and to fully appreciate its performative process, one has to put it in a ‘larger complex of elements’ (Shusterman, 2012: 68).

Although it is impossible to know how the audience felt retrospectively, we can reflect on the injunctions to feel and care made in the films. The controversial films discussed above stand close to another visual genre – that of atrocity images. As with certain photographic cases used in transnational networks – who denounced abuses in the colonies (Twomey, 2012) – atrocity images mainly focus on the exhibition of death and inflicted violence, sometimes with a depiction of the oppressor. They are extreme and more commonly used in human rights investigations. However, atrocity images are not the same as about-to-die images (Zelizer, 2010). The latter were present in many humanitarian films and were instrumental in forging incentives to care and overcome indifference. Many scenes fostered an emergency impulse: an elderly person with all apparent physical signs of starvation (skeleton-like appearance and swelling joints), refugees in ragged clothes, ghost villages and ramshackle settlements. About-to-die images do not show the outcome (as atrocity images do) but rather freeze time and fate to suggest that it is not too late to intervene to prevent impending death. On the opposite of the atrocity scale stands the rewarding image of ameliorative action, which is the meaning-making component of humanitarian appeals because it shows that relief works.

The movies’ cardboards and advertisements systematically linked ameliorative actions to positive emotions. The films emphasized the outcomes of aid through visual editing that reinforced the before-and-after effect: the remedy (relief and care) and the return to the initial state (food, shelter, and education) bring about justice. ICRC movies on prisoners of war thus gave a feeling of ‘weariness’ when prisoners were shown in detention camps, shifting to ‘compassion’ once they returned home (Clouzot, 1921: 367–8). Similarly, the spectacle of ‘parentless, homeless, ragged, starving and ill’ children, sometimes ‘piteously begging’, is presented as ‘disconcerting’, ‘sorrowful’, and ‘wholly wrong’, whereas children shown dancing, singing and learning in orphanages and schools are ‘healthy’ and ‘happy’ (New Near East, 1921: 4). The visual economy of aid (and its underlying moral, political and religious components) thus involved a plethora of emotions not limited to care, compassion, and empathy; filmic narratives also involved indignation that enabled new affective and intellectual engagements of the audience with ‘others’.

The use of blame, which previous studies have underestimated, is of particular interest. Instead of shaming oppressors, the movies blamed the audience for not caring enough. Guilt was thus implied, suggesting that people felt remorseful if they did not help, as underlined in this cardboard from Famine in Russia: ‘Do not forget the emaciated faces of these children, their thin little arms! Let this be to you like a nightmare until you have done your duty towards those who are starving.’ As in humanitarian advertisements, movies used prescriptive requests that commanded people to act – hence the imperative tense and exclamation marks. Phrases included verbs of supplication and affective or emphatic words (e.g. ‘massacre’, ‘disaster’, ‘tragedy’, ‘agony’, ‘dreadful’, and ‘wild terror’). The discourse stressed the unequal (moral) relation between the miserable beneficiary and the potent benefactor through personification (‘you’). This follows the Victorian melodrama style familiar to human rights activists (Newey, 2000) that is still present in today’s humanitarian communication (Wells, 2013).

The feelings induced were instrumental in turning distant suffering into a morally worthy issue. This way, the ‘truth status of images always depends on critical contextualization’ (McLagan, 2006: 192), in which distant suffering is given meaning through specific frameworks. In the early twentieth century, humanitarian action was framed within imperialist incentives (Baughan, 2013), and the body in pain was seen as pornographic (Halttunen, 1995). However, the language of international gospel also prevailed (Torchin, 2006), through familiar tropes of Christian martyrdom and biblical iconography. Together with mother-and-child drawings in the illustrated advertisements and appeals, many of these films called out to a broader Christian community in the West: ‘What has happened to Christendom that it can allow this appalling horror in its midst and lift no more than our little finger to save?’ (Reverend Orchard, 1922: 181). Overall, the performative process of humanitarian films enabled the communication of emotions (audiences were expected to express feelings during the screenings) and acting on these emotions. The magazines of aid organizations regularly emphasized public reactions after viewing the movies, such as donations, spontaneous organization of local fundraising events, or letters to the editor.

Immersive Realities and Witnessing Audiences

The stories in these movies were framed for international audiences and circulated through charity and religious networks, from the field to the theatrical venues back home. The circuit started with the image-makers: relief workers or professional filmmakers, who also became observers (but not impartial observers), shot many of these movies on the spot. Very often, they were the first Western bystanders (besides journalists) to assess the needs, thus offering a unique glimpse on the other side of the mirror. The circuit then ended with local charities and national committees (e.g. the International Socialist Congress, the miners, the International Suffrage Alliance, the International Women’s Committee) contacting film departments of aid agencies to offer theatrical and non-theatrical venues, including meeting halls, clubs, schools, or churches. Moviegoers ranged from working- to middle-class, with a prevalence of female adults. Cinema also became a mobile technology after World War I, traveling from city centers to remote and rural locations. Mixing different filmic genres such as ‘social education’ or ‘scientific publicity’ (Natale, 2004), humanitarian cinema targeted audiences already familiar with aid organizations, as well as those attracted to travel, encounters with the exotic, science and hygiene, war, or entertainment and adventures. More niches were also explored, such as the youth, as seen with the child characters in the NER movies.

The humanitarian film thus ‘provided suitable viewing for mixed audiences. … Part live-action theatre and part film-centred drama as they were, attending a humanitarian film was anything but silent’ (Tusan, 2017: 206). From the field to the venues, humanitarian workers invited to lecture audiences before or after screenings provided first-hand accounts of the hardships or recounted their eyewitness experiences. They acted as witness, speaking ‘on behalf of a local population’ and in their ‘own voice about that population’ (DuBois, 2008: 13; emphasis in original). Although beneficiaries remained anonymous victims whose sufferings the films exposed, the testimonies of humanitarian speakers and the visual presence on screen of caregivers who embodied solidarity and action gave meaning to their suffering. These movies coexisted with field reports or travel diaries of humanitarian delegates who documented the horrors in the first person that the charities’ magazines then published on a massive scale.¹⁰ Humanitarian accounts by delegates provided additional data to contested realities, because even the authenticity of news pictures was disputed (Bulletin de l’UISE, 1922). For example, Famine in Russia was distributed with hundreds of still images from the film, used as propaganda tools to freeze-frame the most memorable parts of the film (including the Buzuluk cemetery). Nansen used slides during his lecture tour in 1922, and printshots were distributed to the press while thousands of postcards and murals were shared throughout Europe. The forensic authenticity Famine in Russia provided, as well as the first-hand, eyewitness account of the High Commissioner, formed evidence-based advocacy.

Moviegoers had to pay to attend a screening or were sometimes asked for donations. In the end, cinema participated in larger campaigning strategies involving a broad mediascape, from pictures, brochures, flyers, posters, postcards, to ads in local newspapers and radios. Cinema added to the immersive experience in such a way, based in the presumption of transparency of images, or the capacity of the motion picture to give intimate access to distant people or cultures. These movies were screened in an increasingly visual world. The public was called to gather for a specific purpose, and then exposed to unprecedented images of suffering that were forcibly ‘put under their eyes’ (Clouzot, 1921: 367). They were then transformed into ‘witnessing publics’ (McLagan, 2006: 194), compelled to judge actions and give their ‘verdict’ as though they were a ‘jury’ (Record, 1922b: 152), with the risk that sensitivities overwhelm genuine witnessing.

Conclusion

Humanitarian cinema in the 1920s shaped visual communication by linking pathos to publicity and propaganda and forged a path to what are now important sectors of humanitarian communication: public relations, promotion and education, fundraising, branding, advocacy and campaigning. The filmic narratives reinforced visual strategies also used in still images. Hence, these films exacerbated the emergency rhetoric (about-to-die images), ameliorative action (before-and-after images) and decontextualization (lack of political explanation of the root causes of famine or genocide) still very familiar in today’s emergency appeals. The regimes of visibility to which these movies relate allowed witnessing audiences to make meaning of other’s sufferings through a multimodal experience. The performativity of the moving image, linked to visual testimony, transformed the immersive spectacle into forensic evidence. These innovative filmic strategies, together with the non-visual forms, helped to mediate humanitarian actions into positive and negative affect.

Thus, movies referred to iconic and familiar tropes – children, refugees, prisoners of war, caregivers, Madonnas and Pietas – that make suffering palatable but establish a moral hierarchy still visible in the twenty-first century. The films fetishized and reified victims into deserving others – those who deserve our attention and those who do not. At the same time, they reinforced the figure of the White savior, incorporating the humanitarian incentive in larger colonialist, imperialist, ideological and cultural frameworks. These movies fit into the needs-based paradigm that prevailed in the humanitarian system until the shift to a more rights-based approach in the 1970s. Yet, these films also highlight contradictions. Although humanitarian organizations commonly idealized humanitarianism as neutral and apolitical in the 1920s, they used cinema as a ‘speaking out’ strategy to raise awareness of the extent of the crises, but not to call attention to the nature of the problems. Far different from other human rights networks that were seeking justice, humanitarian organizations did not use cinema to ‘confront perpetrators with the consequences of their actions’ or to ‘reduce abuses against civilians under attack’ (DuBois, 2008: 12). As they were heading towards protection activities for children or prisoners of war in the 1920s, humanitarian organizations used the films to emphasize their responsibilities to hold themselves (and the donors) accountable. Advocacy was, first and foremost, a public form of bearing witness to.

At the publicity level, it seems that humanitarian films never matched the benefits of commercial movies, despite some fundraising successes. The production was not self-supporting and involved both self-financing and public sponsors (Piana, 2015; Veeder, 1990). Nevertheless, more historical investigation is needed for a proper ethnography of humanitarian cinema, to determine its ‘close connection to community-based organisations’ and ‘the full range of locations and spaces’ related to the circuits of dissemination (Horne, 2012: 14), as well as operating expenses, costs of film crews, and income. Cinema was abandoned in the middle of the 1920s, only to return after World War II. Audiences seem to have fallen in 1922–23 already, despite aid agencies constantly calling for attention in their bulletins. This trivial effect is not surprising for film historians, who notice that ‘particularly in the area of nonfiction film, growing audience familiarity with the unique or exceptional was often claimed to have a counter-effect – boredom’ (Uricchio, 1997: 120). This might be an early example of what has come to be known as ‘compassion fatigue’ in the 1980s.