During the spring and summer of 1945, the American essayist Edmund Wilson travelled to Greece as a ‘reporter at large’ on behalf of the New Yorker magazine. He had the ambition of tracing the origins of Western civilisation, as Classical Greece was being embedded into a broader imaginary that tended to frame the emerging Cold War divisions in civilisational terms.Footnote 1 Throughout his journey, Wilson was accompanied by personnel from the United Nations Relief and Rehabilitation Administration (UNRRA), one of the key actors in Europe’s reconstruction. He quickly became disillusioned with Greek society and politics and thought the Anglo-American involvement, encapsulated by the UN agency, had altered the native population for the worse. In his writings, Wilson also probed into neutrality, a long-lasting principle of humanitarian reason, and wondered ‘whether it is possible, in the world today, for a genuinely non-political organization to obtain from national governments any very large amount of money for disinterested international aid’.Footnote 2
The question of the inherent politicisation of aid was central to Greece, which, on the eve of the Cold War, became a focal point for multiple relief operations. The United States in particular was a key actor, especially with the 1948 enactment of the European Recovery Programme (ERP), also known as the Marshall Plan. The ERP embodied the wider US visions around liberal humanism, the ‘free world’ and Western civilisation.Footnote 3 From 1948 to 1952, the Greek state was one of the major recipients of American foreign aid. A pamphlet, circulated to US officials, noted that 85 per cent of the Marshall Plan was spent on providing ‘minimum necessities to sustain the population’. Half of these expenditures was spent on refugee care and ‘rehabilitation’.Footnote 4 Despite its long history, the notion of ‘rehabilitation’ became a conceptual cornerstone of UNRRA operations.Footnote 5 From spring 1944 to late 1946, UNRRA shipped to Greece commodities with a value of approximately 347 million dollars.Footnote 6 ‘Rehabilitation’ and ‘reconstruction’ were the main mantras of both programmes, which had been funded primarily by the American government. However, the relevant literature often distinguishes UNRRA-led relief efforts from ERP programmes, asserting that the former was multilateral and emphasised ‘emergency’ relief, whereas the latter was unilateral and centred on development and reconstruction.Footnote 7 Johannes Paulmann challenges this distinction, arguing that an accurate explanation of humanitarianism must consider historical conjunctures and contingencies where practices of development and relief converged and overlapped.Footnote 8 Drawing on Paulmann’s insights, this article positions the Greek civil war as a meeting point where politics, development projects and relief programmes both intersected and diverged. To address these multifaceted dynamics, I deploy the concept of the ‘weaponisation’ of aid.
As an analytical concept, weaponisation extends beyond existing frameworks that examine the political manipulation of humanitarian programmes. Overviewing the scholarly discourse on humanitarian independence and neutrality, Antonio Donini argues that there was no ‘golden age’, when humanitarian values predominated over political concerns.Footnote 9 Rather, the history of humanitarianism is full of examples of aid being instrumentalised. Donini distinguishes instrumentalisation from manipulation and politicisation, both of which are in many cases structural elements of humanitarianism.Footnote 10 For Donini, instrumentalisation is the utilisation of humanitarian action or rhetoric as a means to achieve political, military, economic and other goals.Footnote 11 Building on this account, this article coins the term ‘weaponisation’ of humanitarian aid to elaborate on a particular type of instrumentalisation. On the one hand, the concept of ‘weaponisation’ connects with the emerging literature on economic sanctions and the use of classification practices established by the League of Nations as weapons before and during the Second World War.Footnote 12 Weaponisation connects more clearly with the wartime conditions and demonstrates the multifaceted instrumentalisation of aid. On the other hand, weaponisation links to the founding resolutions of UNRRA, according to which ‘at no time [should] relief and rehabilitation supplies be used as a political weapon’.Footnote 13
This article argues that discriminatory provision of aid during the Greek Civil War extended UNRRA operations and followed more complicated trajectories than often assumed.Footnote 14 Aid was a way not only to reward anti-communist and loyalist groups but also, within the context of counterinsurgency campaigns, to establish a categorisation of deserving and underserving refugees.Footnote 15 Ultimately, the contending parties weaponised aid to serve their strategic economic objectives. Drawing on localised approaches,Footnote 16 I ask two related questions: first, how did humanitarian practices become enmeshed in conflict and shape forced displacement? And second, how did the relief objectives of UNRRA, the Near East Foundation (NEF) and the United Nations International Children’s Emergency Fund (UNICEF) clash with the promotion by local authorities of a form of laissez-faire capitalist reconstruction? I adopt a micro-historical and biographical method,Footnote 17 which builds on two bodies of scholarship: the history of humanitarianism and refugee history. Over the last few decades, a thriving historiography on humanitarianism has explored multiple topics, such as international interventions throughout the 19th century,Footnote 18 the links between imperial governance and welfare programmes,Footnote 19 relief and rehabilitation projects during and after the First World WarFootnote 20 and similar humanitarian operations in the aftermath of the Second World War.Footnote 21 Histories of displacement have also prompted closer scrutiny of humanitarian interventions on behalf of refugees, but internal displacement remains largely in the margins of these accounts.Footnote 22 More recently, a number of scholars have emphasised the involvement of relief agencies during the anti-colonial wars.Footnote 23 By expanding on these accounts, this article provides critical insights into the Greek civil war, which has been outside of the scope of these strands of scholarship.
The extent of weaponisation of aid is explored through the experiences of two humanitarian workers: British military officer Ewan John Christian Hare (1903–97) and US social worker Charles (‘Charley’) Schermerhorn (1902–86). Hare initially served in the military before working with UNRRA from 1945 to July 1947. In August 1947, he remained in Greece and worked under the American Mission for Aid to Greece (AMAG) and its successor organisation, the Economic Cooperation Administration/Greece (ECA/G). Schermerhorn had studied religion and social work; he began his humanitarian work during the Second World War with the American Red Cross in Georgia before being assigned to Greece under UNRRA. Later, he spent a few years working with the NEF and UNICEF. Schermerhorn worked extensively in western Macedonia, while Hare stayed mostly in central Greece (specifically Lamia and Karpenissi). Their personal records, alongside papers of Greek welfare specialists, offer first-hand experiences and a different perspective on the history of humanitarianism in Greece.Footnote 24
A close reading of these sources, produced by Hare and Schermerhorn, highlights three key aspects of humanitarianism in Greece, which challenge existing periodisations of humanitarianism and institution-centred narratives. First, they illustrate the dilemmas that humanitarian workers faced on the ground. These records, and especially the correspondence, delve into cases of discrimination, which were often omitted in official reports. Concerning the increasing refugee influx, Hare and Schermerhorn’s reports did not persuade the higher echelons of their organisations to adopt a coherent policy to tackle the root cause of displacement. Second, these two individuals remained in Greece throughout the period of the civil war and their practices did not differ significantly from 1945 to 1949. Therefore, the clear-cut distinction between ‘relief aid’ and ‘development aid’ does not hold in this context. Third, both were foreign citizens and worked in the highly politicised Greek context. Although they can be seen as ‘outsiders’ or simple bystanders, they were actors in the conflict, who, probably unconsciously, shaped the dynamics of the conflict. These individuals demonstrate complicated trajectories in the ways that international organisations operated at a local level.
I have divided this article into three sections. The first offers a historical overview of the forced displacement of civilians during the late phases of the tripartite Axis Occupation of Greece. I also examine the involvement of the UNRRA Greece Mission in the aftermath of the Greek liberation, demonstrating how these operations failed to resolve pre-existing refugee questions. The second section builds upon this foundation by examining the late phases of the UNRRA Greece Mission and introduces the distinct category of ‘bandit–stricken’ refugees. It analyses the main state-led responses to this moment of displacement. The final section demonstrates how these developments culminated in the weaponisation of humanitarian practices. It details specific instances when aid for refugees was withheld or selectively distributed to those deemed deserving recipients. Overall, my analysis reveals the ‘weaponisation’ of humanitarian assistance as it became increasingly entangled with political objectives during this period.
The Legacies of the Occupation
In 1940, Greece became embroiled in the Second World War following an invasion by Fascist Italy, which targeted Greek positions in Epirus. Despite defeating the Italian forces, Greece faced further invasion in April 1941 when Nazi Germany launched an attack from Bulgaria, swiftly occupying Athens within a week. A pro-Nazi puppet government, the Hellenic State, was then formed in Athens. King George II, along with some politicians, sought refuge first in Crete, then Egypt and finally London. They formed a ‘government-in-exile’ and signed a military memorandum of collaboration with the Allies. The country was then partitioned into three zones of occupation: Bulgarian, German and Italian. The Bulgarian zone spanned from Thrace to eastern Macedonia. The German zone took over the areas of Athens and Piraeus, western Macedonia, western Crete, some northern Aegean islands and eastern Thrace. The Italian zone, which covered the remaining mainland Greece, subsequently fell under German control following Italy’s capitulation in 1943.Footnote 25
Throughout the period of Axis Occupation (1941–4), two major ‘emergencies’ caused the involvement of voluntary organisations. Firstly, the combination of German policies and the British blockade of the Mediterranean led to a devastating famine during the winter of 1941–2. This famine, which persisted until at least 1944, resulted in the deaths of approximately 5 per cent of the Greek population.Footnote 26 A successful appeal to the American and British public, as well as funding from Greek expatriates in the United States, culminated in the provision of limited relief to famine-stricken Greeks by foreign and Greek relief agencies.Footnote 27 Many of these humanitarian agencies had been active in Greece for years. In the aftermath of Greece’s defeat in the Greco-Turkish War (1919–22), the subsequent refugee ‘crisis’ and the 1923 Treaty of Lausanne, these agencies facilitated rehabilitation of Anatolian refugees. Organisations such as the NEF, the SCF, the Young Men’s Christian Association and the Young Women’s Christian Association maintained their operations, and their personnel forged robust relationships with local politicians and the Athenian elites.Footnote 28
The second ‘emergency’ was the escalating exodus of the population, especially from the Greek countryside. Between 1941 and 1943, more than 70,000 people sought refuge in the Middle East.Footnote 29 They fled Greece driven by the occupiers’ violent actions and the inability to secure the means for survival.Footnote 30 Concurrently, the Jewish populations of Greece, primarily from Thessaloniki, Volos, Veroia and Athens, were deported to central Europe, where the vast majority faced extermination in concentration camps.Footnote 31 In 1941, the Bulgarian government annexed the area of Western Thrace – an area that was contested throughout the first half of the 20th century – and caused the displacement of more than 150,000 Greeks, including those of Eastern Orthodox creed and Muslim-Greeks.Footnote 32 Wartime displacement was not limited to international border crossings.
In the aftermath of the 1943 Italian surrender, the mopping-up operations of the German authorities caused another influx of internal displacement. From the summer of 1943, the two major resistance groups – the leftwing National Liberation Front (Ethnikó Apeleftherotikó Metopo; EAM) and its military branch, the National People’s Liberation Army (Ethnikós Laikós Apeleftherotikós Stratós; ELAS), and the republican/royalist National Republican Greek League (Ethnikós Dimokratikós Ellinikós Syndesmos; EDES) – divided the liberated areas of Greece into administrative structures.Footnote 33 As these groups based their power on alliances with the local village populations, the German authorities enforced mopping-up measures of counterinsurgency, such as mass executions of civilians, denial of relief provision to remote areas and looting and burning of whole villages, which had supported the resistance guerrillas.Footnote 34 Many Greek villagers went into hiding and others sought shelter in bigger towns. These groups of refugees, whose houses had been destroyed, became known as ‘fire-affected’ refugees (pyropatheis). The International Committee of the Red Cross (ICRC), which received many reports about the German reprisals, labelled them as ‘l’incendiés’. For example, in a report, submitted to the ICRC by a Greek representative, it is mentioned that the village Lechovo in western Macedonia was completely devastated in July 1943 because the German and Bulgarian authorities believed that the residents supported the EAM/ELAS. Almost all of the 2,000 residents were homeless and without basic supplies.Footnote 35 The ICRC supplies could not sustain them, and they did not have any fields that could be cultivated. The same report mentions that 25,000 from the town of Kozani and its nearby villages were refugees due to the mass reprisals. In late September 1943, 27,000 residents of the same area had become refugees.Footnote 36 In spring 1944, refugees from western Thrace and the Greek Macedonia region numbered more than 150,000. More than 50,000 refugees had relocated to Thessaloniki.Footnote 37
Until the October 1944 liberation of Greece, the Red Cross movement – mainly the ICRC and the Hellenic Red Cross (HRC) – worked on behalf of these ‘fire-stricken’ refugees by distributing basic relief to them and negotiating further aid distribution with the occupying forces.Footnote 38 However, the ICRC had to work under the auspices of the Hellenic State. In December 1943, the Hellenic State’s officials claimed that they had to obey the German military command’s order, which denied any relief to ‘fire-stricken’ refugees in western Macedonia.Footnote 39 The HRC also provided medical supplies, soap, corn and flour. In the area of Lamia, for example, the HRC estimated that those affected by the German reprisals were approximately 110,000 people in late 1943. Although HRC workers did not provide detailed statistics about the internal displacement, they noted that many mountainous villages had been deserted and most fire-affected villagers had arrived in the town of Lamia.Footnote 40
In the aftermath of liberation, the management of these dislocated individuals weighed heavily on UNRRA, of which Greece was one of the founding members.Footnote 41 In November 1943, the Greek government-in-exile was among the forty-four states that signed the founding agreement of UNRRA. In April 1944, the Greek, Yugoslavian and Albanian governments-in-exile signed an agreement with UNRRA Balkan Mission at Cairo. The ‘Cairo Agreement’ overviewed the scope of activities that UNRRA regional missions were to manage in the aftermath of Balkan liberation.Footnote 42 This agreement dictated that an Allied Military Liaison (ML) was to control relief for the ‘military period’. As soon as this period had ended, the UNRRA Greece Mission could consult the Greek government in matters of welfare. In accordance with their general policy, the UNRRA Greece Mission had to supervise the distribution of aid, which was to be undertaken by the Greek authorities themselves. The personnel of the Greek mission alone reached a peak of 3,137 Greek and foreign employees, before UNRRA withdrew from Greece in late 1946.
In December 1944, a full-blown civil war in Athens became the first significant challenge for UNRRA workers. The EAM/ELAS forces clashed with British imperial troops, a repatriated regiment from the Greek Armed Forces and former collaborators, who were released by the prisons to contain the ‘communist insurrection’ – the strife remained known as Dekemvrianá.Footnote 43 Direct British intervention swiftly quelled the revolt by mid-January 1945, culminating in the peace Treaty of Varkiza. The Dekemvrianá conflict caused another influx of dislocated individuals, when civilians from the ELAS-dominated areas fled to the British-dominated areas and vice versa. Athenians from the suburbs also abandoned their houses and sought refuge in the countryside.Footnote 44 The conflict stirred reactions within the UNRRA, as the chief of the mission, Laird Archer, resigned due to extensive meddling of British officials in relief provision during and immediately after the Dekemvrianá. Footnote 45 The American deputy chief, Buell Maben, became the new chief of the mission in May 1945 and remained until the withdrawal of the mission in July 1947. After the conflict was over and UNRRA personnel returned to Greece in March 1945, the Greek government and the UNRRA Greece Mission signed a bilateral agreement, according to which the Greek authorities had to provide aid, channelled by the UNRRA. The newly established Greek government had to deal with the repatriation of Greek displaced persons (DPs) from the Middle East and Central Europe, as well as the refugees, whose displacement was caused by the foreign occupation.
Contrary to the legal distinction between refugees and DPs, the UNRRA did not establish a specific status to assign to those that were internally displaced. The term ‘internally displaced persons’ (IDPs) was actually used by UNRRA personnel with reference to the Greek and Chinese internal displacements.Footnote 46 In August 1945, the Third Session of UNRRA Council in London passed a new resolution, revising the previous agreements on refugees and DPs, excluding the post-hostilities refugees from UNRRA’s mandate. The Balkan governments, especially the Greek one, reacted to the new regime, because wartime refugees had not been repatriated.Footnote 47 At the same time, Glen Leet, the director of the UNRRA Welfare Division in Greece, was pressuring the UNRRA director of the Displaced Persons Division, Harry White, to take responsibility for the internally displaced persons, who had still not returned to their homes.Footnote 48 Until the withdrawal of UNRRA from Greece, the issue of IDPs had not been solved, as many wartime refugees refused to return to their villages and opted to stay in the larger towns. For example, in his project closure report, an UNRRA worker from the Division of Displaced Persons noted that in Piraeus the question of IDPs had been handled mainly by the government. The government treated IDPs in the same way as DPs by temporarily settling those who wanted to return to their villages but were indigents in assembly centres and transit refugee camps.Footnote 49
The period between August 1945 and March 1946, when the first post-war elections took place, was crucial for the UNRRA projects, although a set of difficulties arose. Firstly, an economic measure of price regulation was proposed by Minister of Supplies Kyriakos Varvaressos. Varvaressos suggested the fix of prices of UNRRA relief supplies and demanded the shipment of more commodities. Within a few days, the prices of UNRRA foodstuffs were reduced by 50 per cent. However, the government was unable to impose such controls and ‘break the ring of profiteers’ who speculated from the UNRRA aid.Footnote 50 Secondly, the local committees around Greece profited by withholding clothing and other supplies and then reselling them in the black markets.Footnote 51 Thirdly, violent conditions in the countryside prevailed again, especially in the aftermath of the March 1946 elections, when right-wing gangs and militias attacked leftist and anti-monarchist civilians.Footnote 52 As many leftist civilians armed themselves and recruited guerilla groups, the warehouses of UNRRA aid became the main targets of raids. In Pollyrachon, close to Kozani, communist guerrillas raided the warehouses of the village and seized more than 50 per cent of the available food. In other areas around Kozani, the local authorities denied the distribution of aid and claimed that the UNRRA had to regain control of the situation. The UNRRA regional director asked a patrol of the gendarmerie to secure the area around the warehouse, and he organised aid distribution two hours before sunset. Therefore, the guerrilla groups could not set up a raid.Footnote 53 But right-wing gangs were also raiding warehouses, especially in the Peloponnese region. These gangs stole the whole supply of milk and no additional deliveries were expected.Footnote 54 Ultimately, in November 1946, Director General of UNRRA Fiorello La Guardia sent a furious letter to the Greek prime minister, Constantine Tsaldaris, stating that the governmental measures of delaying aid were deliberate. According to La Guardia, the Greek government wanted to wait for the end of UNRRA operations and then sell the supplies on the open market.Footnote 55
Amidst the increasing violence, Charles Schermerhorn arrived in Athens in late spring 1946. He quickly became disillusioned with the UNRRA mission there.Footnote 56 In the opening lines of his memoir, he identified two main flaws of UNRRA in general: first, that it was little more than a three-way arena of powerplay between the British Empire, the United States and the Soviet Union; and, second, that food was rotting in warehouses instead of being provided to ‘people who might oppose the British-supported royalists’.Footnote 57 Schermerhorn criticised the British workers of UNRRA, who seemed to extend their colonial mentality by questioning the Greeks’ ability to be self-governed. Throughout the eight months that Schermerhorn spent with UNRRA as a child welfare specialist and welfare field supervisor, he found himself in a royalist stronghold in southern Peloponnese. Along with the defeatist attitude of UNRRA workers, royalist partisanship created important obstacles to humanitarian work. In areas such as Sparta and Argolis, the welfare directors had developed close connections with royalist groups that often raided UNRRA warehouses, confiscating goods and distributing them to their loyalist networks. According to Schermerhorn’s retrospective account, few Greek workers attempted to distinguish themselves from the deeply politicised environment and the ‘royalist excesses’.Footnote 58
By late 1946, the termination of the UNRRA mission appeared imminent, but the logistical problems caused by the ongoing conflict had not decreased. One of the main challenges stemmed from adverse weather conditions, compounded by deficient infrastructure. This necessitated the formulation of winter rationing plans from late October, with distribution scheduled for early November, in anticipation of approximately 117,000 villagers from regions including Attica, Boeotia, Evvoia, Phthiotis and Evrytania being potentially isolated due to harsh winter conditions. In late spring of 1946, the Ministry of Supply Management opted to transfer the burden of grocers’ fees to consumers, requiring upfront payment for the months during which distribution would be impeded by winter.Footnote 59 Concurrently, any humanitarian distribution was heavily dependent on military operations, the weather and local networks of transportation. Villages that were within the range of rebel activities were excluded from relief provision.Footnote 60
Displacing and Labelling at the Height of the Civil War
In late December 1946, the communist guerrilla bands consolidated to form the Democratic Army of Greece (Demokratikós Stratós Elládas; DSE), directly under the jurisdiction of the Greek Communist Party (Kommounistikó Kómma Elládas, KKE). From early spring 1947 onwards, the KKE were inclined towards a military policy, gaining support from Yugoslavia and the rest of the Balkan states.Footnote 61 In September 1947, the ‘Lake Plan’ aimed to establish a regular army, leading to the formation of a provisional government of Greece at Pindos within two months. By mid-1948, with the United States sending an advisory military mission and expanding aid programmes, the general secretary of the KKE, Nikos Zachariadis, supported a shift towards trench warfare to replace guerrilla ‘hit-and-run’ tactics. Northern Greek villages became the main hotspots of the civil war and peasants became ensnared in the hostilities. Many villagers collaborated with the DSE by providing supplies or intelligence, or joining its ranks, while others were recruited by force.Footnote 62
To control this ground-up support of the insurgency, the government initiated a policy of forced removal. This strategy had already been implemented from 1945, when local authorities evacuated whole villages for ‘economic and military reasons’. This removal was associated with the rebuilding of villages closer to rivers in order to increase their productivity.Footnote 63 The extensive removal, however, started in late 1946, and by autumn 1947 the National Army (Ethnikós Stratós; ES) had evacuated a significant number of villages.Footnote 64 The four overarching objectives of this strategy were (a) to deprive guerrillas of reserves, supplies and intelligence, (b) to decrease the number of troops, which could secure villages, (c) to allow the bombardment of rebels’ positions without major collateral damage and (d) to transfer refugees to towns and ‘neutralise’ the suspicious individuals.Footnote 65 By categorising these populations as ‘bandit-stricken’ or ‘guerilla-stricken’ refugees, the Greek authorities simultaneously attributed blame to insurgents for the displacement, mobilised western anti-communist sentiment and invited relief agencies to act upon this ‘refugee question’.
The characterisation of insurgents as ‘bandits’ had propagandistic connotations. In the eyes of the Greek state officials, the DSE were simply gangs, whose status of belligerence was not recognised.Footnote 66 Furthermore, the figure of ‘bandit’ was tied to the Greek social and political imaginary during the 19th and 20th centuries. Some bandits became irredentist fighters, while others opposed the state administration and were looting villagers to survive.Footnote 67 By labelling the communist guerillas as ‘bandits’, the royalist government deprived them of any ideological motives. According to the official royalist narrative, Greek peasants were passive victims who had to flee to be saved from continuous looting and pillaging perpetrated by the DSE. The military reasons behind the expulsion were kept secret from the public.
Following the standard humanitarian practice of encampment, the royalist counterinsurgents confined the ‘bandit-stricken’ refugees to small camps.Footnote 68 These camps were called either ‘security centres’ or ‘refugee camps’. There, refugees had to obtain specific papers from the army to leave the area.Footnote 69 The material design of these centres varied. For example, at Kozani in northern Greece, many ‘bandit-stricken’ refugees were resettled into an empty field between some low hills. A barbed-wire fence and some military tents were erected by the army and a guard of paramilitaries supervised the area.Footnote 70 In Trikala, Victor Orestis, a Greek AMAG social worker, noted that ‘refugees live under, one could say, Indian living conditions’.Footnote 71 This was one of many references that Greek and American social workers made to the ‘backwards’ element of the refugee living experiences. According to a pamphlet, written in German by Swiss workers and published in Zurich, villagers from Sérvia – a village close to Kozani town – had been displaced initially in 1943 by the Italian occupiers, then in 1944 by the German anti-guerrilla units and once again in 1947. They claimed that the communist guerrillas had done more damage than ‘Germans and Italians put together’. The pamphlet concluded that the ‘constant uncertainty and the active state of war [aktive Kriegzustand]’ had rendered refugees ‘dull and hopeless’.Footnote 72
State strategies varied according to local conditions and insurgent support levels; consequently, civilians experienced displacement in markedly different ways depending on their location. One prominent leftist theoretician observed that his family in central Greece was ‘army-stricken’ rather than ‘guerrilla-stricken’: the ES forcibly relocated them to Karpenissi, making the state responsible for their care.Footnote 73 Working at the NEF hostel near Florina, Schermerhorn encountered a large influx of refugees who had fled from communists. Employing a standard triage approach to identify eligible children for admission to the hostel, the NEF personnel summoned refugee families to gather at a school. However, none arrived, prompting Schermerhorn to investigate the causes of this distrust. He realised that the families had been evacuated by the governmental army and feared that their children were to be taken away.Footnote 74 Yet many of the peasants became refugees due to their fear of retaliation or forced recruitment by the communist guerrillas. For example, in the villages of western Tzoumerka, a part of Epirus – where most of the residents did not support the communist guerrillas – children fled and hid in the forests and valleys to avoid any retaliation.Footnote 75 It is unclear, however, whether these people were included in the refugee statistics. Furthermore, the strategy of forced displacement targeted Ottoman Christian refugees, who had arrived in Greece during the 1920s. Regarding northern Greece, Schermerhorn asserted that the forced displacement, which occurred within villages established by Ottoman Christian refugees. These villages accommodated either Turkish-speaking groups, as exemplified by the Skafi village, Slavic-speaking minorities, or groups that were considered ‘leftist’ populations by the royalist army.Footnote 76 Hence, most of these operations targeted western Macedonia and central Greece, where most Anatolian refugees had been resettled from the mid-1920s.
Between 1947 and 1950, the royalist governments established three specific aid categories within the ‘bandit-stricken’ refugees.Footnote 77 In late 1947, the authorities assigned the ‘bandit-stricken’ status to those that could officially demonstrate either the destruction of their houses or the participation of family members in the auxiliary home guard, the Rural Security Units (Monades Asfaleias Ypaithrou; MAY).Footnote 78 Category A pertained to ‘indigent’ individuals, namely those who could receive supplies for free. By mid-1948, 238,000 refugees in Macedonia were considered completely ‘indigent’.Footnote 79 Category B included the remainder, who were receiving a nominal compensation, but they had to pay for their food and other supplies. The highest incidence of corruption was documented within Category A, which provided the most substantial benefits. Royalist governors, who generally chaired the committees responsible for refugee classification, frequently incorporated members of their partisan networks into these decision-making bodies.Footnote 80
Overall, quantifying the demographic composition of this extensive population displacement presents significant challenges. Any estimation relies on incomplete statistical records and the classification terminology applied to these citizens reflects wartime propaganda and archival omissions. Angeliki Laiou has estimated the refugee population at approximately 850,000 individuals.Footnote 81 Governmental pamphlets increased this number to 1,000,000 individuals.Footnote 82 The fragmented statistical evidence suggests that women, children (who constituted a distinct category) and elderly men comprised the majority of these ‘bandit-stricken’ refugees.Footnote 83 In 1949, when the war was over and the army returned many refugees to their villages, around 8–10 per cent of the country (including approximately 18 per cent of the rural population) had been ‘bandit-stricken’. With Greece’s total population estimated at 7.5 million individuals, and considering that Greek authorities imprisoned and internally exiled nearly 500,000 individuals, approximately one in seven Greek citizens experienced displacement during this period. This experience generated an increased demand for relief provision.
Weaponising Aid during the Late Phases of the Civil War
From early 1947, the DSE attacked warehouses as a response to the discriminatory allocation and retention of aid. The DSE had intensified its activities in Karpenissi and attacked warehouses where aid was stockpiled. By December 1946, the rebels had distributed the spoils to peasants, which were often stockpiled for months. Due to these actions the regional distribution committees decided to cease the delivery of aid.Footnote 84 Concurrently, governmental troops started to bomb warehouses and bridges to impede guerrilla movement. However, this bombardment exacerbated the villagers’ plight, while many of them remained without supplies for over a month. The insurgents took the few available supplies as a levy. Thus, many peasants left their villages and swarmed towards bigger cities.Footnote 85 The bombardment in Karpenissi, for instance, destroyed communication with Lamia. Bridges and trees were damaged and fell on the roads, which were also covered in land mines from both warring parties. A significant stock of food remained in Karpenissi’s centre and was not distributed, while in Phthiotis, 105 out of 174 villages had to depend on rations from October and November 1946.Footnote 86
As the civil war escalated, American aid and the displacement of civilians increased correspondingly. In one report, dated in early December 1947, Ewan Hare reckoned with the ‘refugee question’, which held sway over the Lamia region. The report concluded that ‘[t]he situation is unquestionably appalling, and it is considered that apart from humanitarian reasons, unless immediate and concrete assistance and better organisation are forthcoming, the political consequences may well outweigh any other assistance the Mission is giving to Greece’. Based on his sources, Hare approximated the refugee population to be around 30,000–35,000, out of which 18,000 were in Lamia. Additional to these refugees, there were 8,000 refugee children, aged between eight and twelve years old. Due to the large number of refugees, the ES had ceased the measure of evacuation and the royalist government had ordered an increase of daily compensation from 750 drachmae to 1,500 drachmae. Although most refugees had to be classified as category A refugees, the ES did not swiftly provide them with special documentation and most of them were not entitled to free rations. All the refugee camps that Hare visited in Lamia, Sperchias and Macracomi followed a similar pattern. They were in terrible sanitary condition, and most of them comprised tents, wattle huts and cabins that did not exceed 12 square metres in area and 1.83 metres in height. In most cases, two families, whose average number was seven people each, stayed in such small spaces. The swift exodus also deterred peasants from bringing their livestock. According to Hare, only 5 per cent of them brought their animals, which roamed freely around the camps.Footnote 87
In addition to their squalid living conditions, ‘bandit-stricken’ refugees encountered antagonism from local residents. Given that the vast majority of refugees originated from rural locales, they primarily engaged in agricultural and pastoral pursuits. But, due to the sluggish process of attaining refugee status and the absence of alternative income streams, those who succeeded in relocating their livestock to the refugee camps were compelled to sell their animals to local merchants from Lamia. These merchants regularly visited the camps and negotiated with desperate refugees, who were selling their livestock at very low prices. Moreover, as Hare noted, these animals were taken to the Athenian market for consumption. Consequently, upon return, the refugees would find themselves deprived of access to livestock. To circumvent the red tape and help the situation, Hare proposed that, among other provisions, the prefect himself be authorised to categorise refugees into Category A, that all refugees be entitled to a monthly ration free of charge and that refugees be engaged in collaborative efforts with the Ministry for Reconstruction to enhance the infrastructure of the camps, which could be sanitised by Hellenic Red Cross squads. Furthermore, the remaining UNRRA supplies stockpiled in Athenian warehouses could be allocated to refugees, while specific legislation could be enacted to regulate the resale prices of livestock.Footnote 88
However, by January 1948, most provisions were not implemented and many ‘bandit-stricken’ refugees from communist strongholds had not received aid. In Lamia alone, 37,050 refugees had not received food rations since October 1947.Footnote 89 And although the average compensation rose to 1,500–1,800 drachmae per day by mid-1948, Schermerhorn noted in a letter that refugees in western Macedonia had not received any money for at least three months. Subsequently, the government opted to reduce the compensation to 400 drachmae by November 1948.Footnote 90 A report highlighted the ‘refugee’s racket’, which was made by members of MAY, who were granted the status of ‘bandit-stricken refugee’. These members advocated for the local clientelist networks and pressured the authorities using contacts in the local press.Footnote 91 This advocacy and the increasing pressure of the ‘refugee question’ led to the de facto abolition of refugee categories in late February 1948. Some refugees were able to work in their agricultural fields during the day and return to the camps at night, but there was a growing mistrust of refugees among some local committee members. According to a memorandum sent by the prefect of Boeotia, the refugees were ‘disrupting the economic programme of the Government, and also spy and supply information to the andartes [guerillas] from their safe vantage point’.Footnote 92
The intervention of AMAG became a turning point in the practice of aid withholding with the case of powdered milk being illustrative of the complicated processes of distribution. From late 1947, Schermerhorn was working with UNICEF, which was supplying the Greek government with powdered milk and other foodstuffs. The local welfare committees, under the auspices of the Patriotic Institution for Social Welfare and Awareness (Patriotikó Idryma Koinonikis Prostasias kai Antilipseos; PIKPA), were given the mandate to distribute this milk. PIKPA was staffed with loyalist individuals, many of whom obtained their positions through clientelist networks. In many cases, as Schermerhorn noted, the UNICEF powdered milk was not distributed and was instead stockpiled in warehouses.Footnote 93 Refugees consumed the powdered milk when authorities distributed it as a prepared hot beverage; but, when provided in its dry form, they sold it at reduced prices to merchants supplying local bakeries and candy stores.Footnote 94
The problem of commodity withholding derived from the overlapping tensions between the different sources of supplies. UNICEF milk was intended to be given for free to refugee children under twelve years old. Meanwhile, AMAG milk was sold for relatively low prices to all children, including refugee children. If a portion of this milk remained, it was sold into the open market. Yet in most Greek villages, neither refugees nor other villagers were able to buy milk, which then remained in warehouses.Footnote 95 Deliberate manipulations also exacerbated the tensions between the sources of supplies. Throughout 1948, for example, many PIKPA branches were distributing only AMAG milk and withholding UNICEF milk in the warehouses. The main reason behind such a strategy was to manipulate the refugees’ demand for milk, even though they could not afford to buy it.Footnote 96 Thus, local actors manipulated the aid system by intentionally confusing ‘humanitarian’ and ‘development’ categories, playing these classification distinctions against each other for personal gain.
Gradually, AMAG realised that the financial repercussions of the forced removal outweighed its advantages and undermined any progress in the project of reconstruction. Schermerhorn noted that many peasants were prohibited from bringing their animals, while the crop was abandoned. The rebels were organising raids, reaping the remaining crops. The villagers were then beaten by the gendarmerie, as they were suspected of ‘collaboration’ with the communists.Footnote 97 Already from late 1947, the chief of the AMAG, Dwight Griswold, had requested the immediate return of refugees.Footnote 98 Yet, as noted above, the removals continued. In June 1948, the local authorities in Lamia announced that the relief to refugees had to cease and refugees had to return to their houses. An extensive plan for their relocation was drafted by Hare in late May. According to this draft, AMAG had suggested that a new organisation, under the leadership of a Greek general, had to be formed in order to mediate the return. Of particular interest was the destroyed infrastructure, which would block the distribution of aid. Supplies had to be transported on mules or by human carriers. Furthermore, the villages had to be guarded by semi-permanent squads.Footnote 99 Security was a prerequisite that could facilitate humanitarian logistics.
Other humanitarian actors could not compete with AMAG, which gradually sidelined other humanitarian initiatives. Schermerhorn, like many Americans, saw the arrival of AMAG as a relief for the refugees. Retrospectively, he mentioned how proud he was of AMAG’s achievements. In his memoir, he wrote:
One way to defeat communism was to demonstrate to those misguided souls who had embraced the faith that their best bet was to throw in their lot with Mother Greece. We had to give them hope that, with all of her sons working together as a united Greek family, a better life might come for all of them.Footnote 100
Although AMAG did not claim a purely ‘humanitarian’ role, the US intervention on refugee rehabilitation was framed in humanitarian terms. However, this type of humanitarianism was linked to the policy of containment of communism, outlined by the diplomat George F. Kennan in 1947. Schermerhorn connected containment with the rehabilitation of the nuclear family. This positioned the family as the fundamental unit of modern nationhood, portraying communist insurgents as destroyers of this social fabric through their systematic separation of children from parents.Footnote 101
The return of refugees ushered in a new narrative, one that highlighted the logistical ramifications of the conflict and underscored the necessity for US-oriented modernisation. By the end of 1948, the royalist government halted its provision of relief to refugees, who were deprived of sustaining supplies for a minimum of two months. The delay in distribution derived from the local refugee committees’ inability to comprehend an order, issued by the Ministry of Social Welfare, mandating a total reclassification of refugees. The committees struggled to discern the diverse origins of the refugees. Those originating from valley villages had easier access to transportation and towns, while those from mountainous regions had to store provisions, as their villages had been demolished.Footnote 102 The American policymakers read these logistical impediments within an ideological context, which could strengthen the support of the DSE. In essence, the remote populations and their access to aid were pivotal for political and social stability, as their potential collaboration with the DSE had to be dissuaded. As the royalist government allocated decreasing funds to refugee relief, ECA/G highlighted collaborations with ‘voluntary societies’ such as the HRC, the SCF and the NEF. These organisations had already established soup kitchens within the camps, and further collaboration could bolster the social prestige of the loyalist faction. Therefore, loyalist strata of the Greek authorities could easily weaponise aid as they were the main mediators between relief agencies and refugees.
In January 1949, the DSE attempted one of its last manoeuvres and occupied Karpenissi for around ten days, after which the royalist forces reoccupied the city. During the temporary occupation, the DSE confiscated most of the supplies and ‘abducted’ two HRC nurses. Apparently, the return of refugees to their villages, which surrounded Karpenissi, ceased because the security conditions were in doubt. However, as many villagers stayed in the town during the occupation, the HRC, ECA/G and the government sent Cooperative for American Remittances to Europe (CARE) parcels and supplies as an incentive to strengthen civilians’ morale.Footnote 103 Yet the roads were destroyed and these supplies had to be transported using mules, which were unavailable. The ES would build temporary roads so that refugees could occupy small areas at a time and gradually be returned.Footnote 104
More and more, the DSE had opted for total warfare, which included dispersing land mines on fields and roads.Footnote 105 These mines killed both military officers and civilians. Hare had witnessed such an incident when an AMAG truck was blown up by a mine; five civilians were killed and seventeen were injured.Footnote 106 In northern Greece, insurgents placed mines on the central roads between Florina and Thessaloniki, destroying civilian trucks. Even after the end of the conflict in October 1949, some NEF trucks were blown up by land mines and supplies intended for repatriated refugees were destroyed.Footnote 107 In July 1949, when most of the DSE troops were concentrated in northern Greece, the royalist government kept delaying the repatriation of refugees. Despite establishing security measures in the area, the civil authorities did not send money, food or tractors to facilitate the agricultural rehabilitation.Footnote 108 From the side of the royalist government, the frequent allusions to spatial obstacles became a significant justification, bolstered by the extensive utilisation of mines.
In August 1949, the remaining DSE troops were defeated at Grammos Mountain in northern Greece, and the royalist government announced the end of the insurgency. However, the end of displacement had not come yet. In a memorandum submitted to Hare, some refugees opted to remain in security centres because their futures seemed doomed; they did not have livestock and their houses were destroyed. They also requested money instead of building materials, which could not easily be transported.Footnote 109 Successive Greek governments would grapple with the ‘bandit-stricken’ refugee question for years to come.
Conclusion
From 1943 to 1949, the Greek countryside became a battleground of guerrilla warfare. Already by the 1944 liberation of Greece, more than 150,000 people had been internally displaced because of the German mopping-up operations. The epicentre of this influx was in western Macedonia and Epirus. The intervention of UNRRA did not solve this problem; on the contrary, the outbreak of further hostilities deterred many of these displaced people from returning to their villages. In the following years, approximately one in ten Greeks became displaced as part of state-sponsored relocation initiatives. By labelling these people as ‘bandit-stricken’ refugees, the Greek government weaponised humanitarian aid and used humanitarian rhetoric to battle the ‘communist threat’. Amalgamating pre-existing relief practices with visions for modernisation and economic development, the Greek Civil War became a test case for Cold War humanitarian interventions.
This weaponisation of aid created novel categories of socio-political groups deserving and undeserving of aid. ‘Bandit-stricken’ refugees, like the ‘fire-affected’ ones before them, had to work to receive aid. Furthermore, by claiming additional relief aid on their behalf, many local committees manipulated the relief provision and profiteered by selling the foreign supplies into the black market. After the arrival of the American missions (AMAG and ECA/G) in 1947, the American representatives on the ground insisted on the return of refugees. This return inevitably became tied up with American developmental priorities. This was not an anomaly. On the contrary, it exemplifies an emerging pattern where governments systematically weaponised relief provision within civil conflicts and counterinsurgency campaigns. Similar dynamics emerged in China almost simultaneously with the Greek civil war: by 1947, the Chinese National Relief and Rehabilitation Administration had distributed only two percent of aid to ‘red areas’ – those supporting communist forces – effectively denying assistance to guerillas’ supporters and internally displaced populations.Footnote 110
This weaponisation framework offers valuable insights into subsequent Cold War conflicts across Asia, Africa and the Middle East, where colonial governments repeated campaigns of forced displacement either through counterinsurgency in places like Malaya (1948–60) and Kenya (1952–60), or through refugee management in places like Algeria, Cyprus, Vietnam and Cambodia.Footnote 111 The Greek civil war thus represents not merely a case study but a formative moment in the evolution of modern humanitarian practice, where the strategic manipulation of aid established precedents that would be replicated throughout subsequent decades of global conflict. By identifying these continuities, historians gain deeper insight into how humanitarian interventions became entangled with political and military objectives.
Acknowledgements
I would like to thank the editors of the special issue, the editors of the journal and the reviewers, Professor Bertrand Taithe, Dr Antoine Burgard and Dr Laure Humbert, for their insightful comments. I would like to thank the Hellenic Research Fellowship Program, Tsakopoulos Hellenic Collection, California State University–Sacramento and its funding sources, the Elios Charitable Foundation, the Tsakopoulos Hellenic Foundation, and the Tarbell Family Foundation, which supported my archival research in Sacramento during the summer 2023.
Open access
