The Purposes of Starvation: Historical and Contemporary Uses

Abstract

Mass starvation has throughout history been mis-categorized as a natural phenomenon, or an unfortunate side-effect of conflict and political oppression. The numbers and names of the victims fade into the background, blurred traces of the horrors of history. Yet, this is both an inaccurate understanding of starvation crimes and an injustice to victims. Mass starvation is a process of deprivation that occurs when actors impede the capacity of targeted persons to access the means of sustaining life. In this article, we introduce key features of starvation and note that many of the acts that create conditions of mass starvation are already prohibited under different provisions of international law. We introduce the term ‘starvation crimes’ to capture how these separately criminalized acts, when perpetrated over a long duration can create mass starvation. Implicit in ‘starvation crimes’ is that starvation is produced by leaders’ decisions and serves political, military or economic goals. We discuss nine objectives that can be furthered through mass starvation, offering historical examples to illustrate each. They include: (i) extermination or genocide; (ii) control through weakening a population; (iii) gaining territorial control; (iv) flushing out a population; (v) punishment; (vi) material extraction or theft; (vii) extreme exploitation; (viii) war provisioning; and (ix) comprehensive societal transformation.

1. Introduction

Mass violence was once dismissed as ‘ancient ethnic hatreds’ or ‘tribal violence’. Today, it is widely accepted that widespread and systematic assaults against civilians result from policies implemented by leaders in pursuit of strategic goals. What is more, these policies have been deemed criminal by international tribunals, domestic courts and the court of public opinion.

It is time to shift our thinking about mass starvation in a similar way and with a similar result. While failures of food production or distribution are often central factors, mass starvation results from a broader range of acts that actively debilitate a population’s capacity to survive. Many of these acts are already prohibited under different provisions of international law. We introduce the term ‘starvation crimes’ to capture how these separately criminalized acts, when perpetrated over a duration of months or even years, can create mass starvation.

Implicit in ‘starvation crimes’ is that starvation is produced by leaders’ decisions and serves political, military or economic goals. In this essay, we discuss nine objectives that can be furthered through mass starvation, offering historical examples to illustrate each. They include: (i) extermination or genocide; (ii) control through weakening a population; (iii) gaining territorial control; (iv) flushing out a population; (v) punishment; (vi) material extraction or theft; (vii) extreme exploitation; (viii) war provisioning; and (ix) comprehensive societal transformation.

2. What is Starvation?

Starvation is both an outcome and a process.¹ As an outcome, starvation means deprivation of food unto death, and is very rare even during famines, where the proximate cause of death is usually infectious disease. It takes about two months for a formerly healthy human adult to die from food deprivation alone, during which time the body and mind slowly cease functioning. In this article, we refer to starvation in the transitive sense: a process of deprivation that occurs when actors impede the capacity of targeted persons to access the means of sustaining life. Our concern is with mass starvation: when groups of people (communities, nations, classes, populations) are deprived.

We highlight four key features of starvation.

First, many different actions qualify as acts of starvation. Depriving people of their ability to obtain food is the central component, which includes not only denying or destroying objects such as food stores, but also preventing or obstructing activities such as working, trading and foraging.² Other deprivations are also relevant, including: degrading public health, such as disrupting access to clean water, forcing people to congregate in unhealthy conditions and destroying health facilities; degrading habitation and shelter, and reducing the means and capacities for mothers to care for young children; and tearing apart the social fabric, compelling people to violate norms and taboos, and turn against one another.

Secondly, including a wider purview of acts is crucial to understanding the use of starvation as a weapon of war or repression. Inhibiting access to food is almost never perpetrated against a population in isolation from other actions: direct violence, sexual violence, displacement, destruction of livelihoods and infrastructure. This is particularly true in the contexts of armed conflict and widespread or systematic assaults that we examine in relation to starvation crimes. Understanding starvation exclusively in terms of a reduction in food misses the connective tissue linking a broader health and social context to elevated mortality.

Infectious disease is the leading immediate cause of death in famine. Distress migration, overcrowding and the breakdown of water and sanitation systems create conditions ripe for outbreaks of normally preventable diseases such as cholera, malaria and measles. Susceptibility is increased by undernutrition, exposure to extreme temperatures, poor caregiving for children and other stresses. As Helen Young and Susanne Jaspars have pointed out, the relationship between food insecurity, care and health is ‘synergistic’³: malnutrition soars when deprivations overlap. From this, it follows that acts that increase risk of death through degrading the ecology of health and caregiving should also count among starvation crimes.

The third feature is that acts of starvation intersect with other causes of deprivation. Some of these are external to the society (environmental stress, natural calamity, global economic shocks) and some are internal (economic inequalities and policy that cause economic distress, either deliberately or in error).

In a community that values and protects its weaker members, social ties of mutual assistance and coping will strengthen people’s resilience. During famine, the process of destitution degrades not only physical assets but also social bonds. However, if social ties are weak to begin with, or are sundered so that people turn against one another, then the trajectory of destitution and social fragmentation will be more rapid. If the society is already highly unequal, hierarchical or exploitative, then those features are likely to be exacerbated during a time of food stress. Although ecologies of nutrition, disease and mortality differ across contexts, these social factors are certainly important to the lethal synergies that lead to accelerated increases in death rates.

The tangled aetiology of increased mortality in famine complicates analysis of the purpose of starvation and has implications for establishing legal intent. Generalized food insecurity, underdevelopment, weak healthcare systems, environmental or climatic factors impacting farming or fishing, economic crisis or unintentional ramifications of armed conflict might be brought forward as factors that purportedly mitigate the direct responsibility of a perpetrator regime.

However, and this brings us to the fourth feature of starvation as a process, it takes a long time — months and often years — both to starve an individual and to reduce a population to famine. During this time, the morbid symptoms of societal distress and the degradation of wellbeing, become clear and known to perpetrators to the point of large-scale increased mortality. No modern famine has unfolded in silence. Even under the conditions of strictest censorship, as in Mao Zedong’s China, those in power were aware of the calamity. In contemporary times, humanitarian information systems provide detailed diagnostics of the dimensions and trajectory of food crisis and indicate what might be done to alleviate it. Across examples, it is the failure to respond to signs of mass distress that will prove critically important.

3. Categorizing Starvation

Mass starvation occurs as a result of multiple acts, which reduce a population’s access to food, water and the means to provide essential care for children among other degradations. These acts create the conditions of elevated mortality in a context rife with direct violence over a significant time period.

Existing international criminal and humanitarian law addresses many of the acts that create and maintain the processes of starvation described above. Various bodies of law address acts of destruction, deprivation and displacement as well as impeding delivery of humanitarian aid, and, depending on the precise law considered, require clarity on other issues, such as whether the acts are part of a widespread or systematic attack on civilians, or are committed as a method of warfare, and so forth, as discussed by legal scholars (including in contributions to this journal issue). The law addressing these harms is sometimes articulated as ‘starvation’; in other places, it references, for example, ‘objects indispensable for life’ or ‘conditions of life’, both of which implicitly reference the processes of mass starvation described above.

To focus attention on how existing law might apply to the conditions that produce mass starvation, we propose the term ‘starvation crimes.’ A portmanteau term, that draws on David Scheffer’s coinage of the term ‘atrocity crimes’,⁴ ‘starvation crimes’ does not refer to a legal category as such, but rather brings together a range of crimes under different provisions of law, in a manner that gives them political salience. The relevant legal prohibitions, many of which apply both in times of peace and conflict, may apply to situations that include deprivation of food, water and other objects and activities indispensable to survival, alongside a range of other acts that heighten the adverse effects of such deprivation, by compounding hunger and a degraded health environment with exposure, displacement, direct violence and lack of access to health systems. Together, these acts drive the processes of mass starvation.

Important legal distinctions adhere to different crimes, and prosecution would require in-depth legal analysis of the context, evidence and arguments that govern a given case. As illustration of complexities of the law, we note that starvation is already a prohibited act, named as such, under the laws of war. Attacks on ‘objects indispensable to the survival of the civilian population’ are prohibited in international humanitarian law (see Article 54 of Additional Protocol I to the Geneva Conventions, Article 14 of Additional Protocol II and relevant rules of customary law). Such behaviour is similarly criminalized, though limitedly to international armed conflict, in Article 8(2)(b)(xxv) of the Rome Statute of the International Criminal Court. Other war crimes may also be relevant to creating famine: wilful killing, torture or inhumane treatment, extensive destruction of property, intentionally attacking civilians or civilian objects, pillaging and so forth. Further, various acts of famine can, when directed against a civilian population in a widespread or systematic manner, constitute one of several crimes against humanity: murder, extermination, forced displacement, persecution, or other inhumane acts. If carried out with intent to destroy a protected group, then creating the conditions of life that would bring about a group’s destruction is prohibited under the Genocide Convention (Article 2(c)).⁵ Starvation is also a violation of the right to food.⁶ Lawyers will find ample room for debate about the context, acts and evidence required for the application of specific law.

Across legal categories, starvation crimes have three necessary elements: (i) an indication of intent; (ii) the acts of destroying the means for producing, processing and distributing food and water supplies alongside other objects necessary for life; displacing people away from resources necessary for life; or impeding humanitarian access to vulnerable populations; and (iii) an outcome which includes suffering and increased mortality, associated with deprivation of food, and/or other objects and activities necessary to sustain life (including health and shelter).

Evidence of a strategic or tactical use of starvation should not be limited to examples in which it can be demonstrated that perpetrators had mass starvation as their goal. Rather, we can take the key point from David Marcus’ legal analysis of famine that both intentional creation of famine and recklessness are prohibited under existing international criminal law.⁷ Recklessness becomes criminal when political superiors either know, or have reason to know, that mass starvation will be or has already become the outcome of their actions and then they ‘recklessly continue to pursue these policies despite learning that they are causing mass starvation’.⁸

For a starvation crime to be committed there is also a threshold of severity. Existing famine metrics are indicative, but not exhaustive, of the conditions that apply. For instance, the metrics used by the Integrated Food Security Phase Classification (IPC) scale focus on the severity of malnutrition, mortality and food crisis within a specific geographical locale.⁹ Significantly increased mortality and civilian suffering that might incur criminal responsibility should not be restricted to IPC level 5 (famine); widespread and protracted level 4 (emergency) conditions would certainly be sufficient. If a large population is in IPC level 4 for months or years, it may suffer excess mortality that passes the threshold of a ‘great famine’¹⁰ (100,000 or more dead). Thus, in South Sudan in 2014–2018, just two counties were in IPC level 5 for a few months, where several thousand people perished. But almost half the country was in IPC levels 3 and 4 for two or more years, and the best estimate is that there were 380,000 excess deaths,¹¹ half of them due to hunger and related causes, and half due to violence, over this period.¹² This illustrates how a ‘great famine’ could happen without ‘famine’ being formally declared.

4. Nine Uses of Starvation

What do perpetrators accomplish through mass starvation? The question of objectives points our attention to the linkage between political–military strategic goals and the use of starvation as a weapon of war or repression. We introduce a preliminary categorization of nine uses, developed in reflection on the authors’ existing research on famine and atrocities, and illustrated through historical examples.¹³

The below typology is intended to link mass starvation to strategic or tactical objectives: three caveats are in order. Multiple goals can converge and diverge around the creation of starvation conditions. Thus, starvation crimes perpetrated during a siege can create opportunities for profiting from or punishing a population; it can be a prelude to expulsion or a means to gain direct control over people. Separating and controlling a civilian population may harm them terribly, but also relieve armed groups from the burdens of caring for their own civilians; socio-economic transformation creates opportunities for using starvation selectively against communities suspected of disloyalty. Using militias who ‘pay themselves’ through preying on civilians may appear to reduce the culpability of central authorities, but it may also pave the way for a subsequent famine. Working captive labourers to death may allow captors to accumulate resources while punishing or destroying a targeted population.

Additionally, across these cases and the typology, it is important to pay attention to the interaction of local social, economic, political, military and environmental factors that might render starvation crimes more or less destructive to the civilian population that suffers them. In some places, pre-existing vulnerabilities and the combination of acts targeted against a population create a vortex, in which the destructive impacts are magnified in a non-linear way.

A crucial factor of mass starvation is time. While famine or starvation crimes may emerge from conditions for which culpability is relatively easy to deny, the long duration of maintaining the policies that create these conditions undermines claims of innocence. In most of the ‘uses’ of mass starvation described below, deprivation was imposed on thousands of people over multiple years. In no case were those maintaining the conditions ignorant of the impact their decisions had on entire civilian populations. Yet they chose to continue, and in some cases, deepen the crisis that they themselves had created.

A. Killing: ‘Exterminate the Brutes’

The first and simplest aim is mass killing. When members of a victim group are killed en masse by depriving them of food and materials necessary for life, the crime of extermination may apply. When the group is defined as an ethnic, national, racial or religious community, then acts may meet the legal definition of genocide, as ‘deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part’.¹⁴ Starvation as an indirect form of killing has been underappreciated in the study of genocide — a result, as Sheri Rosenberg argued, of the ‘general failure to understand the link between conditions of life that bring about the physical annihilation of the group and the policy decisions that bring about such conditions’.¹⁵ Helen Fein’s introduction of ‘genocide by attrition’,¹⁶ offers a conceptual corrective.

Just as with killing as an act of genocide, starving people need not culminate in the actual death by starvation of an entire population. What is required for starvation to count as genocide is that the perpetrator must demonstrate the intent to destroy a significant part of the population.

Deploying mass starvation to the scale and degree of lethality associated with extermination or genocide is rare. There are few unambiguous cases. It was used as a tool of genocide throughout the Holocaust; in the genocide of the Herero and Nama under German colonial rule in present-day Namibia (1904); the Armenian genocide (1915–1916); Ukrainian Holomodor (1932–1933); and the Nazi Hungerplan (1941–1945) and its constituent elements including the starvation of Soviet prisoners of war, and the elimination of ‘useless eaters’ in Ukraine.

In two key cases, the Herero and Nama, and Armenian genocides, perpetrators were aided by proximity to a desert. The extermination of the Herero and Nama is a paradigmatic case of colonial genocide by starvation. Even after largely defeating a Herero uprising against the European authorities in German Southwest Africa, the newly appointed commander of the territories,¹⁷ Lieutenant-General Lothar von Trotha decided against allowing surrender. He ordered his forces to drive the Herero to the fringes of the Kalahari Desert and seal off access to water, sentencing those who remained in the desert to almost certain death. Herero who tried to flee out of the desert were shot, beaten, hanged, starved and raped. An estimated 40,000 Herero and 10,000 Nama died.¹⁸

In 1915–1916, Ottoman Turkish forces attacked thousands of Armenian villages, massacring, torturing and assaulting the inhabitants. Those who survived the initial onslaught, many of whom were women and children, were brutally deported in a forced march southeast into the arid lands of today’s Syria and Iraq. There, they were again attacked violently, but most deaths were due to hunger and thirst.¹⁹ Mortality estimates do not distinguish cause of death, but indicate that in total, over 600,000 Armenians died.²⁰

Lacking a complicit geography, perpetrators with the intent to destroy a group through starvation require well-developed organizational capacity. This is typically an advanced security apparatus, such as those found in authoritarian regimes at the height of their power. The primary example is Soviet policies that created the Ukrainian Holomodor (1932–1934). Widespread civilian suffering occurred during a period of grain shortfalls resulting from Soviet collectivization policies. In southern Russia, collectivization was eased and the famine was relieved. In Ukraine, Stalin intensified the harsh conditions, imposing policies that prevented movement, removed food and agricultural supplies, blacklisted villages and restricted trade.²¹ Estimates today suggest that 3.3 million excess deaths resulted from Stalin’s policies in Ukraine, most of which occurred in 1933.²²

B. Controlling a Population: ‘War Assumes an Aspect that May Shock the Humanitarian’

When a perpetrator’s primary interest is in reducing the capacity of a group to resist its policies, there may be only a thin line of interpretation separating an intended goal of killing through starvation, from the lack of concern about how many people die as a result of policies intended to control a population. A common context is counter-insurgency.

Resettlement camps provide a stark, emblematic context for examining starvation crimes during counter-insurgency. In camps, responsibility is heightened because of the direct control exercised by perpetrators over the targeted group. A common feature of colonial, post-colonial and Cold War counterinsurgency, resettlement camps demonstrate how policies ostensibly designed to control and separate a civilian population from armed groups, can become incubators of starvation crimes.²³ In his 1906 Handbook for Small Wars, British Colonel Charles Callwell advised his fellow colonial officers that controlling recalcitrant populations required methods such as burning crops and capturing cattle, that would ‘shock the humanitarian’.²⁴

From the Anglo-Boer War (1899–1902), the birthplace of the term ‘concentration camp’, came the tactic of forcing the civilian population into a contained area to control their armed compatriots. Given the resulting humanitarian consequences — 20,000 Boers and 12,000 Africans died²⁵ — and reputational costs for the British, the camps might have offered a crucial lesson in why ‘population concentration run by the military was guaranteed to become a hearts and minds fiasco’.²⁶ Instead, the camp re-appeared in counter-insurgency tactics for decades to follow: for example, in British colonial wars in Kenya against the Mau Mau and in Malaya;²⁷ Portuguese aldeamentos in Mozambique during the war of independence (1964–1973);²⁸ and, as discussed below, ‘forced relocation’ during the Mozambican civil war (1976–1992); ‘resettlement camps’ in Ethiopia (1984–1988); and ‘resettlement centers’ in Burundi (1996–2000).

In each of these cases, a largely rural, dispersed population was forcibly relocated from their homes and livelihoods into ill-prepared and ill-managed, contained spaces. Violence was perpetrated during displacement; crops or food stocks were frequently destroyed; and existing livelihoods disrupted, leaving the interned population wholly dependent on the authorities for their essential supplies. In some cases, civilians were concentrated in camps within their existing provinces, in others they were transferred elsewhere. Other violations occurred inside camp borders, including rape, torture and murder. But in each of the cases noted above, the most widespread civilian harms arose as a direct result of inadequate food, water, housing and health care facilities within camps.

While important variations exist, the overall policy objective in these cases was to separate civilians from insurgents, increase the military’s ability to prevent the population from providing supplies and conscripts to rebels, and enable close surveillance and intelligence gathering. Three factors increased the negative impact of such measures on civilians. First, resettlement moved people away from their farms and livelihoods. Secondly, policies were carried out in haste with inadequate preparations — the civilian populations’ basic needs were, at best, a minor concern. Thirdly, the context of counterinsurgency is rife with distrust and fear between civilians and the military, and the military personnel in these cases were empowered to use coercion.

Two examples help illustrate the scale and duration of harm involved in resettlement. In each case, tens of thousands died as policies were continued over several years. During the liberation war in Mozambique, according to testimony from Roman Catholic missionaries in the Tete district, an estimated 6–8% of the population in each aldeamento died from malnutrition and disease.²⁹ Extrapolating from this estimate to other affected districts, an estimated 36,000–60,000 Mozambicans died as a result of these two conditions while under direct Portuguese control between 1966 and 1973.³⁰

The Ethiopian famine of 1983–1985 is another case that combines multiple kinds of starvation crime in the context of counter-insurgency. The epicentre of the famine overlapped with areas of counter-insurgency in the northern province of Tigray and neighbouring northern Wollo. The main counterinsurgency tactics included burning crops and food stores, aerially bombarding markets, and restricting local trade and labour migration. These contributed to the collapse of the agricultural economy, causing a simultaneous hike in food prices and decline in livelihoods, contributing to famine. Resettlement, another strand of faminogenic counter-insurgency, was the most controversial.³¹ Three rounds of forced resettlement were carried out, the first (1984–1985) at the height of the famine. At least 80,000 people died as a direct result of the government’s policies, due to illness, lack of food, shelter and water, and the absence of the tools and seeds needed to plant crops.³²

In a context where civilian groups are scattered across a large area, starvation crimes may occur when a party to a conflict attempts to control civilians by attacking their livelihoods, in situ. For perpetrators, the advantage is that it reduces the perceptions of — but arguably, not the legal culpability for — the obligation to care for a civilian population. Such patterns are visible in Yemen (2016–present), southern Sudan (1985–2004), Darfur (2003–2005) and South Sudan (2013–2018).

In these cases, perpetrators disrupted food production and distribution (including the population’s capacity to purchase food); and destroyed access to clean water, housing and health facilities. Perpetrators targeted agricultural sites (fields, irrigation and stocks), transportation routes and markets as well. Humanitarian aid was also regularly obstructed. When such acts are combined and carried out over years at a time, despite clear reports of the harm produced, they should be considered as starvation crimes.

C. Gain Territorial Control: ‘Surrender or Starve’

Siege is the signature context for starvation crimes as part of a military effort to seize territorial control. While, in theory, siege can be lawfully conducted, when perpetrators fail to differentiate between civilian and military populations under siege, acts of deprivation can constitute starvation crimes.³³

A complication of siege is that it requires one side to control outside access to a location, while the other refuses to leave (sometimes with good reason for fearing that a fate worse than starvation may follow should they surrender or evacuate). Siege endures as long as one or both sides are willing to prioritize political–military goals over humanitarian concerns of the besieged population.

For besieged armed forces, a civilian population can serve as a shield, a way to hold ground, a source of conscripted soldiers or other labour, or a source of revenue or aid that is utilized by armed forces. While in some cases civilians support the armed forces that are associated with them, rarely is such support consistent across a group or unwavering over time, especially over a long duration, as resources dwindle and if soldiers prioritize their needs at the expense of civilians.

There is no shortage of examples of siege throughout human warfare. The two World Wars produced the notable examples of the blockade of Germany (1916–1919); siege of Leningrad (1941–1944); and the total blockade of Japan in 1945. Among notable post-1945-era examples are: the Nigerian government’s blockade of Ibo-controlled areas (the self-proclaimed Republic of Biafra) during the civil war (1967–1970); the Sudan People’s Liberation Army sieges of Juba and other southern garrison towns (1984–2004); the Bosnian Serbs’ encirclement of the Bosnian government-held ‘safe havens’ during the 1992–1995 war; the Russian siege of Grozny in Chechnya (1999); the Sri Lankan government’s final assault on Tamil areas (2008–2009); the Myanmar government’s encirclement of Rohingya communities (2017);³⁴ and numerous cases throughout the war in Syria (2011–2019). We highlight one such case: the siege of eastern Ghouta in Syria.

Siege has been a regular element of the Syrian war, primarily, but not exclusively, perpetrated by government-aligned ground forces who announced their policy to the Syrian populace: surrender or starve.³⁵ The siege of eastern Ghouta was the longest, enduring for five years (2013–2018). Eastern Ghouta includes several suburban towns on the outskirts of Damascus. Beginning in April 2013, Syrian government forces surrounded the area, which was a stronghold for two opposition armed forces, the Free Syrian Army and the Islam Army, along with several smaller groups. By early 2015, government forces cut off water supplies, and instituted strict controls over the delivery of food and medical supplies, which were grossly inadequate given the size of civilian population under siege.

Throughout the period of siege, government and (in the later period) Russian forces bombed the area and several times allegedly used chemical weapons in densely populated areas. Civilian infrastructure was regularly targeted: housing, schools, hospitals were bombed throughout the siege.³⁶ The population of some 400,000 people survived under extreme duress by devising ways to circumvent the siege, such as digging tunnels and paying exorbitant bribes to get supplies into the area.

Over the course of 2017, as the government sought to push its military advantage and end the war, it tightened the siege while simultaneously increasing its bombardment campaign. Prices of basic commodities soared,³⁷ along with malnutrition levels.³⁸ A Russian-backed, Syrian government offensive was launched on 18 February 2018, and ended on 12 April, when government forces finally gained control of the entire area.

What follows the ending of a siege varies and may illuminate the besiegers’ goals. In the case of eastern Ghouta, mass killing did not follow the siege. But while government services returned, and no widespread retribution was reported, significant curtailment of human rights continued.³⁹ Endings elsewhere vary: successful resistance (Leningrad, Juba, some Bosnian safe havens); capitulation, with punitive reparations (Germany); occupation (Japan); defeat and reintegration (Nigeria); displacement (Myanmar, and some Bosnian safe havens); and, at times, systematic killing (the Warsaw Ghetto and Srebrenica, Bosnia).

D. Material Extraction: ‘The Benefits of Famine’

Conflicts offer opportunities for enrichment; here we focus on contexts of profiteering through sustaining deprivation conditions.

We continue with the example of Eastern Ghouta. In the early days of the siege, a well-connected businessman, Moheiddine Manfoush, struck a deal with the Syrian regime: he would bring milk from dairy farmers inside the besieged Ghouta area into Damascus, and grain from elsewhere into rebel-held territory. With latitude to negotiate passage — reportedly permitted by paying-off regime higher-ups — Manfoush transformed a dairy trade into control over the al-Wafideen crossing, a critical checkpoint that also served as a transfer point for hard currency.⁴⁰ Everything passing through became a source of profit. When the siege tightened in late 2017, Manfoush’s profits ended. Ultimately, the government’s overarching goal of surrender trumped the financial interests of even well-connected war profiteers like Manfoush.

When profiteering is achieved by taking control of land, as in colonial-settler contexts, starving people to force displacement has been rampant. Assaults against indigenous populations in North America and Australia were characterized by destruction of agriculture and homes, with the goal of displacing indigenous populations so settlers could take their land. Some of the most extreme American examples include the ‘trail of tears’ (Cherokee) and ‘the long walk’ (Diné or Navajo), forced relocation accompanied by starvation conditions.⁴¹ Starving people off their lands has not disappeared from the contemporary scene. Contemporary examples have been documented in Somalia (the ‘Bantu Somali’ groups) ⁴² and in Myanmar, where the government besieged Rohingya towns in concert with acts of violence to force displacement.⁴³

Counter-insurgency-related starvation crimes in service of material extraction can also occur when licence is granted to armed groups to prey upon civilians as a method of payment. This was one of the ‘benefits of famine’ for the Sudanese government’s militia proxies who, in the late 1980s, enriched themselves at the cost of reducing the targeted population of southern Sudan to famine.⁴⁴ The government of Sudan has repeatedly deployed ‘counter-insurgency on the cheap’⁴⁵: pursuing military objectives of defeating armed insurgents by seeking out a local militia, whom they supply and empower to operate in an ‘ethics-free zone’ that has recurrently produced famine or near famine conditions.

Darfur in 2003–2005 provides an example: the Sudanese government empowered militias to assault villagers with violence, systematic theft and destruction of water wells and agricultural reserves, which left the civilian population with few options other than flight across the inhospitable environment. Large numbers died. The best evidence suggests that while 30,000 people were killed outright, another 200,000 died through a combination of hunger, disease and exposure as a direct and foreseeable outcome of the conflict and mass displacement.⁴⁶ The deaths would have been far higher had there not been a large humanitarian aid operation. For the government, this form of counter-insurgency offered several advantages: militias could ‘pay themselves’ through expropriation, and the government could argue that it had no direct control over militia actions.⁴⁷

E. Flushing Out a Population: ‘If You Join Us, We Will Feed You. If Not, We Will Kill You’

Deprivation can be used to flush a population out of a remote area into perpetrator control, with some similarities to both siege and counterinsurgency starvation but following a variant pattern. This section presents cases where the primary goal was population-centric: starvation to bring a population under control.

Two cases provide examples: Guatemala and East Timor (today’s Timor Leste). Both share the physical trait of a mountainous interior. Mountains are not necessary, but they are a form of geography in which civilians can flee to the margins, yet not entirely escape from perpetrator control.⁴⁸ Both also demonstrate a mixture of motivations governing the degree of direct violence deployed, with starvation crimes being most effective for groups over whom the governments ultimately wanted to re-assert control.

In Guatemala (1982–1983), government forces intensified a long-standing counter-insurgency campaign by targeting rural Maya villages it perceived as supporting guerillas, and using selective development to restructure rural life: a policy it described as ‘guns and beans’.⁴⁹ Initial assaults deployed lethal violence against villagers and destroyed village structures. Many civilians escaped into the mountains. The army continued its pursuit, repeatedly attacking people and any makeshift mountain agriculture and livestock reserves they discovered.⁵⁰ Survivor accounts of the time in the mountains are replete with descriptions of extreme hunger and illness.⁵¹ Many people survived for two years in the mountains; continually fleeing and attempting to find or cultivate sources of food, only to be chased by the military and have their crops destroyed.

Civilians and armed groups often mixed, as both were hiding in the mountains. When civilians decided to surrender by moving back into army-controlled territories, guerillas often prevented them.⁵² Nonetheless, by late 1983, many civilians could hold out no longer against the policy described by a Guatemalan intelligence colonel as: ‘If you join us, we will feed you. If not, we will kill you.’⁵³ For the army, which sought to defeat the guerillas and control the indigenous population, hunger was an effective weapon to flush people out of the mountains and into villages where they could be controlled.⁵⁴

A similar use of starvation was seen in East Timor in 1975.⁵⁵ After the end of Portuguese colonial rule of Indonesia, the Revolutionary Front for an Independent East Timor (Frente Revolucionária do Timor-Leste Independente, or ‘Fretilin’) unilaterally declared independence on 28 November 1975. Indonesia invaded nine days later, violently targeting Fretilin leaders and their families. Fretilin forces resisted by fleeing to the mountainous interior with many civilian supporters and prisoners, drawn from people who had collaborated with the Indonesian military (whom they eventually massacred).

Indonesian forces intensified their attacks and disrupted food supplies, resulting in widespread civilian death from hunger and disease. Many civilians wanted to surrender, but the guerrillas prevented them. After a substantial troop build-up and the acquisition of new American and British military equipment, the Indonesia military began to overpower Fretilin bases, systematically destroying livestock and food sources. In late 1978, the last Fretilin stronghold at Mount Matebian fell. Civilians surrendered; Fretilin leaders and followers were pursued, captured and killed. Many civilians who fell under Indonesian control were hurdled into ‘strategic villages’. In December 1978, there were fifteen camps holding about 270,000 people, under poor health and sanitary conditions with limited food.⁵⁶ In total, an estimated 18,600 people were killed by violence, and 84,200 died of hunger and disease.⁵⁷

F. Punishment: ‘Put Them on a Diet’

In some instances, policies of deprivation are intended to alter the victim group’s political calculus, without obliging the perpetrator to take control over either territory or civilian population, and thereby assuming responsibility for governing. This may occur in the form of widespread sanctions, blockade, siege and persecution as forms of collective punishment.

Examples include the sanctions on Iraq (1991–1996),⁵⁸ the siege of Sarajevo (1992–1995), and the blockade of Gaza (2007–present). In each of these cases, those imposing restrictions aimed to calibrate the flow of objects indispensable for life, to create the optimal level of human suffering across a civilian population to force a political change, but without invoking unsustainable reputational costs. As Dov Weisglass, an advisor to Israeli Prime Minister Ehud Olmert, reportedly stated in 2006 regarding Israeli control over goods flowing to Gaza: ‘The idea is to put the Palestinians on a diet, but not to make them die of hunger.’⁵⁹

Economic sanctions became more attractive to powerful states at the end of the Cold War, when UN Security Council consensus made it possible to pass comprehensive sanctions regimes. Sanctions seemingly offered a powerful tool requiring low outlay of domestic political will for those seeking to coerce political change elsewhere.⁶⁰ This was the context in which widespread sanctions were imposed by the UN, led by the United States, on Saddam Hussein’s Iraq at the end of the first Gulf War. The stated purpose of the sanctions was to coerce compliance with UN resolutions demanding that Iraq destroy its long-range ballistic, chemical and nascent nuclear weapons, end support for terrorism, and pay reparations to Kuwait and foreign debts.⁶¹

Independent assessments of the civilian suffering resulting from the comprehensive sanctions suggest that civilian suffering, including elevated child mortality, was substantial.⁶² For example: a UNICEF report found a 72% increase in the level of malnutrition for children under five years of age, between the years of 1991 and 1996.⁶³ For those imposing the sanctions, claims of civilian suffering were brushed aside. U.S. Secretary of State Madeleine Albright, when asked about reports that half a million children had died as a result of sanctions, replied, ‘I think this is a very hard choice, but the price, we think the price is worth it.’⁶⁴ She later expressed regret for the statement, but it summarizes the explicit trade-offs between humanitarian and political concerns imposed by comprehensive sanctions. In response to evidence of civilian suffering, five years after sanctions were put into place, an effort was made to mitigate the harmful impacts on the civilian population, through the United Nations Oil-for-Food programme.⁶⁵

In the Bosnian capital of Sarajevo during the Bosnian war (1992–1995), siege functioned as collective punishment. Sarajevo is in a valley; during the war, the Bosnian Serbs deployed heavy weapons in the high grounds that surround it and pushed the frontlines into the city itself, cutting off outside access. However, it does not appear that the Bosnian Serbs wanted to destroy or entirely capture the city. The Bosnian Serbs expressed a strategic goal of dividing the capital, which they largely accomplished with their initial offensive in 1992. The value of maintaining the siege over the duration of the conflict was, as Robert Donia argued, ‘to use Sarajevo and its citizens as hostages to strengthen their negotiating position’.⁶⁶

Given the swarm of international media in Sarajevo, violence and humanitarian conditions there received much greater attention than elsewhere in the country. Taking advantage of disproportionate attention, the Bosnian Serb army increased bombardment of the city in relation to military or negotiating setbacks elsewhere.⁶⁷ Various health and nutritional studies, conducted in Sarajevo over the course of the war and in its aftermath, documented instances of malnutrition, decreased body mass, and deficits in micronutrients, but no crisis-level deprivation.⁶⁸ In its prosecutorial strategy, the International Criminal Tribunal for the former Yugoslavia (ICTY) focused on acts of killing during the siege. While sniping and bombing deserved legal sanction, by focusing solely on death as an outcome, the ICTY missed the use of starvation crimes as a form of collective punishment.

The Israeli blockade of Gaza (2007–present), while deemed legal in the eyes of the Palmer Report,⁶⁹ follows a similar logic. Instituted in 2007, when Hamas was voted into office in the territory, the blockade functions as collective punishment against the population of the Gaza strip. Israel argues that the restrictions are justified and legitimate to prevent or respond to violence originating from within Gaza. Nonetheless, the restrictions have included broad punitive measures that ‘limit supply of electricity and water, restrict access to medical care and educational and economic opportunity, and perpetuate poverty’.⁷⁰ Due to tightened restrictions in 2018, civilians in Gaza saw significant increases in food insecurity: 47% of households in Gaza were reported as severely food insecure in a September 2018 survey, a point at which the Israelis and Egyptians coordinated restrictions to reduce the flow of materials into Gaza.⁷¹

In each of the above cases, those imposing the restrictions were subject to limitations imposed by international scrutiny. The U.S. and Israeli governments have significant resources dedicated to managing blockades and international perceptions thereof, so as to minimize the legal implications of their actions. This was not so for the Bosnian Serbs, who deployed starvation crimes in conjunction with large-scale, violent ethnic cleansing offensives.

G. Exploitation: ‘Human Raw Material’

Across diverse contexts, creating or sustaining starvation conditions to exploit civilians for their labour or as a means to extract resources reduces humans to what Aleksandr Solzhenitsyn termed ‘human raw material’⁷² in his description of life in the Soviet Gulag.

Slave labour in contexts where human resources are plentiful and food resources are meagre has frequently produced mass starvation. Examples include forced labour of civilians (many of them Jewish, or foreign nationals) under Nazi Germany,⁷³ and within the Soviet Gulag.⁷⁴ A common context concerns prisoners of war (PoWs) or persons deprived of liberty for reasons related to conflict who, under the Geneva Conventions, should not be subjected to starvation, e.g. during World War II by the Japanese anud Germans, and immediately after the war by the USSR (with particularly high-mortality rates). The case of British PoWs forced in starvation conditions to build a railway connecting Japanese-controlled areas in Thailand and Burma — epitomized in the novel by Pierre Boulle (1952) and the subsequent film ‘The Bridge on the River Kwai’ (1957) — provides a stark example from the eastern front of the Second World War.

Here we highlight another case from the same era: German PoWs in the Soviet Union (1940s–1957). Conditions for German PoWs varied depending on their captors.⁷⁵ Across western Europe, German PoWs were forced into labour, but nowhere were their working conditions as bad as in the Soviet Union. Post-World War II Soviet resources were allocated to their own population, and poor postwar harvests worsened the situation for PoWs who were forced to rebuild the war-destroyed country. Many were sent to logging camps in Siberia or mines in the Ural Mountains where they were subjected to ‘brutal assaults on a daily basis, hunger, disease, and the cold’.⁷⁶ Until 1947, the single highest cause of death was dystrophy, caused by undernourishment.⁷⁷ One-third of Soviet-held PoWs — an estimated 1.1 million — perished.⁷⁸ The PoWs who returned to both East and West Germany in the late 1940s were ‘ragged and emaciated’,⁷⁹ with the last survivors returning in 1957.

In a dramatically different context, southern Sudan offers an example of exploiting starvation conditions for civilians as a means to access relief supplies for combatants. Operation Lifeline Sudan (OLS) was launched in 1989, the first-ever UN humanitarian operation that crossed a front line to rebel-held areas.⁸⁰ Initially envisioned as a short-term measure to relieve famine while the warring parties negotiated a peace deal, it soon became entangled in a protracted war. Due to the lack of road infrastructure and widespread flooding across much of southern Sudan, OLS relief was often delivered by air, meaning that it reached a very specific location and, once the aircraft had left, was in the hands of the local authorities. This created a situation in which aid was ripe for abuse. In the mid-1990s, contending factions of the rebel Sudan People’s Liberation Army used severely malnourished children as a lure to attract air deliveries of aid. Based on measured rates of malnutrition, OLS would prioritize locations for relief flights. The food was first distributed to the population, targeting the families with the most severely malnourished children, but when the aid workers left, soldiers from the armed group in control of that particular location would collect the rations at gunpoint and keep most of it for their own supplies. Severe child malnutrition thereby became the means to gain relief resources.

H. War Provisioning: ‘Soldiers Eat, Peasants Provide’

In all non-industrial countries, armies have typically lived off the land, feeding themselves from the local communities. Famines followed in the wake of the armies in Europe, Asia and Africa—including in those armies’ own countries. In response to farmers’ complaints against troops billeted in their village, the nineteenth century Emperor Tewodros of Ethiopia tersely remarked, ‘soldiers eat, peasants provide.’⁸¹ In World War Two, Japanese troops were dispatched to the archipelagoes of Indonesia and the Philippines with rations for only the first few days, after which they were ordered to provision themselves from the local island populations.⁸² In all recorded cases of besieged cities, the greater burden of hunger falls on the most vulnerable, especially civilians, and less so on soldiers.

What singles out these cases is that the hunger is inflicted not only on ‘enemy’ populations, but also on those who are the supporters or the subjects of those responsible for the starvation.

The Bengal famine of 1943 is a significant case of a wartime economic policy that caused famine. Faced with the requirement to raise resources to finance the war effort, the British government pursued a policy that massively curtailed mass consumption in India, but not at home in Britain where extreme austerity would have risked unrest. In India, in contrast, the wartime financial strategy was based on forced savings through a policy that John Maynard Keynes called ‘profit inflation’. The greater burden of this hardship fell in Bengal, which was both the major industrial centre and close to the frontline with Japanese forces. According to Utna Patnaik, this was the principal driving force behind the Bengal famine of 1943: ‘these deaths were the direct result of the Keynesian policy of profit-inflation deliberately followed at that time by the British and colonial governments with a very specific purpose, to raise resources from the Indian population by curtailing mass consumption, in order to finance the Allies’ war in South Asia with Japan’.⁸³ Additional factors causing famine included the impounding of the fishing fleet fearing that the boats might be used by the Japanese army to invade (causing mass unemployment among fisher people) and the refusal of the War Cabinet to divert ships laden with grain, sailing from Australia to Britain, to Calcutta in order to break the rapid food price inflation that drove the cost of rice far out of the reach of many people.⁸⁴

I. Comprehensive Societal Transformation: ‘Silent Violence’

Engineering massive societal transformation is not a crime in itself. Economic policy, however grievously mistaken in its conception or calamitous in its outcomes, is not normally subject to criminal proceedings. The question of culpability is raised, however, when policies clearly demonstrate harmful outcomes, and yet are adhered to over a longer duration even as people die en masse. Criminality in these cases may be attached to ruthless implementation of policies, suppression of dissent, and refusal to alter policies when the scale of the devastation becomes apparent. When investigating potential criminal culpability in cases such as North Korea, lawyers have focused less on the economic policy, and more on brutalities committed during its execution or failure to mitigate extreme preventable suffering that resulted.⁸⁵

The most striking examples of mass starvation through titanic social engineering are attempts at agro-economic transformation by 20th century Communist regimes. Earlier instances include the incorporation of entire economies into imperial systems of production, which caused some of the worst-recorded famines on record. Colonization was a process of coercively re-orienting economies to meet the economic and political goals of the metropole. In many locations, this produced starvation: native populations were often subjected to famine conditions as a result of policies to displace them and appropriate their land (as noted above), or to compel them to produce cash crops for the imperial power. Michael Watts gave the title ‘Silent Violence’ to his study of the political economy of famine in colonial Nigeria.⁸⁶ Imperial British policies in India were perhaps the most vivid example of a colonial power re-orienting a society in a manner that produces famine. These policies contributed to recurrent famines from the 1770s to the 1940s that resulted in the deaths of tens of millions.⁸⁷

The first great colonial famine in India was the Bengal famine of the 1770s, which killed as many as ten million. During this period the East India Company relentlessly increased its taxation and its profits, transferring an ever-increasing quantity of expropriated riches to its shareholders. This was the model for the colonial transformation of subjugated economies, including the Belgian Congo, Dutch Indonesia and French Indochina.

Soviet agrarian reform, the ‘Great Leap Forward’ in China, and Cambodia under the Khmer Rouge, are the leading contemporary examples of forcible socio-economic transformation. The famines of 1930–1934 in the USSR, affecting southern Russia, Kazakhstan and Ukraine, began with forcible collectivization of peasant farming. The focus of liability has been mostly on the targeting of Ukraine; nonetheless, the question of potentially criminal liability for misconceived and brutally implemented agrarian reform should not be overlooked. The starvation of Kazakhstan in 1930–1933 is a particularly egregious case. For a previously semi-nomadic population, forced sedentarization into planned villages, along with collective farming with onerous production quotas, was an exceptionally brutal form of ‘modernization.’ As much as 40 percent of the Kazakh population died.⁸⁸

The highest recorded fatalities in famine occurred as a result of Communist economic transformation in China under Mao Zedong. During the ‘Great Leap Forward’ (1958–1962), an estimated 25–30 million people died as a direct result of the dismantling of peasant agriculture and its replacement with cooperatives and village steel production. Widespread suffering and death were met with stubborn refusal to alter policies.⁸⁹

The policies of the Khmer Rouge in Cambodia were possibly even more extreme than those of Mao, although they impacted a smaller population. Under Pol Pot’s efforts to remake Cambodian society and transform its peasant farming to a collective agricultural system, there was immense upheaval and deprivation. Of the estimated 1.75 million people killed under the Khmer Rouge, 69 percent died of starvation and related causes.⁹⁰ Widespread starvation in Cambodia was, as Randle deFalco demonstrates, ‘a direct, foreseeable and avoidable consequence of Khmer Rouge policies, and that over time the regime’s leaders became aware of their policies disastrous effects, but nevertheless continued to enforce them strictly’.⁹¹

The question of culpability comes into play chiefly as a side effect of the inhuman consequences of flawed policies. First, societal transformations of this scale can be affected only with coercion and repression. Secondly, these policies are often implemented selectively and punitively, in pursuit of other goals such as ethnic repression. Thirdly, as the calamitous consequences of these policies become evident, no efforts are made to change course or mitigate the suffering.

5. Conclusion: Why ‘Starvation Crimes’ Now?

As has become clear in relation to acts of direct violence, clarifying criminal liability does not necessarily provide deterrence or adequate recompense for victims. However, doing so marks an important first step in how such actions are understood and documented, and thereby increases the possibility for legal and political accountability. While successful prosecution for starvation crimes might still be a distant prospect, stigmatizing the acts that produce mass starvation by treating them as criminal can alter the social and political calculus of what is deemed acceptable.

Activating the law concerning starvation is also a timely endeavor. Between 2000 and 2011 there were no episodes formally identified as famine. Due to greater expertise in predicting, preventing, mitigating and responding to crises before they reach famine, humanitarian organizations are capable of greatly reducing mortality during periods of upheaval. And yet, in 2017, the United Nations identified four situations of acute food insecurity that threatened famine or breached that threshold—in north-eastern Nigeria, Somalia, South Sudan and Yemen.⁹² Syria was not included; the UN-recognized Integrated food security Phase Classification (IPC) system was not gathering data in that country.

While each of these cases emerged from a distinct constellation of issues, there are some potentially troubling global trends that may increase or alter vulnerabilities to mass starvation. Today, there is greater attention to direct killing of civilians in conflict, including counting those killed and attempting to achieve political and legal accountability for atrocities. This welcome advance may inadvertently lead perpetrators of violence to deploy acts of deprivation in an effort to mitigate their perceived responsibility for such internationally salient forms of suffering while still achieving the outcome of civilian deaths. Globalization of food markets and urbanization have also altered vulnerabilities: as populations increasingly rely on integrated and trade-dependent systems of food production and distribution, rather than local production, the disruptions of war have the potential to quickly create pockets of profound need. This is not just a matter of food availability: for populations reliant on purchasing food, economic collapse through extended warfare, as has been seen in Yemen and in pockets of South Sudan, creates conditions in which food is available, but priced out of people’s means to access it.

The ever-expanding interconnections of global economies increases exposure to outside shocks: during 2011, in Somalia, a spike in global food prices compounded by other factors, including the actions of conflict actors and US-imposed restrictions on aid to groups designated as ‘terrorists’, created famine.⁹³ In pre-war Yemen, what little agricultural production was possible had been steadily shifting towards an export market, rather than local consumption. In such circumstances, the use of starvation in warfare can cause a food economy to collapse. Finally, the consequences of climate change introduce new challenges, risks and sites of exposure.

It is important to note that those who suffer most from starvation crimes are pregnant and lactating women, children under the age of five—who are extremely vulnerable to severe acute malnutrition, and the elderly. This portion of the population dies in greatest numbers. Young children who survive such deprivation may face lifelong detrimental health impacts. Exposure to malnutrition at a young age can lead to cognitive development impacts that cannot be overcome, regardless of efforts to invest in post-crisis educational initiatives. Even children in utero during famine may experience higher lifelong rates of illness, chronic disease and reduced life expectancy.⁹⁴ Further, these survivor groups often remain at heightened risk if conflict returns and imperils their well-being at a later point. With high rates of relapse, as humanitarian aid expert Patrick Webb noted, populations exposed to famine or near-famine conditions form a ‘reservoir of mortality for the next crisis’.⁹⁵ Starvation crimes are insidious, have a persistent impact, and most harm those who are clearly civilians: mothers and young children.

Mass starvation has throughout history been mis-categorized as a natural phenomenon, or an unfortunate side-effect of conflict and political oppression. The numbers and names of its victims fade into the background, blurred traces of the horrors of history. This is both an inaccurate understanding of starvation crimes and an injustice to victims.

Footnotes