Colonial Soldiers in Italian Counter-Insurgency Operations in Libya, 1922-32

British Journal for Military History, Volume 1, Issue 2, February 2015 Colonial Soldiers in Italian Counter-Insurgency Operations in Libya, 1922-321 NIR ARIELLI University of Leeds Email: [email protected] ABSTRACT The vast majority of the force employed by the Italians to crush local resistance in Tripolitania and Cyrenaica was composed of Libyans, Eritreans and Ethiopians. The article examines why the Italians came to rely so heavily on colonial soldiers. It highlights two key predicaments the Italians faced: how to contend with the social, economic and political repercussions that military recruitment for the counter-insurgency created in East Africa; and the extent to which they could depend on forces raised in Libya itself. Finally, the article offers an initial assessment of how the counter-insurgency exacerbated tensions between Libyans and East Africans. Introduction The insurgency in Iraq from 2003 onwards has created a renewed interest in the comparative study of counter-insurgencies. British, American and French counterinsurgency campaigns in the twentieth century have received most of the attention.2 Little has been said in this context about the lengthy, difficult and bloody attempts by the Italian government and its armed forces to subdue the colonies of Tripolitania and Cyrenaica which were later united and renamed Libya.3 Indeed, events in Libya predated the birth of the term 'counter-insurgency'. But like other colonial 'small wars' of the nineteenth and early twentieth century, the suppression in Libya fits into the definition of counter-insurgency used by David Kilcullen and others – 'all means adopted to suppress an insurgency'; in this case a 'resistance insurgency' in which the 1 The author wishes to thank Bruce Collins, Shane Doyle, Luigi Goglia and Brian Sullivan for their invaluable suggestions. 2 See, for instance: Philippe Francois, 'Waging counter-insurgency in Algeria: A French point of view', Military Review, Vol. 88, Iss. 5, (2008), pp.56-67; Gian Gentile, ‘A strategy of tactics: population-centric COIN and the army’, Parameters, Vol. 39, Iss. 3, (2009), pp.5-17; David French, The British Way in Counter-Insurgency, 1945-1967, (Oxford: Oxford University Press, 2011); Larisa Deriglazova, Great Powers, Small Wars: Asymmetric Conflict since 1945, (Baltimore: Johns Hopkins University Press, 2014). 3 For rare exceptions, which mention Italian counter-insurgency efforts in Libya briefly, see: Alexander B. Downes, 'Draining the sea by filling the graves: investigating the effectiveness of indiscriminate violence as a counter-insurgency strategy’, Civil Wars, Vol. 9, Iss. 4, (2007), pp.420-444; Simon Anglim, 'Callwell versus Graziani: how the British Army applied ‘small wars’ techniques in major operations in Africa and the Middle East, 1940–41', Small Wars & Insurgencies, Vol. 19, Iss. 4, (2008), pp.588-608. www.bjmh.org.uk 47 COLONIAL SOLDIERS IN LIBYA 1922-32 occupying Italians were the initiators while the insurgents sought to preserve the status quo.4 Debates about events in Libya during the colonial period have largely been restricted to Libya itself, Italy and historians of these two countries. Italy's military strategy and operational conduct during the repression of local resistance have been examined by Giorgio Rochat, Angelo Del Boca, Nicola Labanca, Brian Sullivan, John Gooch and, recently, Federica Saini Fasanotti.5 These studies have highlighted the Italians' tactical learning curve. They also brought to light the tensions between the attempts of colonial and military leaders to win local hearts and minds on the one hand, and far more radical actions akin to ethnic cleansing on the other.6 Luigi Goglia has offered an in-depth analysis of the capture, trial and execution of Omar al-Mukhtar, the revolt's most prominent leader.7 Anna Baldinetti and Ali Abdullatif Ahmida studied the impact of the revolt and its repression on Libyan society and national identity.8 One aspect of the counter-insurgency which has not been studied sufficiently is the role played by colonial troops, the vast majority of the force employed by the Italians to crush local resistance. European states were slow to recognise the immense debt they owed to the peoples of their colonies for their military contribution during the two world wars. The discussion of the role of indigenous recruits in colonial counter-insurgencies remains largely in the shadows. The Italian case study provides an extreme, though not unique, example of European reliance on colonial soldiers for the preservation and enlargement of an imperial edifice. The article will address this topic by focusing on four main themes. First of all, it will analyse why the Italians came to rely so heavily on colonial soldiers to do most of their fighting. Secondly, it will illustrate how social, economic and political developments in the Italian colony of Eritrea, and in East Africa more broadly, impacted on patterns of recruitment for the 4 David Kilcullen, 'Counterinsurgency redux', Survival, Vol. 48, Iss. 4, (2006), p. 2. Giorgio Rochat, 'The repression of resistance in Cyrenaica (1927-1931)’, in Enzo Santarelli et al (ed.), Omar al-Mukhtar: the Italian Reconquest of Libya (London: Darf, 1986), pp.35-115; Angelo Del Boca, Gli italiani in Libia: dal fascismo al Gheddafi, (Milan: Mondadori, 1994); Nicola Labanca, La guerra italiana per la Libia, 1911-1931, (Bologna: Il Mulino, 2012); Brian Sullivan, 'A thirst for glory: Mussolini and the Italian military, 1922-1936' (PhD dissertation, University of Columbia, 1983); John Gooch, 'Re-conquest and suppression: Fascist Italy's pacification of Libya and Ethiopia, 1922-1939', Journal of Strategic Studies, Vol. 28, Iss. 6, (2005), pp.1005-1032; Federica Saini Fasanotti, Libia, 1922-1931, (Rome: Stato maggiore dell'esercito, 2012). 6 Some historians have gone as far as accusing the Italians of committing genocide: Eric Salerno, Genocidio in Libia: Le atrocità nascoste dell’avventura coloniale italiana, (Rome: Manifestolibri, 2005). 7 Luigi Goglia, ‘The capture, trial and death of Omar Al-Mukhtar in the Italian press’, in Enzo Santarelli et al (ed.), Omar al-Mukhtar: the Italian Reconquest of Libya (London: Darf, 1986); Luigi Goglia, ‘L’epilogo della lotta di Omar al Mukhtar in un documento della Regia Aeronautica’, Africana, 1 (1994-95), pp. 89-98. 8 Anna Baldinetti, The Origins of the Libyan Nation, (London and New York: Routledge, 2010); Ali Abdullatif Ahmida, Forgotten Voices: Power and Agency in Colonial and Postcolonial Libya, (London and New York: Routledge, 2005). 5 48 British Journal for Military History, Volume 1, Issue 2, February 2015 counter-insurgency in Libya. The third theme is the Italian dilemma regarding the loyalty of the forces they raised in Libya itself. Finally, the article will offer an initial assessment of how the counter-insurgency exacerbated tensions between East Africans and Libyans. This four-levelled approach illustrates the crucial importance of indigenous forces in Italy’s counter-insurgency operations. It also highlights how the recruitment of colonial troops could have social and political implications that go far beyond the immediate repression of uprisings. The Military Context Before examining the role of colonial soldiers, it is necessary to briefly survey the history of Italian efforts to subdue Libya. Italian forces first landed in Tripolitania and Cyrenaica, then two provinces of the Ottoman Empire, in October 1911. When the Italo-Turkish War ended in 1912 the territories under Italian control were primarily those close to the Mediterranean coastline. In 1913 the Italians advanced into the Fezzan in south-western Libya and managed to capture the main oases by early 1914. However, supply lines were long and vulnerable and began to come under attack in the summer of 1914. By the end of the year, the Italians had been driven out of the Fezzan and, at the beginning of August 1915, they only held four coastal towns in Tripolitania. In Cyrenaica, the forces of the Sanusi Order prevented Italian troops from expanding past the coastal areas. In both regions the Italians were forced to sign a series of accords giving local leaders autonomy. In 1918 urban notables and rural leaders formed the Tripolitanian Republic (1918-1920) and a Tripolitanian parliament was set up in June 1919. In October 1920 the Italian government granted members of the Sanusi family a monthly allowance and agreed to a Cyrenaican parliament. The latter was formed in April 1921 and functioned to the end of March 1923.9 The more or less peaceful co-existence between the Italians and the semiautonomous Arab entities in Libya came to an end in January 1922 when the Colonial Minister Giovanni Amendola and Count Giuseppe Volpi, the Governor of Tripolitania, launched the military ‘re-conquest’ of the colony. The first zone to be cleared of hostile tribes was to the east and west of Tripoli. The Tripolitanian nomad tribes were divided and the Italians endeavoured to capitalise on these divisions. They distinguished between ‘rebel’ and ‘submitted’ (sottomessi) tribes and allowed the latter to carry arms for self-defence. The future Marshal Rodolfo Graziani, who earned his reputation as a maverick colonial commander in Tripolitania, also sought to exploit ancient feuds between Arabs and Berbers. In a series of campaigns the Italians managed to extend their control first to the hinterland around Homs and then gradually pushed southwards and eastwards. By the end of 1925 the Italians' 9 John Wright, Libya: A Modern History, (London: Croom Helm, 1981), pp. 29-31; Ahmida, Forgotten Voices, p. 30; Sullivan, ‘A thirst for glory’, pp.213-214; Baldinetti, The Origins, p.45. www.bjmh.org.uk 49 COLONIAL SOLDIERS IN LIBYA 1922-32 position in Tripolitania was much more stable, allowing Benito Mussolini to visit Tripoli in April 1926 (Italy’s military attaché in London would later claim that the Fascist government was ‘sufficiently strong to impose the needful coercion on the territories’ whereas the previous Liberal government had been weak and indecisive). The last major operations in western Libya were the capture of the oases along the 29th parallel and the seizure of Sirtica and the coastal area linking Tripolitania and Cyrenaica in the early months of 1928.10 The subjugation of Cyrenaica proved much more difficult and lengthy. The Italians launched a surprise attack on Sanusi camps in March 1923. Two days later Colonial Minister Luigi Federzoni renounced all agreements with the Sanusi. The initiative soon shifted to the rebels who ambushed and inflicted heavy losses on 7th Eritrean Battalion at Bir Bilal in June. Follow-up attacks pushed the Italians back everywhere. Unlike Tripolitania, where the rebels did not have a central command and the Italians could literally divide and conquer, in Cyrenaica the resistance was united and led by the very able Omar al-Mukhtar. The rebels used small, mobile formations which tended to melt away and evade capture when confronted by a superior force. The mountainous and wooded terrain of the hinterland south of the coastal plain assisted the rebels and hindered Italian efforts. To overcome their difficulties, Italian commanders tried a number of tactics which included, at one stage, the construction of a line of fortified posts.11 Nonetheless, the resistance persisted and the Arabs spoke till very late in the campaign of ‘the government of the day and the government of the night'.12 In late 1928 Marshal Pietro Badoglio, the Chief of the Armed Forces General Staff, was appointed joint governor of Tripolitania and Cyrenaica. In 1929 Badoglio agreed to a truce with Omar al-Mukhtar which lasted a few months. When the truce collapsed the counter-insurgency entered its final and most brutal stage. In March 1930 Graziani, who had distinguished himself in Tripolitania, was appointed as Badoglio's deputy and charged with crushing the revolt in Cyrenaica. Badoglio wanted to isolate the rebels from the rest of the population. The towns along the coast were firmly under Italian control but the hinterland was not. On 25 June 1930 Italian troops began clearing out the mountainous region of Jabal al-Akhdar. Within nine months between 80,000 and 100,000 people were driven out of their homes, made to march great distances and confined inside barbed wire enclosures in a 10 Sullivan, 'A thirst for glory', p.228; Saini Fasanotti, Libia, p.127; A. G. Coppi, ‘Recent operations of the Italian troops in Libya’, RUSI Journal, Vol. 74, Iss. 493, (1929), p. 148. 11 Sullivan, 'A thirst for glory', p.231; Saini Fasanotti, Libia, p.248; Rochat, 'The repression', p.42. 12 E. E. Evans-Pritchard, The Sanusi of Cyrenaica (Oxford: Clarendon Press, 1949), p.173. 50 British Journal for Military History, Volume 1, Issue 2, February 2015 number of concentration camps on the coastal strip.13 Conditions in these camps were terrible. Estimates of the human death toll vary between 40,000 and 65,000.14 In late 1930 and early 1931 Graziani led an expedition deep into the desert to capture the remote rebel stronghold of Kufra. To prevent the rebels from finding refuge in and obtaining supplies from Egypt, a 270km barrier of barbed wire entanglements was constructed along the Egyptian border between April and September 1931. Increasingly isolated, Omar al-Mukhtar, by now in his seventies according to some estimates, was eventually caught under his dead horse during a skirmish. After a brief trial he was executed on 16 September 1931 in the concentration camp at Suluq, south of Benghazi, in front of some 20,000 deportees. On 24 January 1932 Badoglio declared that the rebellion was officially over and praised Graziani. 15 By the end of the uprising, Badoglio would have undoubtedly endorsed David French's conclusion that the 'key to a successful counter-insurgency campaign was not winning the hearts and minds of the population, but establishing physical control over them’.16 'Warrior races' and 'warlike virtues' After independence, the participation of colonial soldiers in crushing local resistance was played down by the Libyan state. In Lion of the Desert (1981), a film primarily funded by Muammar Gaddafi's regime, tarboush-wearing East Africans appear from time to time, but the main brunt of the fighting is carried out by Italian soldiers and Fascist black-shirts. No Libyan soldiers fighting on the Italian side are to be seen.17 However, a quick glance at the units deployed in Libya throughout most of the period of ‘re-conquest’ reveals that Italian counter-insurgency operations were conducted primarily by colonial troops. 13 Sullivan, 'A thirst for glory', pp.242-46; Evans-Pritchard, The Sanusi, pp.88-89; Ahmida, Forgotten Voices, p.42; Baldinetti, The Origins, p.48. On the difficulty of estimating the exact number of Libyans who died in the camps, see: Luigi Goglia, ‘A proposito di una biografia su Italo Balbo’, Africa, Vol. 42, Iss. 1, (1987), pp.152-157. 15 Gooch, 'Re-conquest and suppression', p.1020; Rochat, 'The repression', p.87-92; Archivio Centrale dello Stato [ACS], Fondo Graziani [FG], s.8, f. 12/1, 'Ordine del giorno', 24 January 1932. 16 French, The British Way, p.7. 17 For more on the film see: Clarissa Clò, 'Mediterraneo interrupted: perils and potentials of representing Italy’s occupations in Greece and Libya through Film', Italian Culture, Vol. 27, Iss. 2, (2009), pp.104-109. 14 www.bjmh.org.uk 51 COLONIAL SOLDIERS IN LIBYA 1922-32 Table 1: Troops station in Libya, 1914-192818 Year Colonial troops Italian troops 1914-15 19,000 50,500 1915-16 14,000 61,000 1916-17 18,000 48,000 1917-18 18,000 48,000 1918-19 18,500 57,000 1919-20 21,500 25,000 1920-21 21,300 5,700 1921-22 23,100 4,500 1922-23 31,600 1,900 December 1924 23,792 11,223 December 1925 23,990 13,163* December 1926 22,723 14,130* September 1927 25,820 17,448* November 1928 28,558 12,672* *The number includes a contingent of volunteer members of the Milizia Volontaria per la Sicurezza Nazionale (MVSN), also known as 'Black Shirts'. During the First World War colonial officials in Rome were very reluctant to have 'coloured' troops serving in Italy. For Colonial Minister Gaspare Colosimo the preferred solution for manpower shortages on the Italian front in 1916 was to send Eritrean forces to Libya to relieve 'white' troops, who would then be sent to fight in northern Italy.19 However, when it came to the counter-insurgency in Libya, Italian priorities were reversed. Already in early operations in Tripolitania in 1922, instead of relying on 'white' troops, Italian officers forged a force that combined Eritrean battalions (which had an average of 750 men per battalion), Libyan camel- and horsemounted units, and irregular bands. In 1923, when Colonial Minister Federzoni discovered that no more troops could be sent from Eritrea, he proposed to bring metropolitan troops to furnish garrisons for forts and advanced posts, freeing the 'coloured' troops to be used for offensive operations.20 18 ACS, Fondo Badoglio [FB], b.1, f.3, sf.5; Ministero Africa Italiana, Direzione Generale Affari Politici, Archivio Segreto [MAI, AS], b.1, f.1/2. 19 Archivio Storico del Ministero dell’Africa Italiana [ASMAI], p.115/3, f. 36-37-38, Colosimo to War Ministry, 11 May 1916; Martini to Tripoli, 16 May 1916; Colosimo to Asmara, 23 June 1916. 20 Gooch 'Re-conquest and Suppression', pp.1007-1008. 52 British Journal for Military History, Volume 1, Issue 2, February 2015 Why were colonial soldiers so central for the Italian effort? The use of colonial troops in Libya offered three clear advantages: political expediency, cost and suitability. The political advantage of employing colonial soldiers instead of Italian conscripts derives from what we would now call loss aversion. Captain Rodolfo Corselli, who took part in the war of 1911-12 and remained in Libya until 1913, pointed out in his book La guerra in colonia (1914) that the loss of colonial troops had fewer repercussions in the metropole and consequently placed less pressure on colonial policy.21 The First World War, during which Italy suffered heavy losses, contributed to the reluctance to commit very large metropolitan forces to the Libyan theatre. Furthermore, sending Italian conscripts to fight prolonged campaigns in Africa was ineffective. An inter-ministerial committee, which looked into potential manpower solutions for Libya in 1926, was told by military officials that conscripted troops were 'discontent' to serve in the colonies, exerted pressure to be repatriated and did not tend to perform very well. 22 It is difficult not to agree with Giulia Barrera's comment that: By using Eritreans, and not Italians [...] the Italian government was able to continue pursuing an expansionist colonial policy in Somalia and Libya without running the risk of the political backlash that the death of Italian soldiers could have caused.23 Another key reason why the Italians preferred to use colonial troops was financial: Eritrean and Libyan troops were cheaper to maintain than Italian soldiers. Colonial soldiers were recruited on a voluntary basis for a two-year term. On completing their initial period of service soldiers had the option to sign on for an extra two years. According to the 1926 military regulations for Libya, an Italian private was entitled to a basic daily pay of 2.25 Lire, supplemented by a further 3.50 Lire per day for colonial service. Meanwhile an ascari, who was at the bottom of the colonial military hierarchy, was entitled to a basic daily pay of 1.50 Lire during his first twoyear term (with the sum increasing to 1.75 during the second term and 2.00 during the third). Ascaris who were posted outside their colony received a further allowance of 1 Lira per day.24 In other words, a colonial soldier cost the Italian treasury less 21 Luigi Goglia, 'Popolazioni, eserciti africani e truppe indigene nella dottrina italiana della guerra coloniale', Mondo Contemporaneo, Vol. 2, (2006), p.18. 22 ACS, FB, s. 4, s. 6, ‘Comitato interministeriale per la questioni militari coloniali’, 4 March 1926. 23 Giulia Barrera, ‘Racial hierarchies in colonial Eritrea’, in Patrizia Palumbo (ed.), A Place in the Sun: Africa in Italian Colonial Culture from Post-Unification to the Present, (Berkeley: University of California Press, 2003), pp.84-85. 24 ASMAI, p.115/3, f. 39-40-41, ‘Copie del nuovo ordinamento militare per la Libia’, 16 November 1926. A similar cost ratio between colonial and metropolitan soldiers existed in the British army in 1949. See: French, The British Way, p.36. www.bjmh.org.uk 53 COLONIAL SOLDIERS IN LIBYA 1922-32 than half the sum of an Italian soldier. A further financial advantage was that colonial troops were allocated fewer rations.25 In addition to political expediency and pay, colonial soldiers offered the advantage, in Italian commanders' eyes, of possessing physical and mental attributes relevant to prolonged campaigning in Libya. While not as elaborate as the British 'martial race' theory, early twentieth-century Italian officers and military scholars also developed a set of essentialist beliefs about the levels of effectiveness and virility of the various groups of colonial soldiers they employed. In 1914 Captain Corselli believed that colonial troops were better suited for colonial warfare than 'white' troops because they were more resistant to the weather and could better adapt to the terrain. He held 'Eritrean' troops – a catchphrase for East Africans more broadly – in high regard. His comments about the Arabs were disparaging, describing them as disloyal and deceitful. He nevertheless argued that, with proper discipline, they could make fine soldiers. 26 Writing in 1927, Colonel Guglielmo Nasi, who later went on to become a prominent figure in Italian East Africa, also believed that only Eritrean and Libyan troops should be used in colonial counter-insurgency campaign. He considered that units requiring a high level of mobility, 'white' troops were like 'a ball and chain' on the feet of colonial commanders.27 One of the greatest exponents of the Italian military's typology of colonial groups was General Ottorino Mezzetti who spent a long time in command of troops in Tripolitania before being transferred to Cyrenaica in 1927. Mezzetti believed that 'coloured' troops ‘have the best developed warrior instinct’. Left to their own devices, they knew well enough how to ‘make use of the terrain and fight’. Indeed, even under the most adverse circumstances, ascaris were able to draw on their 'warlike virtues' (virtù guerriere).28 Like Nasi, Mezzetti believed that in combat the 'national' (i.e. Italian) contingent was the most vulnerable and delicate part of the column. They struggled to fight without orders and their behaviour affected the 'coloured' troops. Eritrean troops formed the ‘backbone and the ribs of the columns'. He praised their abilities as marchers, especially in mountainous terrain, commenting that with them and with the Libyans it was possible to cover 50 km in a day. The arrival of an Eritrean reserve, even a small one, could often decide the fate of a battle. Excellent warriors, 'the Eritreans feel excited and develop in the fight the bestial instincts of a warrior race (razza guerriera)'.29 25 26 27 28 29 Gooch, 'Re-conquest and Suppression', p.1008. Goglia, 'Popolazioni, eserciti africani', pp.12-18. Goglia, 'Popolazioni, eserciti africani', p.26. Ottorino Mezzetti, Guerra in Libia: esperienze e ricordi (Rome: Paolo Cremonese, 1933), p.3, p.5. Mezzetti, Guerra in Libia, p.42. 54 British Journal for Military History, Volume 1, Issue 2, February 2015 Eritrean troops, however, suffered from some deficiencies and there were countervailing problems created by their employment in battle. In long marches, he argued, they tended to stray and lose sight of the column. Bearing in mind their tradition of acting independently, they found it difficult to march with other contingents. ‘Even today an excessively aggressive spirit pervades among the Eritreans, preventing them from drawing from fire all the advantages it offers’. If they broke through enemy lines a chase ensued and contact with their column and its commanders was often lost. 30 Mezzetti also advised Italian commanders not to employ Eritrean ascaris in tasks such as breaking rocks for paving roads as such work might offend their warrior pride.31 This type of essentialist thinking was also present in Badoglio's instructions. For instance, in May 1922 he noted that Eritrean troops had always proved themselves to be of the highest quality. Meanwhile Libyans had great potential, especially the Berbers who displayed courage and aggressiveness, characteristics which were always needed in colonial warfare. 32 By the summer of 1930, Badoglio had become convinced that Eritrean troops were the key to crushing the revolt in Cyrenaica, It is absolutely vital to abandon the Arab system of shooting from a distance and regarding it as a success if the enemy is forced to withdraw. This system, capable only of prolonging hostilities indefinitely, is also contrary to the fighting spirit and aggressive instincts of our brave Eritrean ascari.33 Italian military leaders clearly grew accustomed to relying on troops raised in Eritrea. However, the East African colony was not always in a position to furnish all the manpower that was needed. The Eritrean Predicament The Italians were not the first imperial power to raise troops in one colony and then deploy them in another. John A. Hobson, a harsh critic of imperialism, ridiculed this practice in a British context long before the revolts in Tripolitania and Cyrenaica, Why should Englishmen fight the defensive or offensive wars of this Empire when cheaper, more numerous, and better-assimilated fighting material can be raised upon the spot, or transferred from one tropical dominion to another? As the labour of industrial development of tropical resources is put upon the ‘lower races’ who reside there, under white superintendence, why should not militarism be organized upon the same basis, black or brown or yellow men, 30 31 32 33 Mezzetti, Guerra in Libia, pp.42-43. Mezzetti, Guerra in Libia, p.4. Saini Fasanotti, Libia, p.119. Cited in: Rochat, 'The repression', p.80. www.bjmh.org.uk 55 COLONIAL SOLDIERS IN LIBYA 1922-32 to whom military discipline will be ‘a wholesome education’, fighting for the British Empire under British officers?34 As we have seen, Italian commanders thought very highly of Eritrean troops and these were often in demand. From the early months of 1912 until 1934 Italy had used no less than 68 Eritrean battalions and six artillery batteries in rotation for service in Libya.35 However, the over-reliance on Eritrean troops had a negative impact on the colony’s economy. It also introduced an element of tension into relations between colonial authorities in Rome and the administration in Asmara, the colony's capital.36 Eritrea’s population, which was estimated at roughly 510,000 in the second half of the 1920s,37 was insufficient to furnish all the troops Italian colonial and military leaders wanted. Complaints from Italian officials in Eritrea about the strain recruitment placed on the colony were made even before the ‘re-conquest’ of Libya began. In response to urgent requests from Rome for more troops, Governor Giovanni Cerrina Feroni highlighted the ‘grave repercussions’ the subtraction of working hands would have on Eritrea’s economy and agriculture.38 He cautioned that Eritrea must not be reduced into a ‘seedbed’ of mercenary ascaris for the other colonies. With time, he argued, their excellent characteristics and the enthusiasm which distinguished them would be lost.39 The relative calm in Libya in 1920-21 enabled the Italians to send back a number of battalions to Eritrea without replacing them. Instead, colonial officials in Eritrea were asked to furnish men for ‘the establishment and maintaining in efficiency of LibyanEritrean battalions’.40 Officials in Eritrea did not want to undermine the morale and unit cohesion of their ‘organic’ Eritrean battalions. Moreover, recruitment in Eritrea could not sustain more than 12 battalions at any given time. Therefore, the troops that were sent to the so called ‘Libyan-Eritrean battalions’, which were soon much more aptly renamed ‘mixed’ battalions, were raised beyond Eritrea’s borders. The vast majority were to come from Ethiopia (primarily Amharans), but soldiers were also recruited in Sudan and Yemen.41 34 John A. Hobson, Imperialism: a study (Cambridge: Cambridge University Press, 2010 [1902]), p.143. For more examples of colonial recruitment in British colonies in Africa, see: E. Whitehead, ‘A short history of Uganda military units formed during World War II’, Uganda Journal, Vol. 14, Iss. 1 (1950), pp.1-14; Timothy Parsons, The African Rank-And-File: Social Implications of Colonial Military Service in the King’s African Rifles, 1902-1964, (Oxford: Heinemann, 1999). 35 Alessandro Volterra, Sudditi coloniali: ascari eritrei 1935-1941, (Rome: Franco Angeli, 2005), p.43. 36 Uoldelul Chelati Dirar, ‘From warriors to urban dwellers: Ascari and the military factor in the urban development of colonial Eritrea’, Cahiers d'études africaines, 175 (2004), pp.10-11. 37 ASMAI, p.35/4, f.12/13. 38 ACS, MAI, AS, b.1, Cerrina Feroni to Colonies Minister, 25 August 1920. 39 ACS, MAI, AS, b.1, Cerrina Feroni to Colonies Minister, 6 May 1921. 40 ACS, MAI, AS, b.1, annual report for 1920 of the RCTC of Eritrea, 1 May 1921. 41 Volterra, Sudditi coloniali, p.45, p.49; Sullivan, ‘A thirst for glory’, pp. 219-220. 56 British Journal for Military History, Volume 1, Issue 2, February 2015 Recruiting men from Ethiopia was not without its problems. In November 1921, Cerrina Feroni explained to officials at the ministry for the colonies that Ethiopian recruits for the mixed Eritrean-Libyan battalions were scarce because of a veto imposed by Ethiopian chiefs. He added that soldiers recruited from beyond Eritrea’s borders must not be trained in the use of artillery as they might be induced by their chiefs to desert and could use their expertise in the service of another country.42 Furthermore, the strong demand for working hands in Eritrea meant that Ethiopians preferred to enlist as labourers in the colony rather than as ascaris in the colonial army.43 Despite these reservations, Eritrea was able to supply the men requested once counter-insurgency operations in Libya began. In 1924-25, for instance, five ‘organic’ battalions and an artillery battery, consisting of 4,301 men, were raised in the colony and dispatched to Libya. For the ‘mixed’ battalions 3,815 men were sent of whom 3,540, more than 92 percent, were recruited beyond Eritrea’s borders.44 In Rome, military commanders and officials at the ministry for the colonies felt uneasy with the reliance on recruits from Ethiopia. Surveying the situation in an inter-ministerial committee for colonial military questions in 1926, Badoglio explained that in Eritrea, which ‘had the function of furnishing good coloured troops’, a grand programme to revitalise the local economy was obstructing recruitment. Somalia could not furnish anything of value. With regard to Yemen, where the Italians had been recruiting for a few years, the Minister for the Colonies Scalea explained that the local leader, Imam Yahya, could not spare any men because these were needed in case hostilities broke out with Ibn Saud. The committee noted that the Ethiopian regent, Ras Tafari (soon to become Emperor Haile Selassie), was willing to turn a blind eye to Italian recruitment for the time being. However, at the same time he also sought to centralise the powers of the state and create a national army. Hence, the ability to recruit in Ethiopia could dry up completely if tensions between Italy and Ethiopia were to rise. While various alternative options were considered, including recruitment in Albania and the Congo, no viable solution was found.45 The practice of enlisting Ethiopians continued without hindrance 46 According to Brian Sullivan, colonial officials in Rome implored Mussolini to avoid any actions that would undermine relations with Ras Tafari until Libya had been pacified. They did not want the regent to cut off the flow of Ethiopian recruits.47 42 ASMAI, p.35/4, f.12, ‘Conferenza con S.E. Cerrina’, 27-29 November 1920. ACS, MAI, AS, b.1, annual report for 1920 of the RCTC of Eritrea, 1 May 1921. 44 ACS, MAI, AS, b.1, annual report for 1924/25. 45 ACS, FB, sc.4, f.6, ‘Comitato interministeriale per la questioni militari coloniali’, 4 March 1926 and 20 May 1926. 46 ACS, MAI, AS, b.1, R.C.T.C dell’Eritrea, reports for the 1st and 3rd trimesters of 1927. 47 Sullivan, 'A thirst for glory', p.219, p.224. 43 www.bjmh.org.uk 57 COLONIAL SOLDIERS IN LIBYA 1922-32 The tone of Italian colonial officials in Eritrea began to change in September 1927, when the colony was confronted with an invasion of locusts which destroyed crops and severely destabilised the local economy. As a result of the locusts crisis there was a sharp increase in the number of Eritreans willing to enlist.48 While the reports of senior officers in Eritrea noted the dire economic state of the colony, afflicted by further locust invasions and a drought in 1928, recruitment patterns did not change significantly at first. The flow of recruits from beyond the colony’s borders continued and was described as ‘sufficient’. The large number of Eritreans wishing to escape economic hardship enabled the Italians to use more rigorous selection criteria when deciding who to recruit and it is likely that the proportion of Eritreans in the ‘mixed’ battalions rose, though the exact percentage is unknown.49 In 1929, with the lull in operations in Cyrenaica, the very low number of requests for recruits that Eritrea had received created discontent among the local population that had grown accustomed to seeing recruitment as a way of alleviating economic difficulties. In the first trimester of that year, only 77 men were sent to Libya while 2,570 returned, having completed their two-year service.50 Gone were the days of the early 1920s when colonial officials in Asmara pleaded for recruitment to be stopped. An official change in policy finally emerged in the second half of 1929. Article 66 of the official military regulations for Eritrea, which were drafted in the autumn of that year and signed by the King of Italy, stipulated that preference in recruitment was to be given to those born in the territory of Eritrea.51 The commander of the colonial army in Eritrea reported that this preference was given because of the plague of locusts, and also in adherence to the directives of the inter-ministerial committee of 1926. The colonial army wanted to reduce the number of external recruits who received military training and at the same time to enlarge the numbers receiving military training among the colony’s population. However, the numbers sent to Libya during the third trimester still fell far short of the number returning.52 From the point of view of officials in Asmara, the situation began to improve in the first trimester of 1930, owing to the increased demand for Eritrean troops in Cyrenaica. Following the re-organisation of General Graziani’s forces, to which we shall return below, the number of men who were sent to Libya increased sharply. The report for the third trimester of 1930 spoke of an ‘intense rhythm’ of recruitment and during the fourth 48 Volterra, Sudditi coloniali, pp.54-55; ACS, MAI, AS, b.1, R.C.T.C dell’Eritrea, report for the 4th trimester of 1927. 49 Volterra, Sudditi coloniali, p.55. 50 ASMAI, p.35/4, f. 12, R.C.T.C dell’Eritrea, report for the 1st trimester of 1928; ACS, MAI, AS, b.1, reports for the 3rd and 4th trimesters of 1928 and the 1st and 2nd trimesters of 1929. 51 ACS, MAI, AS, b.1, f.1, 'Ordinamento militare dell’Eritrea', 20 November 1929. 52 ACS, MAI, AS, b.1, R.C.T.C dell’Eritrea, report for the 3rd trimester of 1929. No reason was given for the retrospective implementation of the inter-ministerial decision from 1926. 58 British Journal for Military History, Volume 1, Issue 2, February 2015 trimester the number of those being sent (1,135) far exceeded the number returning (369).53 Service during the counter-insurgency in Libya left an impression on Eritrean popular culture and memory. The Eritrean literary figure and clergyman Gebreyesus Hailu wrote a short novel in Tigrinya in 1927, based on the story of an Eritrean soldier who had served in Libya. First published many years later, in 1950, the novel has recently been translated into English and given the title The Conscript, a misnomer as Eritrean recruits in general and Tuquabo, the novel's protagonist, in particular volunteered for service rather than being conscripted. Hailu records two types of responses to the departure of men for Libya. On the one hand there was youthful enthusiasm: This was a time when there was war going on in Tripoli, and it was deemed fitting for the people of Habesha to be willing to spill their blood in this war. The youth were singing, ‘He is a woman who refuses to go to Libya,’ and small children in return sang, ‘Come back to us later, Tribuli... give us time to grow up’54 Conversely, there was also dismay at the immensity of the sacrifice: 'On the other hand, people back home were thinking about them and cried, "Our priests, why don't you speak out? Not even one young man can be found; all have gone to Tripoli"'.55 Libyan Troops and the Question of Loyalty As Ahmida has pointed out, ‘in postcolonial Libyan historiography, collaboration with the Italian colonial state is poorly studied, in part because post-independence Libyan nationalism drew its stock of heroes, martyrs and legends from the anti-colonial resistance’. Those who 'went Italian' were considered to have been lacking in moral character.56 Yet service in the Italian colonial army was certainly not restricted to the margins of society. In just over 30 years of colonial rule tens of thousands of Libyans passed through the military, fulfilling a variety of roles. The Italians began to recruit Libyans to serve in irregular forces as early as February 1912. Later that year the first regular Libyan battalion was raised. 57 Colonial authorities continually debated the trustworthiness of irregular bands. In November 1916, for instance, the Minister for the Colonies, Gaspare Colosimo, implored 53 ACS, MAI, AS, b.1, R.C.T.C dell’Eritrea, report for the 3rd and 4th trimesters of 1930. Gebreyesus Hailu, The Conscript, (Athens: Ohio University Press, 2013), p. 7. 55 Hailu, The Conscript, p.26. 56 Ahmida, Forgotten Voices, pp.26-27. 57 Comando del corpo di Stato Maggiore, Annuario ufficiale delle forze armate del Regno d’Italia anno 1938XVI, (Rome: SME, 1938), p.618. 54 www.bjmh.org.uk 59 COLONIAL SOLDIERS IN LIBYA 1922-32 General Ameglio, the Governor of Tripolitania, not to create and arm an irregular band of 2,100 men. Such a force, he argued, could ‘contain the germ of a betrayal’ that would provide ‘the rebels with two thousand and one hundred 70/87 rifles with which to fight us.’58 However, at various stages of the counter-insurgency, the Italians found these highly mobile and well-adapted irregular troops indispensable. More than a thousand irregulars were used in Cyrenaica each year between 1925 and 1928, while in Tripolitania their number rose from just over a thousand a year in 19241927 to 2,476 in late 1928.59 As for regular Libyan troops, Italian commanders gradually began to appreciate their effectiveness. General Graziani, who was always well aware of the importance of propaganda, praised his locally raised forces after leading the repression of resistance in Tripolitania in the 1920s: ‘In all the period of the conquest and re-conquest of Libya, the Libyan battalions, alongside the Eritrean ones, fought with valour worthy of emulation and admiration’. Italian officers, he added, could feel ‘secure to find in them loyalty without limits, valour, heroism, a warlike spirit of exceptional nature and performance’.60 After a battle at Bir Tarsin on 26 May 1925, Graziani praised the loyalty and valour of the Libyan troops, whose heroism was ‘equal to that of the Eritreans’. He noted with satisfaction that not a single soldier deserted. ‘Our indigenous followers feel today not simply as a mercenary tool in our hands, but an integral and conscious part of the great mother Italy’. When called upon, they rush forward ‘willingly, ready to sacrifice’.61 The notion that the quality of Libyan troops might not fall short of the high standards of the Eritreans was shared by General Mezzetti, who successfully led Tripolitanian troops: ‘Libyan units, today, can compete in performance with Eritrean units’. Compared to the latter, he argued, the Libyans were much more resilient in enduring long marches, excessive heat and thirst. While ‘less aggressive than the Eritreans, in battle they are equally tenacious’. Despite deficiencies in the way they were led, ‘the regular Libyan units maintain [their] loyalty and trustworthiness in every trial’.62 However, in Cyrenaica the satisfaction of Italian commanders with their locally-raised troops was much more circumspect. The British anthropologist E. E. Evans-Pritchard, who conducted research on the Sanusi of Cyrenaica after the Second World War, pointed out that the Bedouins recruited by the Italians 'did not hesitate to assist the Patriots when an opportunity offered itself'. They passed to the rebels both rifles and even pay. Mock battles between the rebels and Italian-sponsored irregulars were used to hand over weapons and ammunition. The submitted tribes were a good 58 59 60 61 62 ASMAI, p.115/3, f.36-37-38, Colosimo to Ameglio, 12 November 1916. ACS, MAI, AS, b.1, f.1/2. Libyans also served as guides, spies and in various other roles. Rodolfo Graziani, Verso il Fezzan, (Tripoli: F. Cacopardo, 1929), p.223. Graziani, Verso il Fezzan, p.366. Mezzetti, Guerra in Libia, p.43. 60 British Journal for Military History, Volume 1, Issue 2, February 2015 source of intelligence for the rebels. Armed sottomessi would sometimes be found among rebel casualties following skirmishes with the Italians.63 After being transferred to Cyrenaica in 1927, General Mezzetti observed that following a hard-won battle, which ‘relieved the rebels of 50 soldiers and 200 camels, the next day the sottomessi will refurnish the rebels with another 50 soldiers and 200 camels’.64 Evans-Pritchard pointed out that 'in the end the Italians came to the conclusion that they could trust no Cyrenaican, least of all a Cyrenaican Bedouin'.65 The dilemmas created by recruiting local troops came to a head in the summer of 1930. In a series of messages to Colonial Minister Emilio De Bono and to Graziani in Benghazi, Badoglio explained that operations in Cyrenaica were put in jeopardy because the population and deserters collaborated with the rebels. In fact, the Italian army recruited from the same pool of men as the rebels did. Omar al-Mukhtar, he explained, ‘is extremely able both as commander and as organiser. A perfect information service enables him to refuse battle when the situation does not definitively favour him'. The Governor of Tripolitania and Cyrenaica wanted to use 'mobile nuclei of troops, notable for their great offensive spirit and guaranteed loyalty, and thus composed of Eritreans’.66 Therefore, parallel to the removal of the civilian population from Jabal al-Akhdar to a series of concentration camps, Badoglio initiated a total reorganisation of the Italian colonial army in Cyrenaica. His goal was to deny the rebels information and ammunition. To reduce their dependency on ‘troops of little efficiency and doubtful loyalty’, the Italians dissolved the 10th and 12th Libyan battalions and also reduced the number of savari (cavalry troops) under arms in June 1930. 200 select men with proven combat experience were maintained and used to form a garrison company. During the same month the 2nd Eritrean battalion was called up and in July the 16th Eritrean battalion was reconstituted. In December 1930 a new Eritrean battalion, the 21st, was established, bringing the total number of Eritrean battalions in Cyrenaica from five in March to eight.67 In his memoirs from the pacification of Cyrenaica, Graziani pointed out that ‘this draconian measure, which struck the hearts of brave commanders and officers, damaged my own popularity and authority’. Officers pleaded with him to reconsider, arguing that desertion from their Libyan units was negligible. 68 He omitted to mention that he tried, more than once, to convince Badoglio to allow him 63 Evans-Pritchard, The Sanusi, pp.162-163. Rochat, 'The repression’, p.46. 65 Evans-Pritchard, The Sanusi, p.163. 66 ACS, FG, s. 8, f. 12/1, Badoglio to Graziani and De Bono, 20 June 1930; Badoglio to De Bono, 1 July 1930; Rochat, 'The repression', p. 74-75. 67 ACS, FG, s.9, f.12, ‘Il R. Corpo delle Truppe Coloniali: sue trasformazioni per aumentare la efficienza’ [no date]. 68 Rodolfo Graziani, Cirenaica pacificata, (Milan: Mondadori, 1932), p.81. 64 www.bjmh.org.uk 61 COLONIAL SOLDIERS IN LIBYA 1922-32 to use loyal troops from Tripolitania instead of Cyrenaican units. However, Badoglio was adamant that Graziani must make do with Eritrean troops.69 It was this reorganisation of the Italian force in Cyrenaica which accounted for the ‘intense rhythm’ of recruitment in Eritrea in the second half of 1930. The irregular bands, which had often given direct or indirect assistance to the rebellion, were also cut in size. Their numbers were reduced to 900 in September 1930, shrinking to 560 in early 1931. However, the Italians could not do without this force altogether. Therefore, to prevent leakage of ammunition to the rebels, all the irregulars had their rifles replaced. Instead of Mannlicher Carcano model 1891 – a rifle also used by the rebels – Italian commanders gave their irregular troops VetterliVitali model 1870/87 that had a different calibre, even though the latter was, in fact, inferior. Some irregulars tried to sabotage their 70/87 in the hope of being reissued mod. 91 rifles, but to no avail.70 In his memoirs Graziani noted with satisfaction that, after his capture, Omar al-Mukhtar admitted that these actions denied him sources of supply.71 But even after most of the Cyrenaican troops had been removed from frontline service, the Italians still relied heavily on Libyan manpower. The construction of the barrier along the Libyan-Egyptian border during the final months of the uprising was carried out by no less than 2,500 Libyans.72 A War of Religion? There is a further factor at work which is far more difficult to quantify or determine: relations between Libyans and Eritreans. Not enough is known about how Italy's colonial soldiers viewed their service during the revolt in Libya. Historians have so far not been able to draw on first hand written accounts. Oral history interviews have shed light on the experiences of Libyan victims of Italian repression and of Eritrean soldiers who served in the Italian colonial army in the second half of the 1930s and early 1940s. 73 When it comes to colonial troops who took part in counter-insurgency operations in Libya the subaltern really does not have much of a voice. However, there are a number of second-hand accounts which suggest that the employment of East Africans in Libya fostered mutual hatred between the two subaltern groups. One of the harshest contemporary critics of Italy’s policy in Libya was the Danish traveller and journalist Knud Holmboe. A Muslim convert who had learnt Arabic in Morocco, Holmboe embarked on a daring voyage across North Africa in a car in 69 ACS, FG, s.9, f.12, Graziani to Badoglio, 18 May 1930; Graziani to Badoglio, 2 September 1930; Badoglio to Graziani, 3 September 1930. 70 ACS, FG, s.9, f.13, 'Per lo stroncamento della ribellione', [no date]. 71 Graziani, Cirenaica pacificata, p.81. 72 ACS, FG, s.9, f. ‘Dieci anni di storia cirenaica’, p.223. 73 Ahmida Forgotten Voices; Volterra, Sudditi coloniali. 62 British Journal for Military History, Volume 1, Issue 2, February 2015 1930. His impressions from the journey were published in a memoir, Desert Encounters, which was translated into English but was banned in Italy. 74 While travelling through Tripolitania he noted that 'lots of Arab soldiers have been murdered in the most foul manner [by the rebels]. And they both spoke Arabic and had the same religion as the Bedouins.'75 However, once he arrived in Cyrenaica, he increasingly commented on the Italian reliance on Eritrean troops, In Arghela [probably Agaila] the difference between Tripoli and Cyrenaica made itself felt at once. Here there was not a single Arab soldier; they were all natives from the Italian colony of Eritrea. ‘Why don't you use Arab troops?’ I asked Colonel de Ronco during the evening. ‘You can't trust them. They fraternize with the enemy.’ ‘But are the Eritreans any better?’ ‘They are more reliable. They hate the Arabs, who are Mohammedans, for they themselves are Christians. The Arabs here in Cyrenaica are exceedingly poor soldiers, while we can trust our brave troops from Eritrea implicitly.’ ‘Then the Arabs round here don't want to be soldiers?’ ‘No. It's not the same as in Tripoli. There is an unquenchable hatred between the Eritreans and the Bedouins. For instance, some years ago some of the Eritrean troops here in Arghela caught three Bedouins. They burnt them.’76 When he asked Lieutenant-Colonel Piatti, stationed at Maraua, why there were so many troops from Eritrea in Cyrenaica, the latter said: Because they are the only troops we can depend on. The Arabs are lazy and unreliable, and run away at the slightest provocation, while the Eritreans on the contrary are absolutely loyal to us Italians. I myself have spent considerable time in Eritrea, and there we never had any trouble or rebellion.77 Holmboe came to the conclusion that the Italians were fomenting religious hatred between the Eritreans and the Libyans. He reports hearing an address by Graziani to his colonial troops, 'New battles lie before you, but I know you will be victorious together with us Italians, who profess the same religion as you. Let us cry "Evvia" for Italy and her colony of Eritrea'.78 As he reached eastern Cyrenaica Holmboe was arrested by the Italian authorities. While in prison in Derna he asked the Libyan guards why they did this job, to which they replied, ‘Don't you think it is better to 74 75 76 77 78 https://www.knud-holmboe.com/biography.html [accessed: 4 December 2014]. Knud Holmboe, Desert Encounters (London: George G. Harrap, 1936), p. 85. Holmboe, Desert Encounters, pp.147-148. Holmboe, Desert Encounters, p.212. Holmboe, Desert Encounters, p.191. www.bjmh.org.uk 63 COLONIAL SOLDIERS IN LIBYA 1922-32 have Moslems to look after the prisoners than Eritreans? When no Italian officers are about we can see that the prisoners have an easier time.’79 Finally, after his expulsion from Libya, Holmboe went to visit the exiled Sanusi leader and future King of Libya, Sayyid Idris, in Alexandria. The latter asked him ‘Do you know why the Italians are using Eritrean troops? […] Because the Eritreans are Christians. The Italians have created a religious war in Cyrenaica’.80 Very few sources can corroborate Holmboe’s assertion of an Italian-induced war of religion. There is some evidence that, in Graziani’s propaganda of 1930, ‘the Askaris were reminded of their Christian duty to kill Moslems’.81 However, Graziani always tried to tailor his messages according to the composition of his audience. For instance, in a message to local notables and Islamic religious officials in Cyrenaica in spring 1930 he underlined the ‘absolute necessity to respect indigenous customs, religion and women.’82 Evans-Pritchard highlighted the importance of religion as a central cause which sustained the revolt, It then became simply a war of Muslims to defend their faith against a Christian Power. […] Without due appreciation of the religious feelings involved in the resistance it would, I think, be impossible to understand how it went on for so long against such overwhelming odds or how the tribes, with a long history of feuds dividing them, co-operated as much as they did under Sanusiya guidance.83 However, he did not comment on how the East Africans fitted into this religious confrontation. It is very important to note that not all of Italy’s East African troops were Christian. A quick look at lists of deceased ascaris from Eritrea reveals several Muslim names.84 In fact, in the East African context, Muslim troops were deemed more loyal than Christians. In 1920 only Muslim soldiers could be recruited to artillery units, which received specialist training.85 In 1923 the ‘mixed’ battalion sent from Eritrea to Libya consisted of 1,179 Muslims and 5,059 Christians, though it is possible that this ratio changed by 1930.86 The East African soldiers in Hailu’s The Conscript are Christians. Tuquabo's relations with his Muslim neighbours in Eritrea are very cordial and there is no sign of religious strife when the novel begins. However, the tone changes once Tuquabo arrives in 79 80 81 82 83 84 85 86 Holmboe, Desert Encounters, p.258. Holmboe, Desert Encounters, p.276. Sullivan, ‘A thirst for glory’, p.247. Graziani, Cirenaica pacificata, p.74. Evans-Pritchard, The Sanusi of Cyrenaica, p.166. ASMAI, Archivio Eritrea, p.657, f. ‘le eredità di eritrèi deceduti e pratiche consimili' ACS, MAI, AS, b.1, f.1, Feroni to Colonies Minister, 25 August 1920. Volterra, Sudditi coloniali, p.49. 64 British Journal for Military History, Volume 1, Issue 2, February 2015 Libya where Italian-inspired negative stereotypes held by East African soldiers regarding the Arabs begin to appear: But for the Tigrinya the Arabs were notorious for bad behaviour. They were untrustworthy, treacherous, and they held grudges. They had the reputation for being merciless killers if they got the occasion. According to the stereotype that was passed along by the Italians, to say that an Arab would respect a deal would be to lie.87 Hailu's account frames the East African-Libyan interaction in terms of historical enmity and racial prejudices rather than a war of religion. On the Libyan side, the poet Um al-Khair described the Eritrean guards at the concentration camp as ‘kinsmen of the devil’.88 Moreover, the association of black Africans with slavery was still prevalent. Considering the brutal nature of the conflict – neither the rebels nor the Italians and their colonial soldiers took prisoners, except for interrogation and/or torture – it is unsurprising that mutual hatred evolved. The colonial involvement in the counter-insurgency in Libya left behind a bitter legacy. Hailu warned that, by becoming ‘instruments to occupy someone else’s country’, the East Africans were committing a grave error that may come back to haunt them, There may be some who think that fighting the Arabs on behalf of the Italians and exterminating them from the face of the earth was forgivable considering that the Arabs and black Africans were historically enemies. But what was being done would one day lead to one’s fall. If one day they come led by a Frenchman or an Italian to fight, didn’t the Habesha know that the Arabs were going to pay back with vengeance?89 Hailu’s warning was not entirely unfounded. At the end of 1931 Libyan children between the ages of 6 and 15, who had been orphaned during the revolt, were placed in special camps that were run on military lines. In 1934 Graziani observed that these individuals were best suited for: the Libyan units of the royal corps of colonial troops, into which the first recruits have already entered from Marmarica, with about a hundred boys from the Margun and Soluk camps due next July; when these reach the 87 88 89 Hailu, The Conscript, p.32. Ahmida, Forgotten Voices, p. 48 Hailu, The Conscript, p.29. www.bjmh.org.uk 65 COLONIAL SOLDIERS IN LIBYA 1922-32 maximum age limit (18 years), they will be enrolled in the VII and IX battalions of the new force.90 These recruits formed part of the Libyan force that was sent to fight for Italy in the war against Ethiopia in 1935-36. In fact, the Libyan division in Ethiopia developed a reputation for taking no prisoners.91 Conclusion To conclude, the counter-insurgency in Libya provides a striking example of the instrumental manner in which Italian authorities treated the troops they had raised in and around their colonies. It took colonial and military leaders time to learn how to adapt their manpower supply to highly varied local conditions, but eventually they succeeded. While it is difficult to accept at face value Graziani's assertions about the sacrificial willingness of his indigenous forces or their enthusiasm for Italy, it is clear that East African and Tripolitanian units were effective. Furthermore, throughout most of the counter-insurgency, East African and Libyan troops were able to serve together despite their cultural differences. In the final stages of pacification in 193031 the Italians' reliance on East African troops increased. Highly mobile, well trained and, by this stage, sufficiently numerous, their trustworthiness offered the Italians a solution to the leakage of information and ammunition to the rebels. That the East African troops despised the local population and vice versa did not hinder, and perhaps even helped, Badoglio who had come around to the idea of crushing the rebels rather than trying to win Libyan hearts and minds. By no means unique to the Italian Empire or to the interwar period, the role played by indigenous soldiers in counter-insurgency campaigns deserves more attention. 90 Cited in: Rochat, 'The repression', pp. 113-115. Angelo Del Boca, Gli italiani in Africa Orientale, vol. II: La conquista dell'Impero, (Bari: Laterza, 1979), pp.666680. 91 66