Bring up the bodies: international order, empire, and re-thinking the Great War (1914–1918) from below

What does international order look like when analysed from its margins? Such a question is the obvious consequence of efforts within International Relations (IR) to take empire, colonialism and hierarchy more seriously. This article addresses this question by examining one of IR’s most important touchstones – the Great War – through the experiences of peoples in southeast Africa. It argues that to do this, we should use the methodological approaches of histories ‘from below’ and contrapuntal analysis. When looking at the Great War from the vantage point of southeast Africa (contemporary Mozambique), the key patterns of interaction organising the international look different to those emphasised in traditional accounts of international order and hierarchy. Notable features are the significant continuities and intersections between structures of war and colonialism, the racialisation of death and suffering, the effects of white imperial prestige as a strategic preoccupation and the deep historical roots of anti-colonial resistance. Reading upwards and contrapuntally from these histories, the paper argues for a redescription of international order as reflecting not predominantly a balance of power or a normative framework for the organisation of authority, but a dynamic matrix of structural violence. Reading order from below in this way helps us better capture how the international is implicated in the production and reproduction of everyday life for many people, as well as in more dramatic political transformations such as those generated by experiences of war and resistance to colonialism.


Introduction
For many in the field of International Relations (IR), the Great War (1914-18) remains a foundational reference point that defines both its origins and its purposes. In terms of origins, as is widely taught, it was in memory of those fallen in the War that the Chair and then first Department of International Politics was endowed at the University of Wales in Aberystwyth in 1919. Relatedly, the purpose of the field became, at least as the liberal tradition has located it, to manage world affairs between Great Powers both peacefully and progressively. This connects to an intellectual mission defined by the investigation of the nature of international order and the causes of change within it. In this regard, the War has also become a central pedagogical object for discussion of what happens in the event of a breakdown of the balance of power and the failure of diplomacy (e.g. Nye, 2001). In many ways, then, the Great War is a critical episode for the field's self-understandings of its origins and purposes.
Yet, as revisionist accounts of the field have shown, this narrative has consistently mystified and obscured the centrality of imperial governance, race war and capitalism as central preoccupations of the field and public debate before and after 1918 (Ashworth, 2013;Barder, 2021;Long and Schmidt, 2005;Owens and Rietzler, 2021;Thakur and Vale, 2020;Vitalis, 2015). 1 Seen this way, the disciplinary field is about 'how to run the world from positions of strength', as EH Carr once put it (cited in Haslam, 2000: 252), in which an imperial peace enables the carving up of conquered and subordinated spaces in a more orderly fashion (Barkawi and Laffey, 1999;Du Bois, 1915). Opening up these questions is instigating, at least in some quarters, a much wider re-think of what international order consists of and who or what IR is for.
This revisionist turn also opens up questions about the disciplinary object itself: the Great War. As Barkawi has argued, despite the centrality of war to the field of IR, there has been a curious lack of theorising about what war actually is, and in particular its coconstitution with politics and society (Barkawi, 2011(Barkawi, , 2013. Wars themselves are often reduced to individual data points rather than dynamic processes which reorder spaces, peoples and polities. Barkawi's (2017) own seminal contributions reinterpret the Second World War and its significance through the dynamics, experiences and perspectives of the British Indian Army, upending received wisdom about how and why armies fight, and how imperial dynamics shape and are shaped by their military affairs. Given the centrality of the Great War to the field of IR (e.g. Lebow, 2014;Levy and Vasquez, 2014), reinterpreting it as a fundamentally imperial constellation (Gerwarth and Manela, 2014) also has the potential to contribute significantly to a re-thinking of how we understand international order itself, and particularly its supposedly peaceful character. This article takes the imperial context of the Great War as a starting point for the re-description of international order via the experiences of some of its colonised subjects.
How might we do this? I argue that IR's recent historical turn (captured in de Carvalho et al., 2021) has yet to exploit fully the methodological resources offered to it by both postcolonial and social histories. In this paper, I propose that the approaches dominant in IR's historical turn can be complemented by a 'contrapuntal' approach which further engages with 'histories from below'. The argument is that while the historical turn has certainly widened and de-provincialised the lenses through which to understand international order, it has not in general tried to look through the other end of the telescope, particularly at the political lives of colonised non-elites. 2 However, re-conceptualising international order from other standpoints offers significant potential for democratising and decolonising IR as a field (Barkawi, 2016;Harding, 2008;Laffey and Weldes, 2008;Rutazibwa, 2018), highlighting dynamics, relations and actors often overlooked by dominant theories.
In this article, I elaborate features of the Great War grounded in the experiences and histories of southeast Africa, where oft-forgotten official combat took place from August 1914 until after the Armistice. In doing so, I make two related interventions into IR. The first, sequentially speaking, is a methodological one about the utility of building analysis from below as a useful approach to rethinking the place of empire within the international. This is particularly important for the field as it contemplates postcolonial critiques about how we conceive and research what we study. The second intervention is a substantive one about the nature of international order, which can be understood holistically as constituted by a dynamic matrix of structural violence, rather than by the balance of power, regulatory institutions or normative questions which are elements within and shaped by it. Within this matrix we can see colonialism and war as importantly continuous with each other, as part of wider interlocking patterns of control, force, punishment and resistance that organise bodies, resources, ideas and territory. In this context, racism as a feature of international order must be understood not principally as a set of outdated norms or individualised ideological prejudices but a hierarchical political regime of disposability and indifference to suffering (Gilmore, 2007) that is central to both the conception of imperial governance and military strategy, and which has widespread consequences for the character of the order itself.
The article proceeds as follows: the first section examines the debates on international order and hierarchy, bringing these into dialogue with postcolonial critique in IR; the second makes the methodological case for engaging with histories from below before elaborating the approach taken in this article; the third elaborates key features of the Great War in southeast Africa; and the fourth section summarises the analytic implications of this account before concluding.

International order, hierarchies and empire
What is 'international order'? The tradition within the field of IR most explicitly devoted to its elaboration has been the 'English School', which has conceived of it as a pattern of interaction among states directed towards the end of upholding a set of values which underpin its own survival (Bull, 1977). 'Order' is in this sense contrasted with 'disorder' and is often held to contrast directly with 'justice' (Bull, 1977: chapter 4). This approach to order has identified a range of institutions that constitute international society and contribute to the maintenance of order among states, including for Bull (1977) the balance of power, diplomacy, the role of Great Powers, war and international law. 3 In highlighting both questions of the distribution of power between states, and of the regulatory and normative order through which their behaviour is organised, this conception of international order has been represented by its adherents as more holistic, pluralistic and useful than other traditions which focus on one over the other (Buzan, 2001). 4 For contemporary descendants of this intellectual tradition, the hierarchical and imperial nature of this order has garnered attention, especially the idea of the 'standard of civilisation'. This operated within the European legal and normative framework to define who was included in the society of states, and therefore entitled to its statuses and protections, and who was not (e.g. Buzan, 2014;Gong, 1984;Keene, 2002;Suzuki, 2005;Yao, 2019). 5 This line of inquiry has given rise to a new 'hierarchies literature', emerging at the intersection of constructivism, historical sociology and the English School, which shows the structure of international orders tends to be hierarchical rather than anarchical, as supposed by conventional IR theory, and examines this in different historical and contemporary contexts (e.g. Hobson and Sharman, 2005;Keene, 2013;Lake, 1996Lake, , 2011Mattern and Zarakol, 2016;Phillips and Sharman, 2015;Towns, 2010;Zarakol, 2017). Much of this literature seeks to interrogate the role of authoritative elite ideas (i.e. norms, rules, discourses) in enabling and regulating hierarchies between different polities (e.g. Phillips and Reus-Smit, 2020;Towns, 2010;Zarakol, 2011). This literature is rich and interesting, adding depth and complexity to the conventional framings of international order that rely on assumptions about anarchy and sovereign statehood, and applying sociological concepts to the structure of the international such as the notion of stigmatisation (Zarakol, 2011) or stratification (Keene, 2014).
However, I argue that the new hierarchies literature tends to adopt an unnecessarily narrow and somewhat conservative sociological imaginary of the international, inherited from the English School, insofar as (a) it is still largely centred on states or 'powers', (b) it focuses very heavily on elite discourses/behaviours and (c) places great emphasis on the normative character of these orders even where the norms do not in fact regulate or constrain more powerful actors. 6 By contrast, historical materialist and postcolonial accounts of the international, long concerned with the question of hierarchy within global social change, have cultivated a wider sociological imaginary which encompasses and often centres the 'masses' of empire -the enslaved and their descendants (Gilroy, 1993;James, 2001James, [1938), peasants (Guha, 1987;Wolf, 1997Wolf, [1982), the indentured and imprisoned (Anderson, 2009) and so on -and the various global structures and processes in which they participate. These challenge the tendencies in the new hierarchies literature to see the international as the inter-state, to limit the recognition of international structures and institutions to those which are normally participated in by elites, and to underrecognise both the constitutive violence of the order and the role of resistance from below in the dynamics of order-making and conflict (Barkawi and Laffey, 2006). While the new hierarchies literature in IR occasionally acknowledges and draws on postcolonial thought in some respects, it has not yet taken on some of its more profound ontological and methodological challenges concerning how we should read and study empire and hierarchy with respect to international order.
How might a postcolonial sensibility approach the question of international order and its hierarchical nature instead? Various answers have been given in IR, which form the platform for the approach I take here to re-thinking the Great War. Barkawi and Laffey (2006), for example, emphasise the need to recognise the asymmetric relations of mutual constitution between the weak and the strong and bring them into a 'common analytic frame'(p. 332). Significantly, this recognises their material entanglements and their violent policing as part of imperial order-making. Krishna (2001) similarly alerts us to the ways in which IR theory abstracts from and sanitises the violence of colonialism and empire, thus treating key conventional narrative elements in IR's story of order such as 'the Hundred Years' Peace' and international law as moments of peace and co-operation rather than periods of conquest, genocide and the extension of imperial control. This perspective does not then see international 'order' as the opposite of 'disorder' and indeed more closely connects it with forms of violence. 7 Jones (2008) further underscores the relationship of imperial and post-colonial order-making to the global political economy, highlighting relations of extraction, exploitation and dispossession in the stabilisation and de-stabilisation of polities. Together, these and other postcolonial readings of the international order work to produce a picture in which wide-ranging relations of hierarchy, violence and extraction are not meaningfully limited or contained by borders or designations of difference, but rather enabled and structured by them. In contrast to some of the emphasis within the new hierarchies literature, which focuses on what it means to join the 'club' of great powers, the postcolonial approach incorporates thinking about the broader social relations and material conditions that underpin the existence of a 'club' at all, and privileging understandings that begin from or re-centre those margins (Blaney and Inayatullah, 2008;Nayak and Selbin, 2010).
Moving back, then, to the Great War as one of IR's most important conventional reference points for thinking about international order, we can begin to see how its depiction as the end of the 'Hundred Years Peace' and the outcome of a shifting or an attempt to shift the balance of power may only be part of the story, occurring as it does at a high point of European and non-European 8 colonial power in the international system. How, then, methodologically, should it be approached? In the next section, I make the case for using contrapuntal reading and histories from below as techniques for re-thinking international order from the margins, to disclose a more holistic picture of its workings.

Reading empire through histories from below
The new hierarchies literature has emerged alongside a wider conversation in the field about the uses of history in IR, expressed in the publication of key texts and arguments and the establishment of a History Section of the International Studies Association. 9 Common among these approaches is (a) a sense of the ways in which the discipline relies, either explicitly or implicitly, on historical claims which are contestable; (b) the sense that a historical appreciation complicates standard categories, narratives and assumptions within the discipline; and (c) a commitment to an ongoing dialogue between 'theory' and 'history' in the study of the international (see especially Lawson, 2012 andde Carvalho et al., 2021).
With regard to the study of empire and hierarchy in IR, and particularly among literature that recognises the problem of Eurocentrism, we can discern at least three dominant historiographical strategies which are currently at work. The first of these conducts critical (re)readings of the imperial archive and examines the relationship between knowledge, power and practice. We can see this in the work of, for example, Bayly (2016) and Manchanda (2020) on imperial knowledge about Afghanistan, as well as work by Bell (2006Bell ( , 2016 on the nature of liberal imaginaries in the imperial period and Branch (2012) on the peripheral origins of the territorial state. Such a strategy seeks to de-mythologise (cf. Rutazibwa, 2016) the rationalities and rationalisations of imperial power.
A second strategy seeks to put the modern European system into a comparative perspective, pointing to other imperial and political formations which had significant impacts of their own on the world. Examples of this include work on the nature of the Chinese imperial system (Hui, 2004(Hui, , 2005, on the character and role of Eurasian and Central Asian empires (Neumann and Wigen, 2013;Zarakol, 2022), and on the nature of the Indian Ocean system (Phillips and Sharman, 2015). 10 This strategy helps to de-centre and 'provincialise Europe' (cf. Chakrabarty, 2000) by pointing to a plurality of systems and imperial polities in IR.
The third strategy builds from the idea of 'connected histories' (Subrahmanyam, 1997) emphasising the interconnected, entangled nature of global modernity, and the reliance of the 'rise' of imperial Europe and the West on its relationship to and incorporation of peoples and systems outside it. Key texts in this space include Bhambra's (2007) re-thinking of quintessentially 'modern' processes from a postcolonial standpoint, Anievas and Nişancıoğlu's (2015) text on the origins of capitalism which connects macrohistorical processes both outside and inside the West through the lens of uneven and combined development, and Buzan and Lawson's (2015) argument for re-locating the making of the international system in the global transformations of the 19th century. This strategy considerably widens the historical-geographic gaze and in doing so helps fundamentally rethink the objects of analysis, such as capitalism and statebuilding, traditionally given a solely or principally European origin story. In line with this third strategy, we can also note Said's (1994) contrapuntal approach to reading history, which explores the 'overlapping territories' and 'intertwined histories' which constitute imperialism and indeed produce 'culture' in different spaces. Bilgin (2016) endorses this as a useful method, ethos and metaphor for approaching the question of diversity in Global IR, insofar as it emphasises not the separateness of different cultures but in Said's words 'an urgent sense of the interdependence between things' (cited in Go, 2016: 111, see also Lowe, 2015).
While all of these strategies have helped expand the analytic and historical imagination of IR with respect to hierarchy and empire, they have not fully realised the ambitions of postcolonial critique with regard to a repositioning of the analytic gaze (e.g. Laffey and Weldes, 2008). At least the first two strategies noted here have remained focused on the locus of imperial power, albeit in a more expansive and critical manner. Regarding the third, while there is certainly more space to imagine the geographical expansiveness of the processes underway and narrate their impacts on ordinary people, it nonetheless analytically foregrounds imperial actors and processes happening at a global or transnational scale.
Yet, within Said's formulation of contrapuntal analysis, following counterpoint in Western classical music, there is another historiographical possibility -namely that of shifting between the two interconnecting lines of melody/history, and taking seriously the idea that each line only has 'provisional privilege' (cited in Chowdhry, 2007: 104). This opens up the question of what exactly we might look at when not looking at the metropolitan and elite histories of empires, which can be connected to the tradition of reading history 'from below', that is, getting to grips with the lives and realities of those whose traces on the historical record are few but who may more properly represent 'the people', or, in tracing out the effects of colonialism on the 'global majority'. 11 This is exemplified in CLR James' The Black Jacobins (2001 [1938]), which centred the enslaved and their agency in bringing about the Haitian Revolution. Relatedly, Chakrabarty (2000) articulated a distinction and intertwining between 'History One' and 'History Two' in which the former emplotted the development of capital and the latter the lifeworlds of those historical subjects who were not fully colonised by its logic. The work of Mitchell (1991Mitchell ( , 2002 has also been significant in understanding the colonial nature of capitalist expansion from the bottom up. If we are interested in having that more holistic, pluralistic assessment and analysis of international order, then a methodological strategy in which we bring 'histories from below' into counterpoint with imperial practice seems a useful starting point. We now return to the reading of the Great War -as noted, a central reference point in the field's conception of international order. In the field of IR, of course, wars are generally understood as discrete events, having a definitive beginning and end, and a key institution organising international society. Yet as Barkawi (2016) has argued, 'decolonising war' gives us an opportunity to consider war alongside other forms of force. To contribute to this project, I re-situate the historical sketch of the war in the context of southeast Africa, narrating its significance through the impact on the people living there, and the ways in which their agency and negotiation of the colonial condition in turn impacted events.
In this reading, it is necessary however to reflect on issues of sources and representation / interpretation. The problem of sources is of course particularly acute in a colonial context, where the majority of available written records are kept by and for colonial powers, and in which ordinary people leave only a trace when interacting with officials. 12 The historians of southeast Africa on whose work I depend (such as the Isaacmans, Newitt, Vail and White) work with established strategies for this, including a critical reading of the colonial archive, a tracing of the agency and presence of subaltern actors in that archive through, for example, agricultural reports or police records, a use of oral histories from African communities that speak to their collective experience and wider secondary literature. I supplement these and other historians' work with the autobiography of a Portuguese soldier, Cardoso Mirão, who served in southeast Africa for a report on the conditions and agency of the Africans they captured, plus some other primary and secondary sources. 13 These strategies are consistent with those established for doing histories from below, for example, by James (2001James ( [1938), EP Thompson, (2013[1963) and the Subaltern Studies collective (e.g. Guha, 1987). While they are necessarily imperfect and mediated they do offer some kinds of methodological accountability for the direction of travel and opportunities for substantive interpretation (Ortner, 1995, in response to Spivak, 1988.
What is offered is thus a kind of 'best guess' at reconstruction, based on a wider understanding of how humans work in general and under particular conditions such as enslavement, war, kidnap and so on. I have sought to apply these principles to interpreting the nature and significance of the historical events discussed in the sections below, while remaining cognisant of the provisional and partial nature of such interpretations. While this work is challenging, it is necessary to ensure that the gaze of historical IR can attempt to relocate itself from its conventional birds' eye approach to historical phenomena and think in counterpoint from their supposed margins. The next section aims to do this with the Great War in southeast Africa, organising the historical material into four key themes -conscription and forced labour; practices of porterage; imperial racial anxieties; and uprisings and resistance.

(Wartime) conscription and forced labour
Conscription -enforced military enrolment -was an acute wartime phenomenon in Europe during the Great War as it became clear that the volunteers for the trenches were drying up. However, it had been a chronic phenomenon in colonial Mozambique ('Portuguese East Africa'), where it was accentuated by the war from 1914. 'Conscripts' were not called conscripts, and colonialism was not called war. Accompanied by the whip and palmatoria 14 rather than the promise of glory, many captured and enslaved people were sold to become the armed enforcers of large colonial estates. The historic phenomenon of military conscription in this territory operated within a broader but fragile apparatus of colonial extraction and predation, in which forced labour -agrarian, construction and military -was a constitutive component.
The colonial system in much of the territory until the 19th century depended on an inherited landholding system known as the prazos (Newitt, 1969). These were private estates granted to initially Portuguese settlers by the Portuguese crown, which entitled the holder to utilise the resources and peoples inhabiting the lands as they broadly wished. This entailed in most cases attempting to force the peoples inhabiting the land (the 'tenants') to pay off what they owed in hut taxes or 'rent' with their labour on the prazo estates. Yet, the settlers were poorly equipped to manage their lands, barely able to stave off tropical disease and ill-supported by the Portuguese crown in terms of manpower. As a result, there was also internal demand for Africans to become the armed enforcers within huge colonial estates. While some did volunteer in return for a promise of no further re-sale by the owners, many of the private armies raised by the prazoholders consisted of men who had been captured and enslaved elsewhere, and sold to prazo-holders.
This kind of 'conscription' was of course not particularly stable. While many of those sold chose to remain attached to the estates into which they were sold and enforce the predatory structure of the estates, prazo-holders also experienced multiple rebellions and desertions by their enslaved armies of enforcers. In particular, during the 19th century, productivity on the estates was so low due to the capacity of African communities to evade compulsion and the absence of investment, that prazo-holders began to try to sell off their enslaved armies to the lucrative slave trade headed for Brazil (Isaacman and Peterson, 2003). The Chikunda -the enslaved armies of the prazos in the Zambesi region -mobilised and deserted to form autonomous and migratory communities from the late 19th century onwards.
In the wake of the Berlin conference in 1884-1885, where Portugal had just about clung on to nominal imperial control of its African territories, it was forced ultimately to lease them to mostly British, French and German capital through the establishment of colonial companies from 1891. The Nyassa Company, the Mozambique Company and the Zambezia Company took over the vast lands formerly occupied by the prazos (Neil-Tomlinson, 1977;Vail, 1976). Although better supplied with capital and administrative capacity, and run more like plantations than the former estates, the foundation of their extractive capacities was fundamentally the same -heavily coerced military and agrarian labour based on 'taxing' the surrounding African populations, and, increasingly, supplying demand for migrant labour in South African mines. Between 1891 and 1929, these companies were the foundation of the colonial rural economy. Labour was not however limited to agricultural production -the rural infrastructure of roads and railways developed by these companies was also created by forced labour, known as chibalo. 15 When war was declared in 1914, then, numerous Africans were already working within a system which had for decades, and in some cases centuries, conscripted their collective labour forcefully as instruments of production and of violence. The forced growing of cotton, for example, certainly required both in large quantities, being labourintensive and therefore in general producing famine among those who had little time to grow food (Isaacman et al., 1980). Yet as demand for cotton surged with the war, so the demands on the peasants were stepped up. Resistance and resentment towards this system were however widespread, as documented through acts of sabotage, evasion and low productivity rates (Pitcher, 1996).

Invisible bodies of the 'Great War': porters, soldiers, villagers
What did the declaration and conduct of war mean, then, in Portuguese East Africa, in a context where compulsion and violence were already written into the fabric of everyday life? In a general sense, conscription intensified and developed new and more aggressive forms. This produced a sharp acceleration in suffering among many of the populations -yet the enormous pile of bodies produced remains largely invisible to commemoration practices, and our understandings of World War I and modernity more generally.
One key aspect of this intensified conscription was in the practice of press-ganged military porterage. Tens of thousands of Africans from the north of Mozambique -a relatively sparsely populated area -were forced to carry supplies and equipment for the Portuguese, British and German forces, following around the military detachments with even less equipment and provisions than the already poorly prepared European and colonial soldiers (Strachan, 2004).
Estimates of the total numbers of porters involved in the whole Allied campaign in East Africa are around 1,000,000, with up to 95,000 dead among those used by the British alone (c.600,000) (Paice, 2010). A death rate of around 20 percent of the porters of the East Africa Carrier Corps is more than double the 9 percent death rate for British troops, reflecting active policies of imperial indifference to their conditions. Given the much worse conditions experienced by both troops and porters within the Portuguese army, the death rates are likely to be higher still. The biggest killers were not bullets but disease and malnutrition. Yet, for the Germans and the Portuguese, there are no clear numbers -the porters were never counted, their ultimate disposability being a founding assumption of those colonial war machines. One particularly chilling number is however offered by the then Director of the Zambezia Company, who claims to have sent 25,000 men as porters, of whom only 5,000 returned, and those who returned being in 'such a state as to cause horrors to onlookers' (Carvalho, 2014). An 80 percent death rate, if accurate, is comparable to periods of combat in the widely commemorated Battle of the Somme.
The writings of Cardoso Mirão (2001Mirão ( [1926), a Portuguese sergeant serving in 1917, testify as much to the systematic de-humanisation of these men. His writings were suppressed by the fascist dictatorship which took power after 1926 for bringing the army into disrepute: They are not men, because they have no name. They are not soldiers, because they have no number. You do not call them, you count them. They are placed into formation by force of the stick, you put a cargo on their heads and that's it. (Mirão, 2001(Mirão, [1926, quoted in Carvalho, 2014) 16 Other were press-ganged or kidnapped not into porterage but into combat roles, to serve as Portuguese soldiers. The recruitment process for these fighters, as for the porters, does not appear to different radically from the slave-raiding practices of the previous century. Mirão (2001 [1926]) writes, These coarse people, short thongs between their legs and dirty curly hair on their heads, were soldiers caught running through the fields, people recruited through the bush and the forest, chained by the strap to each other in long lines attached by wires and led to the barracks under the constant threat of shotguns. They were people caught by surprise, as they trap gazelles or beasts to the fold, and led by the neck into a high cordon, with guards in sight to prevent deserters and the escape of future combat troops that would serve under our command. And it was these soldiers who were going to minister to military education and the art of modern warfare! (Cardoso Mirão (2001Mirão ( [1926): 28-29) 17 Reading Mirão's (2001[1926: 31-32) account further, an episode is recounted of an attempt to make the soldiers perform in a military parade in front of medal-bedecked senior officers, which results in a lot of chaos, confusion and people walking into each other. After some grim looks from the senior officers, it is all blamed on the innate stupidity of the Africans, and eventually the Portuguese commanders decide to dispense with this part of the training and carry on with war preparations.
One can only imagine -and reading from 'below' suggests that we should try to imagine in the absence of alternative archives 18 -what the kidnapped, whipped African cultivators made of their cruel capture, the bizarre marching rituals, incomprehensible shouting and strict linear formations of European military training, insistently carrying on in the hottest parts of the day, with no visible enemy, threat or purpose -not to mention the ghastly physical condition of the men who were supposed to be their civilised superiors. What they made of their own condition must be in little doubt; the level of violence and coercion required to bring them to the barracks, and their near-constant guarding thereafter described by Mirão indicates that there was little will or acceptance of their conscription.
Given the ongoing systems of taxation by the colonial companies and chibalo labour in this same time and space, we can imagine that both porters and soldiers feared for their families left to satisfy these demands in their absence, and with fewer protectors than before. Indeed, the loss of male heads of families in patrilineal cultivation systems might have entailed the further break-up of land and livestock holdings, or their resettlement under other male relations -in this sense, their kidnapping would have not only represented the likely death of their individual bodies, but more profound extinction and structural terror -the wholesale destruction of their family structures. Their now 'orphaned' children may have themselves been offered up to other recruiters and overseers in exchange for leaving the rest of the new household alone, or some other manufactured goods.
Among the inhabitants of northern Mozambique, which was being raided by German forces with regularity, forcible predation was not limited to those who were taken as porters and soldiers, but imposed on the villages of the region, which were forced to deliver up food and supplies to passing armed forces of all stripes. The Nyassa Company had practised similar kinds of predation for decades -episodes are recorded where the women of Makonde villages were kidnapped and held as hostages until their husbands and families delivered some kind of 'taxation' to the company. In these periods of evasion and depopulation, famine and disease spread easily among the general population whose coping systems had been destroyed. Drought in 1918 compounded the numbers dying from hunger and also forced migration. At Versailles, the Portuguese claimed that the war had resulted in the loss of 120,000 indígenas (natives), for which it wanted compensation. 19 War and colonialism, in Portuguese East Africa, then, functioned in profoundly similar ways in terms of how the African populations were organised, exploited and ultimately killed by the demands of extractive predation. Clearly, there was a new intensity to the system in the north of the territory when operated by multiple hungry militaries rather than by a sparsely staffed monopolistic colonial company. For example the demand for road and rail infrastructure exponentially increased, meaning an increased demand for chibalo. Moreover, the more competitive character of colonial relations between the Europeans in this period -as compared with their previously collaborative natureopened up a series of further opportunities for rebellion and war among the peoples across the region, some of whom had been 'pacified' only recently, to which we will return. In this sense, the declaration of war between European powers was not trivial for the peoples of the regions. Yet, neither was it a transformative moment of rupture that marked the entry of African peoples into 'modern' systems of organised violence and labour extraction; rather it was precisely a continuity and intensification of long-standing patterns of interaction and coercive organisation, that is, of structural violence.

Imperial anxieties and racial prestige
The specifically strange and destructive shape of the war in northern Mozambique revealed by contrapuntal analysis requires, as part of the explanation, an understanding of the historical role of white colonial prestige and racial anxiety within the imaginaries of the European parties to the conflict. Speaking of the war, Du Bois (1915) argued strongly that the global colour line was intensifying as part of the struggle between labour and capital within the white nations. Moreover, he argued that it was Africa itself which was at the time understood as having the greatest exploitative potential for furthering European economic expansion. A desire to retain control within Africa was thus a key animating concern for all the imperial powers of the period, even if not directly strategically related to the war in Europe.
While Du Bois argues forcefully about the centrality of imperial aspirations for control of Africa in the period, in this argument he downplays the significance of imperial status and prestige in their own right as a political force in favour of a more materialist account. The fate of the armies in the Portuguese Republic in East Africa during the war, and the collapse of the Republic in the years afterwards, however seemed to turn in large part on an attempt to avoid imperial embarrassment and the shame of degeneracy among its European counterparts. While this aspect of Portuguese colonial history is known in general terms, it is not often brought into dialogue with the questions about the global significance of the war.
Although at the Berlin conference Portugal seemed to have salvaged recognition of its African claims, a series of secret British-German treaties had made plans for the partition and re-allocation of its territories. The last of these was signed in early 1914. This was on top of the fact that the colonial companies in Portuguese East Africa after 1891 were heavily capitalised by other Europeans. Its failures functionally to occupy the territory and suppress slavery, its chronic poverty and its degenerate old habits of miscegenation with African populations were all seen by its imperial rivals as evidence of Portuguese backwardness. This backwardness, if admitted, undermined Portuguese claims to administer African territory, since an ability profitably to cultivate or develop colonial possessions was a cornerstone of the justification for the division of Africa in the first place.
When Portugal entered the war, its expressed primary objective was to protect its colonial status, even though it could barely afford to do so (Paice, 2010). It could not have been said to be doing well out of its colonial possessions as profit-making enterprises in 1914. But this did not prevent the committal of thousands of troops and state money by the Republicans to the defence of the colonies. Despite this, it won no military victories, lost equipment and supplies to the German raids and largely failed to assist its allies. 20 More Portuguese soldiers died in Africa than in Flanders, but the vast majority simply from disease and poor conditions rather than through rare instances of combat. The gulf between its imperial anxieties and its tactical capacities extinguished tens of thousands of lives, both in terms of the European and askari troops, 21 and in terms of the Africans who died in the effort to keep this colonial war machine afloat. This is not to claim any ethical prestige or advantage for the relatively more successful war machines of the Germans or the British in East Africa -efficient producers of death and suffering in their own way, although both would claim superiority and restraint on this front. But the Portuguese engagement with the war makes the point more starkly about the centrality of colonial possession and imperial prestige to 'great power' global political status at the time -in this sense, the war arguably provoked a greater existential crisis for the political identity of the Portuguese than the British or Germans.
Imperial anxiety also contributed to the rise of authoritarian political tendencies within Portugal. Although the long-running fascist dictatorship of Salazar did not emerge in Portugal until the late 1920s, two authoritarian coups took place during the war which mobilised themselves around the threat of losing the colonies and the need to reinforce its imperial status. Authoritarian and later fascist politicians bought strongly into the image of crumbling and decadent Portuguese colonial settlers who had 'gone native', running the colony into chaos and vice. It was only a firm, Catholic hand that could redeem the glory of the Portuguese empire.
The response from this point onwards would be to extend the presence of Portuguese settlers and administrators into Mozambique rapidly and to classify it as an 'Overseas Territory' of the Portuguese nation, broken into provinces which largely picked up the boundaries of the colonial companies. Re-writing the shameful history of the war, the administration censored works which had criticised the campaigns, such as that of Mirão, and chose instead to affirm it as a moment of multi-racial national co-operation. This attempt is inscribed in Mozambique's war memorials from the 1930s, in particular the Padrão da Grande Guerra. Located not at the sites of battlefield in the north of the country but in the capital in the far south -in front of you as you step out of the train station -is the monument to the Great War. As noted by Verheij (2011), this is not a monument to the past dead, as other European monuments of the time were -it is the projected imaginary of a flourishing colonial present. Interestingly, it is one of the few places where Africans are projected as equal participants in the war -not a concern at the time of war, but only afterwards within a project of authoritarian national-imperial integration.

Uprisings, collaboration, repression and escape
The largest and most destabilising armed confrontations of the period 1914-1918 in 'Portuguese East Africa' however were not between Europeans working through imperial rivalries, but multiple African uprisings against colonial predation and rule, which a contrapuntal analysis throws into relief. These uprisings are commemorated in presentday Mozambique as key moments in national and nationalist histories, preserved in history textbooks, banknotes, political speeches and so on. Clearly, they offer a point of departure for a historical narrative of African agency and political resistance which resonated with the anti-colonial projects of the liberation fronts at the time, and their attempts to preserve their hegemony in the present. Yet, the rebellions need not only be woven into nationalist dogma -they can also be read back into a larger historical fabric of violent and non-violent resistance to extractive and authoritarian tendencies past and present. They also speak to the mobile and transnational character of resistance strategies in southern Africa, and to the historical power of rivalries within the territory.
The Chilembwe rebellion, which took place against the British in neighbouring Nyasaland (Malawi) in 1915, was naturally a concern for the Portuguese colonists. Chilembwe, a Christian preacher trained in the United States, had been a vocal and active anti-colonial campaigner for some time. In particular, he widely protested the racist basis of colonial rule, advocating instead the equality of men before God, and demanded an Africa for the Africans. The uprising consisted of planned co-ordinated attacks on the British settlers and infrastructure by hundreds of Africans, and for about 2 weeks succeeded in mobilising large numbers of people and controlling much of the non-urban space. The rebellion was violently suppressed with a mixture of British-African and Portuguese-African troops and ultimately failed. Nonetheless, it decisively changed attitudes among the British about the assumed docility and loyalty of the peoples of Nyasaland and demonstrated to some extent the significance of inter-imperial collaboration against anti-colonial rebels (Paice, 2010: 161-165).
The Portuguese on the other hand did not need lessons about the potential of rebellion in East Africa, particularly in Zambezia. This region resisted integration successfully into Portuguese rule for centuries, culminating in a series of long armed struggles with the autonomous Barue people from 1880 which had only ended in 1902. Others further south, such as the Gazan king Ngungunhana, had only been finally defeated in 1895. Relations with the Ngoni people in the region had also been complex, turning largely on managing rivalry between them and the Barue.
The emergence of a huge uprising in 1917 by the Barue / Makombe leaders was only a surprise then to the extent that it successfully co-ordinated other groups such as the Sena and Yao to mobilise simultaneously against colonial rule instead of being divided by it. They were also joined by the Chikunda, the descendants of the enslaved armies from the prazos (Isaacman and Isaacman, 1976). The Barue recognised the spiritual and political leadership of the role of the Makombe (leader), although there was competition for that role. The immediate cause for the rebellion was an upsurge in the demand for forced labour, to work on infrastructure projects in support of the war effort, as well as the breaking of various promises around taxation by the Portuguese. Under this rebellion, which lasted from March until November, 100,000 fled and hundreds of loyalists were killed. In comparison to the rather feeble numbers and state of troops deployed against the Germans, the Portuguese rapidly deployed 20,000 well-armed, well-paid Ngoni mercenaries, with a licence to plunder women and resources, to put down the rebels with as much force as possible. Many of the Barue leaders and the Makombe fled across the border to Malawi.
In the north of the country during the war, the Makonde increased their ongoing resistance against the Portuguese, facilitated by promises of German abolition of the hut tax and chibalo, which had also been made towards the Yao and Makua. As with the Barue, the demands for chibalo had increased as a result of the war demand for infrastructure, and the peoples of the plateau found that the autonomy that they had been used to was under threat. While the Germans supplied arms, they did not fight alongside the Makonde, who had also never been fully incorporated into Portuguese rule. Rather, fighters mobilised by Malapende, a powerful figure in Makonde history, managed to fight the Portuguese forces in the bush and in the villages until 1917 (West, 2005: 87-98). The Portuguese finally mobilised 2,000 well-armed Yao against the Makonde, torching 150 Makonde villages for suspected collaboration with the Germans during between April and June in 1917 (Carvalho, 2014). Malapende could not bear to see his people subjected to the cruel conditions of forced labour, so he fled north over the border and waited.
Forty-five years later, when the Mozambican liberation activists were looking for a base and an entry point into Mozambique, it is not surprising that they began in Tanzania and moved through the Muedan plateau (Israel, 2006). The Makonde had been to some extent historically defined by resistance to the slave trade, Portuguese colonial rule and the predation of the Nyassa Company throughout the 20th century. In this sense, the war of 1914-1918 was an episode which re-inscribed the character of colonial domination, and which would have a long-standing impact on the peoples of the region as political actors, but which did not capture the totality of their historical presence. The sudden massacre in 1960 of Makonde protestors who had been called to the governor to make their grievances known was a key moment in the lead-up to the founding of Frelimo in 1962. The spirit of Malapende was seen by the Muedans to have returned within the mapiko dances after independence -commemoration practices of the 'living' spirit of resistance rather than its dead bones (West, 2005).

Reading the Great War and international order contrapuntally and from below
Coming back to our opening question, then, we are now in a position to suggest how international order itself can be understood differently when we read the Great War from its margins. In this section, I advance four analytic conclusions that speak to this concern with international order, which foreground the dynamics of order read from below, and in counterpoint to narratives of order that focus on the imperial metropole. These conclusions necessarily underscore the importance of the methodological approach taken here to rethink the international.
The first point is that in the colonial context, and particularly when looking at the experiences of colonised peoples, 'war' is not a significant rupture from 'peace'. In fact, as has been shown, there were significant continuities with colonial violence and predation throughout these two phenomena, albeit with some forms of intensification and transformation that served evolving imperial objectives. This speaks back to the literature which notes the hierarchical and dualist character of international order (e.g. Keene, 2002) in which 'civilisation' rather than 'toleration' characterised the imperial part of the order. Reading this from below and in counterpoint to the discursive projections of empire, however, the dynamic is not so much 'civilisation' as it is exploitation/ predation institutionalised through regimes of violence, that cut across both 'order' and 'disorder'.
Second, the racialised disposability of colonial populations was a constitutive element of international order -slightly ameliorated if they were tied officially to imperial service but basically not at all if the people in question simply had the misfortune of living for centuries in space claimed by colonial powers as property. This was systematic across political, economic, military and social spheres. Racism, in this understanding, is not reducible to racial slurs, open violence and eugenic science, but the studied and deliberate indifference to one's death and suffering based on assumed descent. This brings us closer to understanding its structural implications as a governing principle of 'modern' human experience -or what Mbembe (2003) has called 'necropolitics'.
This understanding of racism as indifference and disposability resonates with a broader set of discussions about the place of racism in global modernity (see Hesse, 2007). Recent writings in de-colonial theory have advanced the claim that the constitutive racism of modernity/coloniality (being two sides of the same coin) is not only integral but in some sense analytically prior to thinking about capitalist, nationalist and industrial transformations, although the latter are contained within them (Grosfoguel, 2002;Robinson, 1984). By reading modernity itself contrapuntally, the racism and 'body politics' (Mignolo and Tlostanova, 2006) inscribed within this structure are certainly integral and indispensable to the ways in which the patterns of violence, coercion and extraction develop, suggesting that these dynamics should be an analytic priority in thinking about order itself.
The third point is that colonial conflicts, anxieties and strategies were significant and existential issues for European powers throughout the war. Indeed, when we look at the kinds of conflicts leading up to the war, from the 1905 confrontation between France and Germany in Morocco, to the Italian invasion of Libya in 1911, it is clear that rather than the outbreak of war in 1914 being a 'surprise' to many in Europe, inter-imperial violence was already underway (Gerwarth and Manela, 2014). During the war, the Portuguese government was twice replaced because of the anxiety over the loss of its imperial standing, Britain aspired to control the territories of its supposed ally, and in the treaty of Versailles, imperial losses and reparations constituted a significant sphere of negotiation. When we think about international order, it is classically argued that states organise it and wage war in order to avoid 'state death' -yet the actualities of the war suggest that imperial status and 'imperial death' has been at least as significant a driving force.
The final point is that the intensification of exploitation stemming from the War also dynamised forces of resistance to colonial rule itself, requiring the ongoing 'modernisation' of colonial governance and increasing attempts to accommodate / re-categorise / memorialise colonial rule as something else. This was not through a 'contagion effect' from Europe, relating to ideologies of self-determination necessarily, although colonised peoples clearly shared ideas (e.g. between Indian and Irish thinkers on demands for home rule). Rather, its roots were in the capacities of people experiencing violence to organise themselves against it, eventually leading to the wider military, political and normative conditions in which decolonisation became more widespread in the 20th century. By locating some of these dynamics in the Great War, rather than some kind of change of heart by European powers after the Second World War, we have a better capacity to understand the structural nature of change in international order that took place as a result of this resistance.

Conclusion
Bringing these conclusions together then, we can argue that international order -that is, patterns of interactions between polities -in this period and perhaps beyond, is constituted not principally by a balance of power, or a 'standard of civilisation' but by a dynamic, hierarchical matrix of structural violence, cutting across 'wartime' and 'peacetime', characterised strongly by racialisation, imperial anxieties and forms of resistance. The conception of a 'matrix' of violence helps us put both its horizontal (i.e. inter-imperial) and vertical (i.e. colonial) dimensions into the same analytic frame, to appreciate international order more holistically, to understand the interdependences between processes located thousands of miles apart, to nonetheless appreciate the colonial experience on its own terms and to foreground the violence and suffering of humans in a way which honours the aspirations of many in the early field to end violence. Therefore, our understandings of international order, violence, hierarchies and modernity itself may be systematically inadequate if we do not consistently aim to read phenomena 'from below' or in counterpoint across different sites of the system as a whole. Within IR's historical turn, then, we cannot content ourselves principally with reading the archives of elites and their correspondents or antagonists, but must also find ways to grasp the experiences, agencies and processes of the majorities that make up societies and social orders, in order to reconnect our understanding and theorisation of the international to the people that constitute it. Such a reading can be called 'contrapuntal', insofar that it thinks laterally and more 'pluralistically' about its subjects, objects and epistemologies, and which can thus reach different conclusions about the forces and patterns structuring global order. By utilising such approaches for the contemplation of the international more broadly, as a field we can more adequately meet the challenge of democratising and decolonising its gaze. 6. Some of what falls under this literature does not necessarily over-emphasise norms but material enablers and constraints at the level of states/powers, for example, Buzan and Lawson (2015) and Phillips and Sharman (2015). 7. See also, for example, Biswas (2014) on how the asymmetric control of the means of violence produces forms of injustice and disorder. 8. Specifically Japanese and Ottoman imperial power, although the latter had been considerably eroded over the 19th century. 9. See, for example, de Carvalho et al., 2021;Elman, 1997, 2008;Hobden and Hobson, 2002;Lawson, 2006Lawson, , 2012Hobson and Lawson, 2008;Rosenberg, 2006). 10. See also volume edited by Phillips and Reus-Smit (2020). 11. We recall here that the label 'Third World' was developed in reference to the 'Third Estate'. 12. On some of the issues raised, see Ginzburg, (2013Ginzburg, ( [1976) 'The Cheese and the Worms'. 13.  reads IR from below using the autobiography of a French soldier and prisoner of war in World War II. 14. A large paddle for beating people. 15. The Portuguese were not alone in relying on forced labour, of course. Across the border in Nyasaland (now Malawi), the thangata system required Africans to offer free labour to European estates -so much during the war that it was locally named the 'war of thangata' (Page, 1978). 16. 'Não são homens porque não têm nome; também não são soldados, porque não têm número.
Não se chamam, contam-se. Formam-se a varapau, põe-se-lhes uma carga à cabeça e pronto . . .' (Mirão (2001(Mirão ( [1926), quoted in Carvalho, 2014). 17. 'Aquela gente rude, de tangas curtas entre as pernas e carapinha suja na cabeça, eram soldados apanhados a monte pelas machambas (plantações), gente recrutada a laço pelas roças e pelos sertões, acorrentados pela cinta uns aos outros, em longas filas presas por arames, e conduzidos para o quartel sob a ameaça constante das espingardas do recrutados. Era gente apanhada de surpresa, como se apanham gazelas á ratoeira ou feras ao redil, e conduzida pelo pescoço para dentro de uma cerca alta, com sentinelas á vista para evitar desertores e a fuga da futura tropa de combate que haveria de servir sob o nosso comando. E era estes soldados a quem íamos ministrar a instrução military e a arte de fazer a guerra moderna!' (Mirão (2001(Mirão ( [1926): 28-29). 18. See Guha's (1987) methods for investigating 'Chandra's Death'. 19. Given the context, the number is likely inflated / invented, but may not be so far off. Statistic from Carvalho (2014). 20. As an indication of the poor preparation and state of the troops, in one military memoir it is recorded that they had sent 26 trucks but only six drivers for the large campaigns of 1917 (Martins, 1934: 183-189). 21. The role of askari troops in this region is huge, constituting the vast majority of those fighting under European flags. Time and space do not permit a proper treatment of the complex and ambiguous role that they occupied; sworn and faithful to empire, yet themselves subjected to colonial racism and forms of violence. Moreover, the Yao were known for switching between fighting for the Germans and for the British interchangeably. See Barkawi's (2006) work on Indian troops in the British army for insightful reflections on the significance of this mediating war-space.