accepted version of article forthcoming in Journal of Historical Geography

abstract

Keywords

Introduction

[1] AISL Discussion Paper for the Groupings Meeting: Fundraising in Amnesty International’, September 1990. Folder 145, Amnesty International Nederland archief. International Institute for Social History, Amsterdam (hereafter AI NL).

[2] Edward Armston-Sheret, ‘Diversifying the Historical Geography of Exploration: Subaltern Body Work on British-Led Expeditions c.1850–1914’, Journal of Historical Geography 80 (2023) 58-68 (p. 59).

[3] Adom Getachew, Worldmaking after Empire (Princeton: Princeton University Press, 2019); Peggy Levitt and Sally Merry, ‘Vernacularization on the Ground: Local Uses of Global Women’s Rights in Peru, China, India and the United States’, Global Networks 9 (2009) 441–61.

[4] Ruth Craggs and Hannah Neate, ‘What Happens If We Start from Nigeria? Diversifying Histories of Geography’, Annals of the American Association of Geographers 110 (2020) 899–916; Jake Hodder, Stephen Legg, and Mike Heffernan, ‘Introduction: Historical Geographies of Internationalism, 1900–1950’, Political Geography 49 (2015): 1–6 (p. 2).

[5] Ann Laura Stoler, Along the Archival Grain: Epistemic Anxieties and Colonial Common Sense (Princeton: Princeton University Press, 2010); J.K. Gibson-Graham, ‘Reading for Difference in the Archives of Tropical Geography: Imagining An(Other) Economic Geography for Beyond the Anthropocene’, Antipode 52 (2020) 12–35; Caroline Bressey, ‘Surfacing Black and Brown Bodies in the Digital Archive: Domestic Workers in Late Nineteenth-Century Australia’, Journal of Historical Geography 70 (2020) 1–11.

[6] Martin Müller, ‘Opening the Black Box of the Organization: Socio-Material Practices of Geopolitical Ordering’, Political Geography 31 (2012) 379-388 (p. 379).

Non-governmental organisations and non-state worldmaking

[7] Jake Hodder, Michael Heffernan, and Stephen Legg, ‘The Archival Geographies of Twentieth-Century Internationalism: Nation, Empire and Race’, Journal of Historical Geography 71 (2021) 1–11; Federico Ferretti, ‘History and Philosophy of Geography I: Decolonising the Discipline, Diversifying Archives and Historicising Radicalism’, Progress in Human Geography 44 (2020) 1161–71.

[8] Hodder, Heffernan, and Legg, ‘The Archival Geographies of Twentieth-Century Internationalism’; Ferretti, ‘History and Philosophy of Geography I’; Patricia Clavin, ‘Roundtable: “Governing the World” by Mark Mazower’, History Workshop, 8 October 2013, https://www.historyworkshop.org.uk/ideas/roundtable-patricia-clavin/ (accessed 7 December 2023); David Featherstone, Richard Phillips, and Johanna Waters, ‘Introduction: Spatialities of Transnational Networks’, Global Networks 7 (2007) 383–91; Mark Philip Bradley, The World Reimagined: Americans and Human Rights in the Twentieth Century (Cambridge: Cambridge University Press, 2016).

[9] Carolien Stolte, ‘Introduction: Trade Union Networks and the Politics of Expertise in an Age of Afro-Asian Solidarity’, Journal of Social History 53 (2019) 331–47; Duncan Bell, ‘Making and Taking Worlds’, in Global Intellectual History, ed. by Samuel Moyn and Andrew Satori (New York: Columbia University Press, 2013), pp. 254–80; David Featherstone, ‘Maritime Labour, Circulations of Struggle, and Constructions of Transnational Subaltern Agency: The Spatial Politics of the 1939 Indian Seafarers’ Strikes’, Antipode 55 (2023) 1411–32.

[10] Stoler, Along the Archival Grain, p. 1.

[11] Stoler, Along the Archival Grain, p. 2.

[12] Florian Weisser, ‘Practices, Politics, Performativities: Documents in the International Negotiations on Climate Change’, Political Geography 40 (2014) 46-55 (p. 49).

[13] David Featherstone, Ben Gowland, and Lazaras Karaliotas, ‘Solidarity, Worldmaking and Inter-Connected Geographies of Authoritarianism: Trade Unions and the Multiple Trajectories of Chile Solidarity’, Alternautas 11 (2024) 16-58.

[14] Kevin O’Sullivan, The NGO Moment: The Globalisation of Compassion from Biafra to Live Aid (Cambridge: Cambridge University Press, 2021).

[15] ‘Report on Sections and their Development Issues’, November 1984; ‘Correspondence from Bogotá group to International Executive Committee’, 16 January 1985. Microfilm 285, Amnesty International, International Secretariat archive. International Institute for Social History, Amsterdam (hereafter AI IS).

[16] For conferencing as an important site for internationalism, see Placing Internationalism: International Conferences and the Making of the Modern World, ed. by Stephen Legg, Mike Heffernan, Jake Hodder and Benjamin Thorpe (London: Bloomsbury, 2021).

[17] Müller, ‘Opening the Black Box of the Organization', p. 384.

[18] Roland Wenzlhuemer, ‘Globalization, Communication and the Concept of Space in Global History’, Historical Social Research / Historische Sozialforschung 35 (2010) 19–47; Michael G. Müller and Cornelius Torp, ‘Conceptualising Transnational Spaces in History’, European Review of History: Revue Europeenne d’histoire 16 (2009) 609–17.

Decentralisation and effective action

[19] Amnesty International, Amnesty International Report 1989 (London: Amnesty International Publications, 1989), p. 16.

[20] Stephen Hopgood, Keepers of the Flame: Understanding Amnesty International (Ithaca: Cornell University Press, 2013).

[21] Samuel Moyn, The Last Utopia: Human Rights in History (Cambridge: Harvard University Press, 2010); Jan Eckel, The Ambivalence of Good: Human Rights in International Politics since the 1940s (Oxford: Oxford University Press, 2019).

[22] For Wickremasinghe, see Correspondence, Microfilm 551, AI IS.

[23] Hopgood, Keepers of the Flame, p. 84.

[24] Hopgood, Keepers of the Flame, p. 84.

[25] Hopgood, Keepers of the Flame, p. 84.

[26] ‘Report of the Crash Committee Appointed to Advise on AI Development & Planning’, July 1977. Folder 118, AI IS.

[27] ‘Draft two-year plan for 1980 ICM’, February 1980. Folder 158, AI IS; ‘Discussion Paper on Development’, March 1980. Folder 158, AI IS; Michelle Carmody, ‘Making Human Rights International? Amnesty International, Organizational Development, and the Third World, 1970–1985’, Human Rights Quarterly 43 (2021) 586–606.

[28] Dennis Rondinelli, ‘Decentralization and Development’ in International Development Governance, ed. by Ahmed Shafiqul Huque & Habib Zafarullah (London: Routledge, 2006), pp. 391-404.

[29] Sam Halvorsen, ‘The Political Opportunities of Urban Decentralisation: Mobilising Local Governance in Buenos Aires’, Political Geography 74 (2019) 102038, https://doi.org/10.1016/j.polgeo.2019.102038; David Featherstone, Anthony Ince, Danny Mackinnon, Kendra Strauss and Andrew Cumbers, ‘Progressive Localism and the Construction of Political Alternatives’, Transactions of the Institute of British Geographers 37 (2012) 177–82.

[30] Featherstone, Ince, Mackinnon, Strauss and Cumbers, ' Progressive Localism'.

[31] Nathan Schneider, ‘Decentralization: An Incomplete Ambition’, Journal of Cultural Economy 12 (2019) 265-285 (p. 266).

[32] ‘Report of the Committee on Long Range Organisational Development to the 1987 International Council’, 1987. Microfilm 619, AI IS.

[33] ‘ Report of the Committee on Long Range Organisational Development’, Microfilm 619, AI IS.

Knowledge production and national sections

[34] ‘Basic/dependent sections with acute problems’, July 1980. Microfilm 251, AI IS.

[35] ‘Report on Visit to Pakistan section’, November 1980. Microfilm 251, AI IS.

[36] Wenzlhuemer, ‘Globalization, Communication and the Concept of Space in Global History’.

[37] ‘Draft proposal for Criteria for National Sections and Groups’, March 1980. Folder 158, AI IS.

[38] ‘Draft proposal for Criteria for National Sections and Groups’; ‘Principles Underlying AI’s Development Work’, 13 November 1986. Microfilm 512, AI IS.

[39] For an accessible introduction and overview, see Susan George, A Fate Worse than Debt (New York: Penguin, 1988).

[40] Jeff Haynes, Trevor W. Parfitt, and Stephen Riley, ‘Debt in Sub-Saharan Africa: The Local Politics of Stabilisation’, African Affairs 86 (1987) 343–66; A. B. Zack-Williams, ‘Sierra Leone: Crisis and Despair’, Review of African Political Economy 49 (1990) 22–33.

[41] A B Zack-Williams, ‘Crisis, Structural Adjustment and Creative Survival in Sierra Leone’, Africa Development 18 (1993) 53–65; 'Application for 1992 Budget Grant', 22 April 1992. Folder 145, AI NL.

[42] Zack-Williams, ‘Crisis, Structural Adjustment and Creative Survival’, p. 57; 'Letter to Nelleke Dieleman from Samuel Sankoh', 16 April 1991. Folder 145, AI NL.

[43] ‘Application for a Budget Grant’, 20 January 1992; ‘Meeting between Nelleke Dieleman (SDC) and David Gbao (AISL) on 4/5 March 1992’, 6 March 1992; ‘Meeting between Nelleke Dieleman (SDC) and David Gbao (AISL) on 4/5 March 1992 (re-written report)’, 9 March 1992. Folder 145, AI NL.

[44] Joel Kibazo, ‘Meridien Collapse: the inside story’, African Business (1 June 1995); Ifedapo Adeleye, Franklin Ngwu, Nkemdilim Iheanachor, and Ebes Esho, ‘Banking on Africa: Can Emerging Pan-African Banks Outcompete Their Global Rivals?’, in Africa’s Competitiveness in the Global Economy, ed. by Ifedapo Adeleye and Mark Esposito (New York: Palgrave, 2018), pp. 119–20.

[45] David Gbao, Oral history interview, Bristol, UK. 3 March 2023.

[46] Gbao, Oral history interview.

[47] ‘Meeting between Nelleke Dieleman and David Gbao (re-written report)’, 9 March 1992. Compare with ‘Meeting between Nelleke Dieleman and David Gbao’, 6 March 1992. Folder 145, AI NL.

[48] ‘Meeting between Dieleman and Gbao (re-written report)’. Folder 145, AI NL.

[49] Martin Mulsow, Knowledge Lost: A New View of Early Modern Intellectual History (Princeton: Princeton University Press, 2022), p. 4.

[50] ‘Application for a Budget Grant’, 20 January 1992. Folder 145, AI NL.

[51] Letter from AISL to Nelleke Dieleman, 16 April 1991. Folder 145, AI NL.

[52] Kenema Special Meeting report to 4th AGM, 1990. Folder 145, AI NL.

[53] AISL 4th AMG: National Development Plan’, 1990. Folder 145, AI NL; AISL Fishing Project memo, N.D., Folder 145, AI NL.

From local knowledge to decentralisation

[54] Frederick Cooper, ‘What Is the Concept of Globalization Good for? An African Historian’s Perspective’, African Affairs 100 (2001) 189-213 (p. 190).

[55] ‘Jaarverslag Nederlandstalig Afdeling België’, 1971. Inventory n. 455.00001, Archief Amnesty International Vlaanderen, AMSAB Institute for Social History, Gent.

[56] ‘AI Development and inter-section cooperation’, January 1983. Microfilm 264, AI IS.

[57] Letter from AISL to Anita Klum, AI Sweden, 7 December 1990. Folder 145, AI NL.

[58] 'Kenema Special Meeting report to the 4th AM’, October 1990. Folder 145, AI NL.

[59] Peter van Dam, ‘Challenging Global Inequality in Streets and Supermarkets: Fair Trade Activism since the 1960s’, in Histories of Global Inequality: New Perspectives, ed. by Christian Olaf Christiansen and Steven L. B. Jensen (New York: Palgrave, 2019), pp. 255–275; Peter van Dam and Andrea Franc, ‘Trajectories of Global Solidarity. Fair Trade Activism Since the 1960s: Introduction’, Contemporary European History 28 (2019) 512–17.

[60] Benjamin Möckel, ‘The Material Culture of Human Rights: Consumer Products, Boycotts and the Transformation of Human Rights Activism in the 1970s and 1980s’, International Journal for History, Culture and Modernity 6 (2018) 76-104.

Competing theories of decentralisation

[61] See José Zalaquett correspondence, Microfilm 551, AI IS ; ‘IEC Member reports : Bacre’, 19 November 1985, Microfilm 377, AI IS.

[62] ‘Minutes Subcommittee for Development’, November 1985. Microfilm 377, AI IS.

[63] Author correspondence with David Hinkley (AIUSA), 12 September 2024.

[64] Letter to David Gbao from Gerson Gu-Konu, 31 July 1990. Folder 145, AI NL.

[65] Letter to David Gbao from Gerson Gu-Konu. Folder 145, AI NL.

[66] ‘AISL Discussion Paper for the Groupings Meeting: Fundraising in Amnesty International’, September 1990. Folder 145, AI NL.

[67] Letter from David Gbao to Anita Klum & Goran Sveden, re: intersection cooperation, 7 December 1990. Folder 145, AI NL.

[68] Letter to David Gbao from Gerson Gu-Konu, 31 July 1990. Folder 145, AI NL.

[69] Proposal for Informal Discussion at Grouping Meeting, 1 August 1990. Folder 145, AI NL.

[70] Proposal for Informal Discussion at Grouping Meeting. Folder 145, AI NL.

[71] Letter to AISL from Gerson Gu-Konu, 8 October 1990. Folder 145, AI NL.

[72] Letter to AISL from Gerson Gu-Konu; Letter to David Gbao from Gerson Gu-Konu, 31 July 1990. Folder 145, AI NL.

[73] Letter to Section Development Committee from Swedish section, 25 February 1992.Folder 145, AI NL.

[74] Mark Granovetter, ‘The Strength of Weak Ties’, American Journal of Sociology 78 (1973) 1360–80.

[75] Davd Gbao, oral history interview.

[76] Ruth Craggs, ‘Hospitality in Geopolitics and the Making of Commonwealth International Relations’, Geoforum 52 (2014) 90–100.

Conclusion

[77] Hopgood, Keepers of the Flame, p. 114.

[78] Stoler, Along the Archival Grain. For work on counter-archiving, see Paul Ashmore, Ruth Craggs, and Hannah Neate, ‘Working-with: Talking and Sorting in Personal Archives’, Journal of Historical Geography 38 (2012) 81–89.

[79] Mulsow, Knowledge Lost, pp. xii, 1-9.

[80] Su Lin Lewis, ‘Skies That Bind: Air Travel in the Bandung Era’, in Placing Internationalism : International Conferences and the Making of the Modern World, ed. Stephen Legg, Mike Heffernan, Jake Hodder, and Benjamin Thorpe (London: Bloomsbury, 2021), pp. 234–52.

[81] Carolyn Steedman, ‘The Space of Memory: In an Archive’, History of the Human Sciences 11 (1998) 65-83 (p. 67).

[82] Getachew, Worldmaking after Empire.

Documentation, decentralisation, worldmaking: Amnesty International’s Sierra Leone section, 1980s-1990s

In this article I examine the process of using documentation to engage in institution building and (re)ordering. As such I contribute to understanding ways that actors built non-governmental institutions for worldmaking, the shaping of geopolitical relations. I also contribute to the project of writing a history of decentralisation, a key form of (re)ordering in the late twentieth century. The actors at the heart of this article are Amnesty International’s Sierra Leonean section, who used documentation to interact with the rest of the organisation and develop and circulate a theory of the decentralised international organisation. They did this by documenting what was actually happening on the ground and circulating this information up to the organisational centre of Amnesty in an attempt to shape its policy. While using the case of Amnesty’s Sierra Leone section, this article illuminates broader patterns in the organisation and its findings will be relevant to scholars interested in understanding how non-state actors engage in worldmaking via non-governmental organisations (NGOs).

worldmaking, decentralisation, international organisations, NGOs, Amnesty International, Sierra Leone

In 1990 Amnesty International’s Sierra Leonean section circulated a discussion paper amongst a handful of other national sections. ‘Is it’, they asked, ‘that the International Secretariat as a central coordinating structure has in its role been very irresponsive/insensitive to “new” thinking of weaker Sections?’[1] This critique of centralised structures echoed broader developments. Beginning in the 1970s the concept of ‘decentralisation’ became increasingly ubiquitous within a multitude of different organisational and institutional structures, invoking connotations of democratisation, participation and the redistribution of centralised power. Amnesty, like many non-governmental organisations, embraced the discourse of decentralisation in its planning and strategizing around how the relationships between different organisational scales should be routed. Yet for all the democratisation and participation that this term implied there were still power structures that remained resistant to change, as the Sierra Leoneans’ comments attested to.

Making an international movement, network or organisation is a ‘collective project of work’ involving not just leadership decisions but also ‘subaltern labour’, the everyday work of members in places like Sierra Leone.[2] As new frameworks emerge for understanding subaltern actors as theorists and shapers of global politics, so does the chance to illuminate the role of non-state actors whose contribution is so often reduced to one of ‘vernacularisation’, the translation of international politics into local contexts.[3] As the comments of Amnesty’s Sierra Leone section hint at, these activists were internationalists not just in the sense of seeking change at the international scale but in the sense of thinking about the mechanics and arrangements of international connection and cooperation that facilitates this change. Cognisant of the disparities in power within the organisation when it came to setting these arrangements, they invoked decentralisation’s abstract gesturing towards greater membership involvement and movement democracy and invited others to question whether the organisation’s leadership was fulfilling its obligation to lead change or whether it was in fact perpetuating existing distributions of power and influence. Circulating this challenge within a membership forum, they created an opportunity for themselves to engage in institution-building.

In this article I examine this process of using documentation, the material circulations that made up the international organisation, to engage in institution building. The aim of this article is to extend the concept of archiving as worldmaking, and the ordering of relations that this involves, to the endeavour of recognising diverse spatial imaginaries. Historical geographers have called for ‘starting from’ a ‘wider array of sites, people and politics than is often considered when addressing internationalist thought and practice’ as a way to uncover these more diverse histories, an approach taken in this article.[4] In terms of sources, existing methods and theories of the archive such as reading against the grain or reading for difference have provided blueprints for how alternative visions of the world can be surfaced from within the materials it contains.[5] But these methods, as immensely useful as they are, do not highlight the act of ordering. Martin Müller has called for the development of methods for ‘opening the black box’ of organisations, uncovering the ‘socio-material processes of ordering by which organisations are assembled and become more or less coherent entities.’[6] Like archives, organisations are sites where attempts are creating order out of disorder occurs. I propose looking at documentation as a vehicle for this ordering, ‘starting from’ Sierra Leone via the traces they have inserted into Amnesty’s organisational archive.

The actors at the heart of this article are Amnesty’s Sierra Leonean section, who used documentation to interact with the rest of the organisation and develop and circulate a theory of the decentralised international organisation. By the late 1980s, when the Sierra Leonean section was formed, a broader reordering of organisational and governmental structure was occurring under the banner of decentralisation. An empty signifier whose contemporary history is yet to be written, decentralisation offered a range of actors, from elites and other existing holders of power to grassroots actors seeking to redistribute or limit that power, a way of framing new theories of order. The Sierra Leoneans took up this opportunity. They called for a recognition that knowledge about how to run the organisation existed at all scales, a recognition that would in turn thicken and consolidate the ties between its constituent parts. They also proposed developing horizontal relations between parts of the organisation, based on prior affinity and relationships. They did this by documenting what was actually happening on the ground and circulating this information up to the organisational centre of Amnesty in an attempt to shape its policy. This version of decentralisation clashed with others being proposed throughout the organisation, giving rise to attempts to claim the authority to engage in ordering.

Discussions within historical geography on archives and internationalism have pushed us to diversify what we talk about when we talk about internationalism.[7] However, ‘traditionally’ international movements and organisations have acted as magnets for people across the world who have ‘reached out’ to them for various reasons, pointing to the existence of multiple ‘internationalisms’ within these spaces and, by extension, the continued importance of those spaces as sites where the world has been reimagined.[8] Recent scholarship on both internationalism and worldmaking encourages us to recover and revisit forgotten ‘chronologies of possibility’ to understand the multitude of ways that people have exercised their agency and ‘constructed and reconstructed worlds, drawing on the existing resources available to them, worlds carved from the materials of other worlds.’[9] Recovering these forgotten possibilities calls for specific methods informed by the insights of debates on archival practice and worldmaking.

Archiving imposes an order by making connections, but the archive still contains fragments that point to a new or different way of imagining order. The fragments at the heart of this article are a case in point. I develop this enquiry through an examination of a small series of documents produced by Amnesty International’s Sierra Leonean section, kept by Nelleke Dieleman, a Dutch member of the organisation’s Section Development Committee (SDC). The SDC was Amnesty’s internal subcommittee responsible for overseeing membership and organisational development in the global South during the 1980s and 1990s, and Dieleman was the SDC member responsible for Africa. They are documents that have spilled out from SDC meetings, pre-reading like position papers, draft resolutions and contextual information as well as (draft) repots and general correspondence. Dieleman most likely left them behind in the office of the Dutch national section (to which she belonged), resulting in them becoming preserved in that section’s own archive.

These ‘documents with itineraries of their own’ evidence worldmaking in the sense of a shared imaginary.[10] But they also do something more. These documents and the itineraries they contain, ‘produced the administrative apparatus as it opened on to a space that extended beyond it.’[11] They were worldmaking tools in that they contributed to institution-building, functioning as ‘hidden scripts that attempt to order and organise what is usually unordered.’[12] Recently some have called for extending this understanding of worldmaking, as institution-building with a view of bringing about geopolitical change, beyond the focus on state actors and political elites to non-state actors.[13] From the 1960s onwards, non-state actors’ access to the realm of global politics was increasingly mediated by non-governmental organisations (NGOs), the rise of which was itself part of changing North-South geopolitical relations in the wake of decolonisation.[14] Differently situated individuals and group have contributed to the ongoing constitution of these organisations through flows of information, ideas and resources. In doing so, they engaged in the ongoing project of ordering and structuring this vehicle for geopolitical change.

In the case of Amnesty, organisational order and structure was intimately connected to ideas about now non-state actors could wield this geopolitical power. In the early 1970s the organisation’s leadership began a concerted effort to develop national sections across Asia, Africa and Latin America. Incorporating a membership base beyond Western Europe aimed at allowing them to legitimately lay claim to being an ‘international’ rather than a western organisation, a claim that in their mind would facilitate influence with states in these parts of the world. While membership remained small compared with the organisation’s strongholds in Western Europe and its settler colonies, in the following decades significant numbers of human rights activists from across the global South decided to join this international organisation.

But these new members did not see their membership as merely something that allowed the organisation to claim the mantle of ‘international’. As they gained experience in setting up and running national sections and in participating in the broader organisation, they reflected on the ways it was structured and run and wanted to put these ideas forward. Some chose to resign from the organisation, preferring to pursue their political work within other organisational spaces.[15] Some pursued change by participating in organisational committees or via interventions into meetings and congresses, key spaces where the international network was coordinated.[16] Yet for those at the centre of this article, they pursued this goal in the course of their everyday work in the organisation. Documentation, encompassing the production of documents and their circulation, was a key way of communicating these ideas and forging associations, ‘critical for linking together the parts and creating organisations as actors.’[17] Communicating across borders and distances, reporting on activities, formulating budgets: these are all activities that aim to make the various parts of the international body cohere into something capable of effective action.[18] They were also activities in which the struggle over that ‘something’ was conducted.

In Sierra Leone a handful of individuals, primarily lawyers, had affiliated to Amnesty during the 1970s. But it was not until the mid-1980s that people with a more activist-orientation, keen to carry out Amnesty activities, formed a local group. These members were celebrated by the organisation for their ingenuity and dedicated: the Annual Report for 1989 enthusiastically relayed the fact that they had managed to get one in every 1000 inhabitants of Sierra Leone to sign a petition as part of the organisation’s international Human Rights Now! Campaign.[19] By 1990 they had applied for section status, meaning they would now constitute a formal part of the international organisation.

The organisation they joined had, by this time, evolved from its loose, amateur roots into a more professional, disciplined and legible structure that was designed to ensure, among other things, effective incorporation of these new members across the world. But the question of what kind of structure could most effectively fulfil this goal was not a settled one. Amnesty was organised in two, interrelated arms, with the International Secretariat (IS) in London incorporating the paid staff who carried out research and organising work, and the membership side which was organised into national sections, consisting of local groups, who elected representatives to the International Executive Committee (IEC) to direct the IS and oversee the implementation of decisions made at annual International Council Meetings (ICM) on their behalf. The relationship between the two arms, the professional IS and the membership side, was another source of tension, with both seeing themselves as the true heart of Amnesty.[20] But within the membership side, there were also questions as to how best to structure sections, how to create them, how to support them, and what the role of the IEC and the IS should be in this process.

These questions became more urgent when in 1977 Amnesty found itself facing a crossroads. Human rights were now becoming mainstream, the central organising principle of global politics.[21] This brought with it opportunities in the form of a massively expanded membership, but also new challenges in the form of managing both this expanding membership and the concomitantly expanding expectations placed on the organisation to shape global politics. Decentralisation, a term which had already been used to plan and structure the organisation’s publications and translation strategy, was floated as a way to manage these challenges. That year, a small group of mainly European leaders (one American was on the committee, while the sole appointee from the Third World, Suriya Wickremasinghe from Sri Lanka, had to temporarily suspend her work with Amnesty due to martial law being declared in the country), formed what they called the Crash Committee to strategize both the structural and the financial implications of the expected growth in the movement over the coming years.[22]

In the subsequent discussion, the question of power was central. One of the key members of the Crash Committee was Thomas Hammarberg, from the Swedish section (and who would soon take over from Ennals as Secretary General). Hammarberg was deeply committed to protecting what he saw as the essence of Amnesty, a certain spirit or ‘ethos’ that he considered under threat from growth in membership, staff and organisational profile.[23] He was also a strong proponent of decentralisation, which he saw as a way to diffuse power away from any one growing threat while preserving capacity at the centre to protect the ethos.[24] The Crash Committee agreed that by devolving routine tasks to the national sections the organisation’s centre would be more effectively able to control core issues like protecting organisational identity from change.[25] To this end, the committee argued for decentralisation with the proviso that it would ‘necessitate a strict observance of AI’s principles and procedures in order to ensure that a single AI identity is preserved and that AI speaks with one voice at all levels of the organisation in all parts of the world.’[26] Upon these recommendations, a new series of structures and regulations were enacted that aimed at ensuring a standardised and legible format for all national sections.[27]

Amnesty’s embrace of decentralisation mirrored the term’s rise within the broader development sphere. Strongly linked to the rejection of central economic planning and management, during the 1970s and 1980s the term became a way of framing political-economic interventions whose definitions remained vague yet invoked an implication of what they were not: centralised control, power and agency.[28] Despite this lack of positive definition decentralisation became a central ordering practice and theory of political institutions during this period. The open and undefined nature of the term meant it was embraced by a range of actors, from grassroots groups who saw it as a way of framing calls for direct democracy and citizen participation, to regional political elites looking to shore up their power base, to international institutions like the World Bank and the International Monetary Fund (IMF) who promoted it across the Third World.[29] And, like any ordering process, there was struggle over the political implications of the term and its processes in each of these and in other instances of its invocation.[30]

As Amnesty’s engagement with the concept shows, while it invoked a sense of redistribution power this was not an essential characteristic. As Nathan Schneider points out, ‘even the most apparently decentralised systems have shown the capacity to produce economically and structurally centralised outcomes. The rhetoric of decentralisation thus obscures other aspects of the re-ordering it claims to describe.’[31] Amnesty’s Crash Committee used decentralisation as a way to shield the organisation’s existing power structure from the impact of external change, institutionalising a certain ‘single AI identity’ in the fact of a diversifying membership. But this attempt to control change also sparked challenges. This discontent, however, remained articulated within, rather than against, the discourse of decentralisation.

Many parts of the movement challenged the structures and procedures established in the wake of the 1977 committee, arguing that they were interfering with the overriding and agreed upon goal of becoming a truly international organisation. The arrival of a significant sum of money, raised by the Norwegian section in 1984, galvanised these dissenting parts of the organisation to call for a new strategy committee to be convened to strategize how this money should be used. Called the Committee for Long Range Organisational Development (CLOD) this committee took the view that ‘to preserve what has been seen as the essence of AI, in the last decade especially, a plethora of sometimes rigid guidelines has been developed. Expressed by the slogan “one movement, one voice” … the effect, though never intended, is to limit severely the accessibility, flexibility, dynamism, therefore the diversity and effectiveness of the whole organisation.’[32] Yet this critique was of the limits of prior decentralisation, not of decentralisation itself, and the committee recommended moving forward by implementing horizonal rather than exclusively vertical relationships with ‘collaboration, cooperation and interaction in planning, in program and in the exercise of democracy.’[33] This created an opening for new theories of decentralisation. These new theories of decentralisation were forged in the actual practice of running national sections and local groups, practices which took place at the interface of Amnesty’s organisational rules and structures and the conditions across the world in which sections and groups operated.

The Crash Committee’s understanding of decentralisation shaped the creation of a set of minimum requirements for Amnesty national sections to adhere to in order to obtain and maintain ‘section’ status. This status, in turn, allowed them to participate fully in the organisation by sending delegates to the International Council Meetings, the annual membership gathering where policy was decided. These requirements were developed and monitored by the Subcommittee for Development (created in 1980 and renamed the Section Development Committee (SDC) in 1988), a body of the International Executive Council that coordinated the ‘newer’ parts of the international network, which were largely located in the Third World. But this coordination relied on documentation. In the early days of the Subcommittee, its members discussed the need to shut down sections from whom no correspondence had been received. Pakistan, Peru, Costa Rica, Cyprus, Nepal and Bangladesh were all assumed to be no longer functioning – or functioning outside of the rules – since no news had been received from them in quite some time.[34] Yet when an IS representative visited Pakistan with the intention of shutting down the section they found, to their great surprise, a vibrant and active group of members undertaking an impressive program of action in pursuit of Amnesty’s goals.[35] Given that it was not practically feasible to visit every section to monitor their work documentation took on greater importance, allowing the Subcommittee to make accurate evaluations and decisions.

Documentation allowed for the ‘time-space compression’ necessary for the Subcommittee to operate on an international scale.[36] The resulting rules and procedures stipulated that new sections report to the IS and the Subcommittee every month for the first six months.[37] This reporting was intended to demonstrate progress towards the various minimum criteria for groups and sections. But it invariably became something more, as these criteria were applied not purely in the space of the organisation but in the various physical locations of the sections and groups. As such, this documentation became a place where the interaction between these criteria and the conditions in which these groups and sections operated was documented.

One of the requirements for section status was that sections have a budget and a ‘plan for self-financing’.[38] For sections like the Sierra Leonean, this budget recorded the intersection between Amnesty’s rules and another set of rules and procedures: IMF-mandated structural adjustment.[39] With its requirements to reduce state presence and therefore expenditure and open the economy to the global financial system, this was itself another decentralisation initiative. These policies were implemented in Sierra Leone from 1986 onwards as a response to the debt crisis or ‘lost decade’ of the 1980s, characterised by both high levels of foreign debt and stagflation.[40] Yet this response led to further debt and stagflation, invoking more IMF support, which came with further conditions provoking further crisis and so on. AISL developed a series of strategies that allowed the section to survive the crisis.

These strategies drew on broader knowledge and practices developed in response to structural adjustment. Kinship networks had become important ways for ordinary Sierra Leoneans to navigate the socioeconomic instability that accompanied the crisis, and for AISL these networks also became important sources of cash to help tide the section over during periods of shortfall or when the Section Development Committee (as the Subcommittee was now known) was late in approving budget grant requests.[41] Another key challenge was inflation. Nationally, it stood at 124%, but for the types of goods and services that AISL required it was much higher: In 1991 Treasurer Samuel Sankoh reported that clearing charges for imported materials like newsletters had increased 340%, external postage (airmail) 1000%, and internal postage 2500%.[42] Yet in their 1992 budget AISL accounted for merely 30% inflation throughout the year, with the shortfall managed through a series of self-devised initiatives. Principal among these were keeping salaries fixed for as long as possible, as well as buying consumables like stationary in bulk at the beginning of the budget period, allowing savings to be shifted to cover the costs in areas where inflation could not be avoided, such as transport to the AGM and catering costs.[43]

AISL General Secretary David Gbao also opened a sterling-denominated account at a new bank, the Meridien Bank, which would allow them to maintain the purchasing power of their internal funding despite the wildly fluctuating Leone. Meridien Bank itself was a new bank that took its inspiration from pan Africanism and envisaged itself as a network ‘out of Africa, and for Africa’, while at the same time being a product of the very financial liberalisation that created the conditions AISL was struggling to deal with.[44] This dual-sided nature of the bank was emblematic of the complex way in which multiple decentralisation endeavours impacted each other, in this case with financial liberalisation and the structural adjustment it was a part of shaping the conditions in which Amnesty’s rules and procedures were applied and, in turn, contributing to the production of new knowledge that was used for organisational development.

This knowledge production was also in large part due to Amnesty’s own approach to decentralisation after the Crash Committee. Much of the financial literacy that supported this knowledge could be traced to Gbao, a highly motivated and organised individual who had joined what was then only a group of Amnesty supporters in 1982. In 1987 he became their volunteer chairperson and led the process of formalising into an Amnesty section. But in early 1991 he took a full-time job in a local bank, leaving him with significantly reduced time to invest in AISL activities.[45] Concerned about the effect that this would have on the Sierra Leonean group and its prospects for gaining official section status Gerson Gu-Konu, who worked at the IS an organiser for the African region, convinced Amnesty to create the paid position of General Secretary in Sierra Leone and offered it to Gbao in order to recover the immense organisational capacity that he brought to the section. In doing so, Gu-Konu inadvertently enriched the section’s financial literacy, an equally crucial skill for section survival. Gbao also brought in other members with these skills, like Sankoh, who was a colleague from his time at the bank.[46]

This decentralised exercise of financial authority was not sanctioned by Amnesty’s rules and procedures. Rather than hide it, however, or simply leave these manoeuvres undocumented, the section assiduously documented this knowledge in their budgets, annual reports and other associated paperwork, even when this required additional work to shoehorn the specifics of the situation into the standardised forms. They then circulated this knowledge in a specific direction: to the SDC and to Dieleman herself. The direct communication with Dieleman in particular functioned as an alternative strategy for scaling up their ideas, enlisting her as the Africa-focused committee member to advocate for them and their capacity to act as a fully fledged section. Following a meeting with her in Amsterdam Gbao carefully wrote and then re-wrote his minutes of the meeting, ‘so that the points are better presented’.[47] In these notes he took the time to explain the economics of currency exchange to the SDC, outlining for them how the instability of the Leone gave a distorted picture of the section’s ability to manage finances and encouraging them to look at costs in pound sterling to get an accurate picture of the economic situation, both in Sierra Leone and elsewhere.[48] Circulation, then, was a key aspect of documentation that sought to allow the documented ideas and knowledge to have an impact.

The impact it sought was the thickening of the ties between the organisational centre and the national section, an objective that no doubt emerged with under the awareness that sections could easily be earmarked for closure. Indeed, this was the basis of the Crash Committee model of decentralisation, where the administrative centre held the power to decide if the national sections were able to remain part of the organisation. Under this model, new sections remained in a precarious situation, their status ‘unsure, tenuous, awkward, problematic, revocable.’ Their knowledge, then, was ‘precarious knowledge’, easily lost or forgotten.[49] Budget reporting that documented local, specific challenges but that put these in terms of the SDC could understand and that was circulated in ways that aimed at maximising their reception aimed at removing the precarity of new section status. In doing so, it aimed at creating a strong institutional connection that would allow this knowledge’s truth claim to be considered valid, strengthening the organisation’s ability to expand internationally.

Yet knowledge circulation requires reception for it to have an institutional effect, and the Sierra Leoneans could not control this. AISL used the budget reporting and associated correspondence to request that budget grants and other funding sent from the IS be deposited in sterling into the Meridien account Gabo had opened.[50] But the SDC regularly ‘forgot’ to make its transfers into this account, a mistake they said arose from the fact that the section did not use the correct paperwork to inform them.[51] Despite the fact that AISL repeatedly asserted that their work devising methods to combat the effects of inflation was aimed at improving the organisation’s bottom line, there was conflict over which kind of documentation – which kinds of knowledge circulated in which ways – could be considered to bring about an effect within the organisation.

Rather than a once-off mistake, this lost information was part of a broader pattern of institutionalised forgetting that spoke to the difficulties in circulating knowledge and scaling it up. In addition to developing a budget, the rules and procedures required sections to work towards being financially self-sufficient. To achieve this, they were asked to engage in fundraising. But this presented practical challenges for sections like the Sierra Leonean. The domestic economy, AISL repeatedly explained in annual reports, was simply too tight and Sierra Leoneans did not have money to donate or spend on Amnesty materials.[52] And while they recognised the importance and value of traditional Amnesty forms of fundraising and communications such as the distribution and sale of newsletters and other publications, they also explained that high levels of illiteracy in the country meant that this strategy had serious limitations.[53] They used their reporting to educate the SDC about the socio-economic conditions in the kinds of places the organisation wanted to be active, highlighting that sections – and therefore Amnesty as a whole – didn’t operate outside of these conditions.

While the Sierra Leoneans used documentation to thicken and strengthen the relations between their own section and the organisation itself, they did not necessarily envisage an organisation that was structured hierarchically, with relations emanating out from the centre to national sections. They also used documentation to circulate examples of how a decentralised organisation might be structured. The version of decentralisation that they practiced was in line with Frederic Cooper’s description of globalisation as ‘filled with lumps, places … where social relations become dense amidst others that are diffuse.’[54] This was the reality of Amnesty’s internal network: a mixture of thick and thin ties criss-crossing the organisation, emerging and fading at different moments. Despite the best efforts of the organisation to standardize and consolidate its international presence, events in different parts of the world meant sections would be shut down, or go into a period of reduced activity, or experience efflorescence outside of the control of organisational rules and procedures. While the CLOD had gestured towards horizontality, it had still operated with an assumption that the space of the international organisation was hermeneutically sealed, existing independent of other spaces, giving it the ability to achieve homogeneity, with all ties of equal and ordered strength. In documenting that this was not actually the case, the Sierra Leoneans made an argument for the utility of lumpy ties as a model of decentralisation.

Horizontal ties, in the form of section-to-section collaborations and cooperation, had underpinned Amnesty’s growth for many years. Even within Europe certain sections owed their start to the involvement of others. During the early 1970s, for example, the Belgian Flemish section had relied heavily on resources and support from the neighbouring Dutch section to build up a local presence.[55] This kind of ad hoc mutual assistance became more formalised following the Crash Committee’s recommendations and subsequent creation of the Subcommittee for Development. Along with this formalisation came a conceptualisation of assistance flowing in one direction, from larger and medium sized section to smaller ones still in a precarious organisational position. Within the rules governing this assistance the IEC was required to approve all proposed initiatives, and assistance was to be purely the provision of human rather than financial resources.[56]

The Sierra Leoneans fostered several collaborations with larger sections. In partnership with the British section they created a calendar, designed by a Sierra Leonean member and sold across the UK. Another exchange involved sending traditional Sierra Leonean gara clothing to the Swedish section, who would then facilitate its sale. In the pilot trial of this project, they sold 50 items at the 1990 Swedish AGM.[57] Buoyed by the success of this pilot initiative, AISL planned to expand the sale of gara clothing across Europe, with a plan to work with around ten different European sections, who would act as retailer.[58] This initiative evidenced AISL’s multi-dimensional understanding of global trade and finance. At a basic level, objects produced in Sierra Leone could be sold for a higher price in the European market. More specifically, however, fair trade consumerism was an increasingly significant market that the Sierra Leoneans, in partnership with the British and the Swedish, knew to exploit. The postwar period had seen consumption become a key focus of western European social movements, with major campaigns encouraging the public to act around global inequality through fair trade, the purchase of equitably traded and renumerated goods from the global south by northern consumers.[59] The Sierra Leoneans saw an opportunity to engage with these new trade circuits, upending the flow of expertise envisaged in the rules governing inter-sectional assistance.

In addition to bringing financial benefits to the organisation, these engagements also served to connect Amnesty to a broader public. As Benjamin Möckel has argued, fair trade helped bring human rights out of the sphere of international law and foreign policy and into the everyday materiality of civil society.[60] By providing items that the European sections could sell, the Sierra Leoneans helped expand the sphere of engagement with Amnesty in Europe, a complete reimagining of how inter-sectional collaboration contributed to organisational development.

This engagement was with the broader public that now imagined and engaged with the world through their consumption, as well as with other organisational networks. Following discussions between the Swedish section and Gbao at the 1991 ICM the former sent member Lennart Asker, who also worked for development organisation Diakonia (formerly Swedish Free Church Aid), to Sierra Leone to advise on a new fundraising project the Sierra Leoneans were proposing. Indeed, this connection had also occurred when Gbao had opened the foreign currency account with the Meridien Bank, connecting AISL to the resources of entrepreneurial pan-Africanism. This version of decentralisation practiced by the Sierra Leoneans further upended the organisational understanding of where resources came from and how they flowed throughout the organisation, demonstrating the interdependence between Amnesty and others for internationalisation.

The opening for new theories of decentralisation created by the CLOD was not only filled by ‘section’ actors like the Sierra Leoneans. Amnesty’s ‘centre’ was a diverse body of staff and officials, many of whom by the 1980s came from Asia, Africa and Latin America. Among many examples, Chilean José Zalaquett worked at the IS after having chaired the IEC for several years at the beginning of the decade. Senegalese Bacre Waly Ndiaye was on the IEC in the later part of the decade, while Togolese Gerson Gu Konu had been the staff organiser (officially called the Membership Development Coordinator, MDC) for Africa since the late 1970s.[61] Gu Konu in particular had a clear vision for how he saw organisational development in Africa proceeding.

Gu Konu was recognised throughout the organisation for his hard work: one IEC member joked that if he claimed all his overtime he would bankrupt the entire organisation.[62] This dedication was expressed as a strong belief in the role of MDC as the one in a position to structure the organisation across Africa. His approach was not necessarily shared by the rest of the staff or IEC members, who at times considered it to be overly top-down.[63] But for Gu Konu, who as an organiser needed to be able to create a shared vision in order to mobilise membership development, the main audience for his version of decentralisation were the African sections themselves. Even with the organisational status of MDC, his ordering project needed to be accepted as true by the membership if it was to be implemented. This involved documenting and circulating these ideas.

Gu Konu was well aware of the important role that documentation played in ordering Amnesty on the continent. Even while in the field, attending to other sections, he composed lengthy communications that aimed to persuade the Sierra Leoneans away from their approach to decentralisation. This was no simple activity. While on location in Tanzania, for example, he composed a lengthy letter to the Sierra Leoneans. He then relayed this letter to the typing pool at the IS, which wrote it up and sent it on to Freetown.[64] An eight page missive, it covered a wide range of issues, many of which were not urgent or administrative but that spoke to the conflict between visions of ordering occurring between the two.

Much of this conflict centred on inter-sectional assistance. Gu Konu argued that ‘a strong/lasting/credible/truly respected S. Leonean Section of AI can come into being only through well thought-out/coherent/consistent/disciplined approach and methods’ and not the ‘superficial and incoherent international relationship[s]’ that they had fostered based on trade.[65] He believed that only by following the existing organisational processes to the letter could African groups become legitimate parts of the organisation, operating on an equal footing with the more powerful European sections. For him, it was through equal treatment, complying with the same rules and regulations as everyone else, that the African sections could claim their place at the table.

The Sierra Leoneans vociferously pushed back against the idea that the relationships they formed with the Swedish and other sections were superficial and inevitably dependent. The problem, as they put it, was that ‘to date the movement (AI in a collective sense) has failed to formulate appropriate policies/mechanisms to consolidate fundraising in the weaker sections through cooperation between Sections without necessarily developing undesirable dependence of weaker Sections to stronger ones.’[66] For them, relations between richer and poorer sections were beneficial to both and to the organisation itself. In correspondence with the Swedish section, a carbon copy of which was to the SDC, Gbao explained in detail his vision for internationalisation fostered through horizontal connections and exchanges:

Any cooperation should necessarily be of mutual benefit to the parties involved. The individual gain may be quite different and unrelated but are in essence vital for the development of Amnesty International as a truly international organisation. When we visit other sections for example, we benefit in that our members get experience and are in effect exposed to modern ways of doing things … In the larger/richer sections, members get a deeper understanding of how things are in the smaller/poorer Sections and get to know that despite our circumstances we from the smaller/poorer Sections/Groups are quite capable of contributing to international debates and representing AI to the public.[67]

Rather than pure independence as an antidote to the unequal power that resulted from uneven levels of material resources, Gbao proposed a re-evaluation of what resources were considered valuable for the task of internationalisation and a fostering interdependence based on exchange of those resources.

Gu Konu’s willingness to invest time and energy in these lengthy communications while undertaking a hectic schedule on the road indicated the importance of documentation in the process of membership and organisational development, even for staff members. He occupied a more powerful position in the organisation and could control the activities of the Sierra Leoneans by withholding funding or ejecting proposed initiatives. In 1990 he quashed AISL’s planned visit to the Mauritius groups’ AGM, saying he felt ‘strongly that the Mauritian membership and you (the S.L. membership) have not had the necessary time to prepare the scheme properly.’[68] But he was aware that his role was to create order, not simply prevent other orders from consolidating. Documentation allowed him to use the occasion of rejecting proposals to circulate his own vision of an order built on careful, central planning.

The competition between Gu Konu’s and the Sierra Leonean’s visions of order directly challenged each other’s positions as legitimate holders of knowledge. The Sierra Leoneans distributed a proposal to the entire membership for a ‘reinvigorated African Regional Meeting’ (ARM), ‘an essential structure for the sharing/exchanging of views among AI Sections and Groups in Africa as well as initiating and developing realistic strategies for contributing to the international planning and decision making processes.’[69] In this proposal they directly challenged Gu Konu’s capacity to conduct organisational development, arguing that the ARM offered the chance be a ‘major way out of our problems of developing AI in Africa.’ These problems, they argued, stemmed from ‘the Africa Membership Team (more specifically the MDC) … through excessive interference and virtually directing the ARM.’[70] While Amnesty had long encouraged regional groupings as forums for skill sharing (and as a way to rationalise policy debate and decision making), AISL was proposing the removal of paid staff like Gu Konu from these forums to make them grassroots – and effective – spaces.

Gu Konu, of course, disagreed. Formal structure, rather than knowledge exchange, was what was most needed, with it being ‘crucial for the AI membership in Africa to possess an organised and efficient servicing structure i.e. a national office.’[71] Alternative visions were based on ‘amateurish ideas/understanding of “development”’ and on people pursuing ‘their individual/personal interests which, unfortunately very often, have either nothing or very little to do with the real interests of the AI movement.’[72] This effort to mark the Sierra Leonean’s ideas as precarious, unable to lay claim to truth die to their unstable and ill-informed basis, became the key way that Gu Konu could use his position to assert control over their reception across the organisation.

But the connections that had been forged between sections through collaboration meant that these strategies had limited success. The SDC, guided by their conceptualisation of resources flowing from the larger, more powerful sections down to the smaller ones, took it upon themselves to begin corresponding directly with the Swedish section when it wanted to discuss one of the Sierra Leonean’s fundraising projects that the Swedes had offered support to in the past. The SDC would approve it, they sad, as long as the Swedes remained closely involved in its execution, essentially in a supervisory role. This attempt to assert a certain organisational order was met with indignation on the part of the Swedes. ‘This,’ they wrote back, ‘is not development policy, it is cowardness.’[73] That the Swedes, traditionally one of the most conservative sections within the organisation, so forthrightly challenged the SDC’s own competence demonstrated the ‘strength of weak ties’ across the organisation, the very ties that the Sierra Leoneans were highlighting as crucial tools in organisation building.[74]

The Swedes’ actions speak to the diffuse and unexpected ways that actors mobilised their power within the organisation in the context of the opening to ideas around decentralisation and order. As the example of Gu Konu – an African within the IS – shows, there was not a clear-cut layering of North-South divisions onto the different visons of decentralisation. Likewise, relationships traversed North-South lines, shaped by unexpected cultural connections. The Sierra Leoneans’ strong relationship wit the Swedes, for example, was partly facilitated by the mobility that Amnesty’s meetings and congresses occasioned. But the relationship was particularly strong because of the Swedes’ practice of billeting people in members’ homes when they hosted international Amnesty meetings. This facilitated an intimacy of contact that did not develop in other places like Britain or the Netherlands, where visitors tended to be put up in hotels.[75] Hospitality, then, became important way of building ‘affective social worlds’ that connected unlikely places and sections within the organisation.[76] These connections, in turn, became the basis of the ideas and practice of decentralisation documented by the Sierra Leoneans.

In his ethnography of Amnesty’s International Secretariat, Stephen Hopgood quotes a former IS staffer looking back at the CLOD with a certain sheepishness. ‘The whole starting premise was wrong,’ they say. ‘I mean, look at the name. The committee for long range organizational development. It was about Amnesty. It was about building Amnesty … it didn’t say, you know, the committee for more effective strategy for ending human rights violations.’[77] But what this reflection forgets is that Amnesty saw organisational development as directly underpinning any effectiveness they could have in shaping state behaviour and ending human rights violations. Both the leadership and the membership of Amnesty agreed that it was only through incorporating a membership in the Third World that the organisation could claim the mantle of ‘international organisation’ and therefore act as an institutional part of the post-war liberal order. The CLOD was driven by debates between the leadership and membership on how exactly this incorporation should be structured.

As a membership organisation, Amnesty offers a unique chance to understand the ways that non-state actors have attempted to shape global politics and the ‘forgotten’ possibilities behind these attempts. Communications, central to global space making, occurred largely though documents that were created and circulated in specific ways and with specific content. The documents produced by the Sierra Leoneans at the heart of this article were and handful of the great many that circulated within the organisation, creating connections while also transmitting ideas, reflections and questions regarding those connections.

The ordering of relations that occurs within organisations, like that which takes place during archiving, creates certain worlds, yet this ordering and classification that occurs is often an ‘unruly and piecemeal venture at best.’[78] Other visions, other ‘worlds’, are contained within these organisations (and within these archives). Focusing on forgotten possibilities also requires recognising that forgetting is an active practice. Power within the organisational or institutional structure is unevenly distributed and the knowledge of those with less power is the most vulnerable to loss either from being pushed underground by those with more power or through lack of access to the resources for recording and preserving it.[79] In the former case, as I have shown, struggles within the organisation over ordering saw alternative proposals pushed away. This practice is most visible at conferences, congresses and large meetings where resolutions are voted down or perhaps never tabled and where physical attendance is subject to a host of factors, contextual and structural.[80] But it is also evident in documentation, where agendas were prosecuted, connections made, and orders asserted both within the content of the documents and in the ways they were circulated.

In the latter case, the unruly and piecemeal nature of archiving itself means that the power to shut out certain knowledge is never total. As Carolyn Steedman points out, ‘the archive is made from selected and consciously chosen documentation from the past and from the mad fragmentations that no one intended to preserve and that just ended up there.’[81] This article draws on both ‘ordered’ archived documents, and some of these mad fragments to examine another attempt at ordering and the forgotten or excluded responses to it.

While ordering is central to organisations generally, the context that I examine speaks to a specific framework of reordering that became hegemonic from the 1970s onwards. Decentralisation, while it has a longer history, experienced a rise in popularity at this time, embraced by a wide range of organisations and institutions. Yet the specific order that the term refers to, and the location of power within that order, is not self-evident. By looking at struggles over ordering in the context of decentralisation, I aim to contribute to building up an understand of the different ways that actors understood and theorised decentralisation. Adom Getachew’s great contribution with her mobilisation of worldmaking is to open up ways of seeing new actors as theorists of international order in their own right.[82] I take this contribution, and the provocation to consider the possibility of non-state actors also as theorists of international order, has been at the basis of this articles claim that the Sierra Leoneans, among many others, advanced serious theories and practices of decentralisation that aimed to shape the international organisation and, by extension, enhance its effectiveness on the world stage.

Acknowledgements

This paper has benefited from the comments of the special issue editors, Sneha Krishnan and Jake Hodder, and the participants of the Archives of Internationalism panel at the 2023 American Association of Geographers annual meeting. Thanks to them and the anonymous reviewers for their generous engagements with earlier versions of this paper. Recognition of funder

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101031140.