The myth of humanitarian action supported by data and scientific evidence has had its day – or so argues the author here, drawing on the history of the most widely known standards in the sector. He does not refute the essential role that science can and must play within it, provided it does within the framework of a permanent dialogue with the stakeholders and not on the basis of the exploitation of science.
Humanitarian agencies love to mimic the language of science. Quantified indicators, technical jargon and percentages given to the nearest decimal point all jazzing up reports and press releases. The use of technical language is nothing new, but the recent growth of technology (mobile telephones, satellite imagery and machine learning) has given it a new dimension. Since the Secretary-General of the United Nations (UN) called for a true “data revolution”, the positivist discourse of “evidence-based” humanitarian work has become an official lineUnited Nations General Assembly Economic and Social Council, Strengthening of the coordination of emergency humanitarian assistance of the United Nations: report of the Secretary-General, … Continue reading.
Obviously, this discourse does not convince everyone. Many critics highlight the contradictions concealed behind the technical knowledge facade, arguing that coloured computer graphics in fact disguise uncertainties, approximations, and even manipulationsRóisín Read, Bertrand Taithe, and Roger MacGinty, “Data hubris? Humanitarian information systems and the mirage of technology”, Third World Quarterly, vol.37, no.8, 2016, p.1314-1331; Daniel … Continue reading. The harshest critics believe that humanitarian statistics are more political than knowledge-based, with some even seeing an avatar of the neoliberal project behind this craving for figures.
However, the debate on the value of humanitarian knowledge is also turning towards a deeper, epistemological issue, namely the relationship between knowledge and power. Does humanitarian knowledge come from politics or academic knowledge? As a matter of fact, humanitarian knowledge brings together academic knowledge and politics in a very specific way.
We need to closely observe the origin of humanitarian knowledge in order to gain an understanding of how this works. We will be looking at the “Sphere standards”, for they are undoubtedly the most well-known humanitarian standards, as Sphere is used by hundreds of humanitarian organisations, non-governmental organisations (NGOs), funding bodies, governmental and intergovernmental agenciesThey are grouped together in the Sphere Handbook. Humanitarian Charter and Minimum Standards in Humanitarian Response, published by the Sphere Project for the first time in 2000. The fourth and most … Continue reading.
The Sphere standards – a paradigmatic example
The Sphere standards aim to ensure that the services provided to victims in crisis situations (political crises, or in the aftermath of a natural disaster) achieve a minimum acceptable level. They define a “minimum for survival”Doctors Without Borders, Performance Standards in Humanitarian Relief. Project Proposal, International archives (box 192), Geneva, September 1996.. For instance, it should be ensured that “all people have safe and equitable access to a sufficient quantity of water for drinking, cooking and personal and domestic hygiene”. This involves making certain that “at least 15 litres of water per person and per day are collected”All the quoted indicators and standards are taken from: Sphere Project, The Sphere Handbook. Humanitarian Charter and Minimum Standards in Humanitarian Response, first edition, 2000.. Likewise, at least “2,100 kcal per person”, including “10-12%” of protein, “17%” fat, and “adequate micronutrient intake” need to be provided. The Sphere standards set the minimum living space in a tent that should be available for each person (3.5-4.5 m² per person), the maximum number of people using communal toilets (20), and the amount of water needed for handwashing (1 to 2 litres per person per day).
“Sphere’s advocates stress this idea of modern, objective and rational humanitarian aid.”
Consequently, science is at the heart of the story put forward by the NGO Sphere ProjectA good overview of the project’s official line is found in the video clips published on the Sphere Project website: Sphere Project, The Sphere story, a video documentary, 30 June 2015, … Continue reading. This official story has two distinct stages: firstly, the NGO gathered together the most in-depth possible technical knowledge on human needs. It then got a large number of NGOs to agree on this knowledge, which was regarded as being objective and beyond question. In the official story of Sphere, technical knowledge helped everyone to agree. Knowledge made the humanitarian consensus possible. It brings an end to chaos and arbitrariness. Sphere’s advocates stress this idea of modern, objective and rational humanitarian aid. It would be ushering in a radically new era for humanitarian work. They speak of a “radical” invention, a “landmark” watershed, and the “end of the age of humanitarian innocence”Sphere Project, Sphere standards: Radical but inevitable. Interview with Peter Walker, 21 October 2012, … Continue reading. The language used is deliberately borrowed from the history of scientific discoveries: the members of Sphere speak of a “revolution” and a real “paradigm shift” in line with “Kuhn’s thesis of a paradigm”John Borton, Sphere Project, 2015; Margie Buchanan-Smith, How the Sphere Project came into being: A case study of policy-making in the humanitarian aid sector and the relative influence of research. … Continue reading.
However, this story does not stand up when reading through the archives. As luck would have it, the first Sphere handbook was written between 1996 and 2000, at a time when email was already used as standard, but people still took the time to print and store hard copies of emails. This means that today we can read the thousands of messages that preceded the publication of the Sphere standardsThe archives were held in the basement of the International Council of Voluntary Agencies, in Geneva, when I had access to them. I wish to thank the Sphere Project management team for allowing me … Continue reading. A very different tale to the official story then comes to light.
We can see that the Sphere standards are not the result of academic research. Or more accurately, they are not solely the result of it. Technical knowledge was not absent from the debates, but at no time were technical knowledge and politics stringently separated. The real story of Sphere is that of a series of negotiations between humanitarian workers and experts, with an array of proposals, criticisms and counter-proposals, which admittedly make use of academic arguments, but also and simultaneously utilise political, religious, cultural arguments, etc. Power struggles were always a part of the negotiations. At no time was academic knowledge cut off from these power struggles. Academic and expert knowledge was most definitely present, but it did not precede the non-academic arguments; instead it accompanied them.
Science as a tool of legitimisation
The recourse to technical knowledge took place in a specific context. In 1996, when Nicholas Stockton and Peter Walker launched the Sphere Project, the primary aim was to restore the legitimacy of humanitarian work. The intention was to guarantee a certain service level to both funding bodies and aid recipients. Ultimately, this guarantee should enable the good NGOs (those that comply with the standards) to be differentiated from the bad ones (those not adhering to the standards, or that do adhere to them but are unable to attain them due to a lack of professionalism).
Humanitarian organisations were being heavily criticised at the time. The humanitarian intervention in Rwanda had become a debacle. Many organisations were flaunting their incompetence to the mass media: misappropriation of food rations, antibiotics being handed out in unruly conditions, injections being applied with insufficient safeguards, and ignorance of local languagesLindsey Hilsum, “Reporting Rwanda: The media and the aid agencies”, in Allan Thompson (ed.), The Media and the Rwanda Genocide, Pluto Press, 2007, p.167-187.. An international report concluded that 100,000 deaths could have been prevented if aid had been more effectiveJohn Borton et al., The international response to conflict and genocide: Lessons from the Rwanda experience. Study 3: Humanitarian aid and effects, Joint Evaluation of Emergency Assistance to … Continue reading. The major funding bodies therefore significantly cut their humanitarian spending. Furthermore, at the same time, competition for funding exploded because the creation of more NGOs saturated the market. The large organisations quickly grasped the danger and wanted to increase guarantees of their professionalism, developing ethical codes, oversight, audit and assessment mechanismsFor example, the Turku Declaration (December 1990), InterAction’s Providence Principles (1993), the Mohonk criteria (1993), the People in Aid Code of Good Practice (1997) and the Red Cross Code … Continue reading. In the eyes of Stockton (who was working for Oxfam at the time) and Walker (an employee of the International Federation of Red Cross and Red Crescent Societies), technical knowledge should enable stakeholders in the humanitarian sector who were worthy of funding bodies’ trust to be differentiated from those who were not. They dreamt of an ISO-type quality standard applied to humanitarian work. Any NGO that failed to meet the standard would be barred from accessing international funding.
“Academic and expert knowledge was most definitely present, but it did not precede the non-academic arguments; instead it accompanied them.”
Stockton and Walker gathered together some fifteen or so major humanitarian organisationsThe Sphere Project’s Governing Board brings together fourteen major organisations or NGO alliances: Oxfam, InterAction, Save the Children Alliance, CARE, Doctors Without Borders, Lutheran World … Continue reading and then sought funding, which was quickly provided by ten or so major funding bodiesThe leading backers are Australia, Denmark, Netherlands, United States, Sweden, Switzerland, France and the European Commission’s Directorate-General for European Civil Protection and Humanitarian … Continue reading. They then hired a handful of managers to draft the handbook. These managers consulted experts and academics. Most of the work was undertaken by a core group of six or seven managers who were close to Stockton and Walker: this group was responsible for drafting the handbook. Around them, a first circle of a few dozen NGOs and funding bodies gave them funding, information and expert knowledge. Finally, a third larger circle was made up of several hundred experts, researchers and consultants approached to provide an opinion on a specific issue, chapter, or the whole handbook. The Genevan team could therefore draw on a vast network of individuals working in highly diverse organisations (NGOs, universities, laboratories, think tanks, governmental institutions, etc.), while retaining control over the final version.
A series of “furious negotiations”
The Genevan team made the proposals, and then sought feedback from the experts. The feedback was often understated, sometimes enthusiastic, and even passionate. It was often far-removed from the cold and detached tone that one would imagine in technical discussions. The team was in a series of talks that Nicholas Stockton himself would brand as “furious negotiations”.
For example: how should the amount of food needed to survive be defined? An expert suggested 1,900 kcal per person per day as a minimum energy requirement. Another expert proposed 2,300, and another 2,070. The International Committee of the Red Cross recommended 2,400 kcal, while the World Food Programme said 2,100. A decision needed to be made and the Sphere managers set the threshold at 2,100 kcal but specifying that this figure equates to a theoretical average calculated on the basis of a sedentary population with “normal demographics” and an “ambient temperature above 20°C”Sphere Project, Progress Report for Sphere Project Second Quarterly Report. Nutrition Sector Quarterly Report, archive box no.2, October-December 1997; Sphere Project, Sphere Handbook…, op. cit., … Continue reading.
“Each positive or negative decision on an indicator was simultaneously a decision in favour of a given organisation.”
How much water is needed for survival? An expert suggested 2 to 3 litres per person per day as a minimum water requirement. However, another said 20 litres. The first expert focused on drinking water, while the second included cooking and personal hygiene requirements. The figure of 15 litres was chosen. This figure is a compromise, high enough to require effort but low enough to be realisticSphere Project, Comments on the draft by JA, archive box no.2, January 1998; Satterthwaite, 2011, p.922.. For a proposal to prevail, it needed to not only be based on technical rationale; it also needed to be championed by an institution. Each positive or negative decision on an indicator was simultaneously a decision in favour of a given organisation. What is the minimum living space required for a refugee? The World Health Organization recommended 30 m² of shelter per person, while the UN Refugee Agency allowed 150 m² per household55 m² of living space for a family, including a garden area. Sphere Project, Progress Report on the Sphere Sector: Shelter and Site, archive box no.2, December 1997., so choosing an indicator meant choosing a camp.
At no time could the humanitarian stakeholders allow themselves to put politics to one side. Stockton and Walker suggested that the Sphere standards should become a universal right for crisis victims, and a legal guarantee of obtaining certain services. Unacceptable, responded an American funding body (in this case the Office of U.S. Foreign Disaster Assistance, a United States Agency for International Development entity), and the US NGOs stated that an “entitlement” would lead to endless legal proceedingsSphere Project, Email, archive box no.10, 7 July 1997.. The issue of religion also turned up in the discussions. The specialised NGOs wanted sexually transmitted diseases and family planning to be included in the handbook, but several religious NGOs were steadfastly opposed to itSphere Project, Email from International Pregnancy Advisory Services (IPAS), archive box no.2, 6 July 1998.. The subject of abortion was carefully avoided. Condom distribution was an issue. Finally, the handbook ended up recommending the use of condoms, but not without Caritas Internationalis, headquartered at the Vatican, ensuring that specific reference was made to the fact that this was not an unanimously agreed standardSphere Project, Minutes of the Sphere Management Committee Meetings: 22-24 Sept. 1999, Geneva, archive box no.10, September 1999..
The project almost fell apart at each stage. Stockton and Walker wanted to make the Sphere standards binding, get them adopted by the UN General Assembly, and ostracise any NGO that refused to adhere to them. However, a group of French NGOs opposed this stanceDoctors Without Borders, Fax by Action Against Hunger, Institut de l’Humanitaire, Doctors of the World, Doctors Without Borders – France and Groupe URD for the Sphere Project, archive box no.197, … Continue reading. They explained that these standards were not universal, and only worked in some specific situations, in refugee camps, but were not appropriate in every setting.
Matters had to be discussed and negotiated, and compromises had to be reached at each stage. Numerous articles, faxes and emails were sent, position papers and reports were produced, and meetings were held behind closed doors in Geneva, London and New York. Each step forward in one direction was a step backwards in another. Each concession made to an NGO called into question the backing of another NGO. If Stockton and Walker became closer to Doctors Without Borders (Médecins Sans Frontières) – France, the Americans from InterAction were then likely to leave the project. If they drew closer to Action Against Hunger, the Lutheran World Foundation threatened to leave. In other words, at no time did the experts really separate academic knowledge from politics.
Does this mean that the indicators are purely arbitrary? No. All the Sphere participants vindicated their stances by referring to evidence and references, notably academic evidence and references (leading academics, laboratory research, specialised journals). However, the arguments were never solely academic. The experts alluded to reason and strength, academic and diplomatic arguments, the opinion of researchers and the opinion of funding bodies.
From evidence-based humanitarian work to reflective humanitarian work
The Sphere Project’s mode of veridiction is not similar to a scientific mode of veridictionBruno Latour, La fabrique du droit. Une ethnographie du Conseil d’État, La Découverte, 2004, p. 207-260.. Instead, it is similar to what sociologists call decision-making “by apparent consensus”Philippe Urfalino, « La décision par consensus apparent. Nature et propriétés », Revue européenne des sciences sociales, vol. 45, n° 136, 2007, p. 47-70.. This method of decision-making does not involve unanimity or voting (votes are not counted). A decision is taken at the end of a long discussion process aiming to incorporate, to a certain extent, the disagreements, and then coming to an end when minority dissent falls silent. The decision is adopted when there are no longer any overt objections. Not everyone agrees with the decision, but there is a lack of evident opposition to a proposal. In the case of Sphere, disagreements continued to be expressed, but the project was successfully completed (the handbook was published), drawing on the involvement of a large number of stakeholders. For this to happen, the Sphere team had to manage the context of the debate (setting deadlines, raising questions, and retaining control of the final version, etc.) and agree to yield on the most controversial issues (removing an entire chapter from the handbook, taking out the concept of an “entitlement” to assistance, the idea of an ISO standard, and abandoning the idea of getting the text adopted by the UN General Assembly). On the other hand, these concessions enabled the number of project participants to be spectacularly increased (641 individuals took part in the first Sphere manual, and over 4,000 in the second edition). However, it is clear to see that this very broad consensus did not stem from an agreement on technical indicators. From a chronological perspective, the opposite actually happened: a consensus emerged first, and the indicators came second.
“It is fanciful to expect academic knowledge and technical know-how to solve conflicts in the humanitarian sector.”
We shall leave it to humanitarian professionals to judge the quality of the Sphere standards. The intention here is not to side with any specific technical instrument. Instead, the aim is to reiterate the inherent difference between technical knowledge-based decisions and humanitarian decisions. These are two different systems of truth, each with their own rationale. It is clear that technical knowledge can be sought by humanitarian agencies. However, it is fanciful to expect academic knowledge and technical know-how to solve conflicts in the humanitarian sector. Expecting technical progress to resolve the humanitarian sector’s thorny issues is confusing realms. This would be cultivating the illusion of an infallible and watertight humanitarian sector. Furthermore, close observation of humanitarian decision-making shows that discussion is at the heart of the process.
An improvement in the quality of the decisions taken therefore comes from improving the quality of the discussions. One of the major challenges involved in addressing the myth of humanitarian work based on big data is ensuring that the sector has healthy discussion mechanisms; discussion in which academics and experts obviously have a place. Indeed, rather than fantasising about a new form of evidence-based humanitarian work, it might be better to champion humanitarian work based on quality deliberation.
Translated from the French by Gillian Eaton
This article is the outcome of research conducted within the Africa Multiple Cluster of Excellence at the University of Bayreuth, funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC 2052/1 – 390713894
ISBN of the article (HTML) : 978-2-37704-913-4