Words of Relief: Translators without Borders' Local Language Translation for Emergencies

"Overall, the innovation process has been successful in creating enhanced learning and evidence around the importance of language translation in disaster response, and has succeeded in producing a measurably improved mode of communicating with affected people."

This case study examines Words of Relief, a Translators without Borders (TWB) project designed to provide local language translation services to non-governmental organisations (NGOs), United Nations (UN) agencies, and other actors during humanitarian response. The project includes 3 strands: (i) an online multilingual library of location-specific disaster messages translated before a crisis into local languages, to be openly disseminated through digital platforms; (ii) a spider network of professional diaspora and community-based translators who would provide voluntary rapid translation services through an online platform; and (iii) the Words of Relief digital exchange: an online platform funded by Microsoft Technology for Good that facilitates translation of content generated from the community affected by the disaster, via social media networks. The tool was also designed to incorporate a library of 300 translated messages in 18 languages, including the top disaster terms, medical terms, SMS messages, and tweets.

To understand how this transpired, the research team conducted a review of project documents and 11 interviews with key project and partner staff in early 2015. The resulting case study uses a model based on 5 stages:

1. Recognition of a specific problem, challenge, or opportunity to be seized - A number of translators with informal links to TWB were active in Haiti when the earthquake struck. Despite French being an official language, the affected population mainly spoke Haitian Creole, and aid workers were unable to communicate effectively. Shortly afterwards, between 2011 and 2013, several TWB members were involved in crisis mapping exercises with the crowd-sourcing platform Ushahidi in Kenya and the Digital Humanitarian Network (DHN) in the Balkans and northern India. During this phase, TWB identified 2 key challenges to providing local language translation in humanitarian scenarios: (i) Language is intimately linked with the location of a disaster; and (ii) volunteers would be linked to the location of the crisis; therefore, volunteers would be affected themselves in some way.
2. Invention of a creative solution or novel idea that addresses a problem or seizes an opportunity - In September 2012, TWB's new Director, Rebecca Petras, attended a conference presentation by Will Lewis, from Microsoft, on lessons learnt from their project Mission 4636, which she describes as a catalyst in her - and TWB's - understanding of how to build a real-time crisis response team for translation. There has been a large emphasis on crowd-sourcing within the translation community in recent years, and the Words of Relief project built on this momentum, as well as one of the key lessons from Mission 4636 on the importance of using technology to engage a wider community of translators. In this sense, the Words of Relief project was a new iteration of an innovation Mission 4636 had piloted in Haiti.
3. Development of the innovation by creating practical, actionable plans and guidelines - Words of Relief was piloted between December 2013 and May 2015 in Kenya. The project focused on northern Kenya, which suffers frequent drought and food insecurity, and western Kenya, which has suffered repeated flooding and cholera outbreaks. Eight translators worked out of the Nairobi office, translating over 400,000 words of crisis-related content into Swahili. A pilot spider network of 11 volunteer community translators was established to translate crisis messaging from English into 11 local languages. A code of conduct was written and a Facebook page set up to support communications within the network. Dublin City University conducted a study on the effectiveness of the translation training and exercises on improving the quality of translation. The study reported high confidence levels in translating quickly and accurately, determining the importance of messages, and overcoming cultural and language issues. The main challenge during the simulations was internet connectivity. The development phase involved ongoing refinement of the concept; for example, TWB programmed the translation engine with over 425,000 words of crisis-related information to improve its accuracy.
4. Implementation of the innovation to produce tangible examples of change, testing it to see how it compares with existing solutions - In July 2014, a study by the United Nations Children's Fund (UNICEF) indicated that poor information was fuelling the spread of Ebola; TWB described it as a "crisis of language". TWB set up a spider network of around 12 translators and translated information on the Ebola virus into local languages in Guinea, Mali and Sierra Leone. Three teams of translators were recruited through community and translation networks. They received TWB's Rapid Response Team online orientation followed by further training, via Skype, on crisis translation. Over 100 items - approximately 81,000 words - were translated into 30 languages. TWB worked with more than 10 partners to collect, translate, and disseminate local language materials.TWB primarily provided translation for written materials because TWB's partnership model means that their translation service is reliant on communication formats chosen by the non-governmental organisations (NGOs) that use it, and most of these partners continued to work with written materials despite the low literacy rate. A video, "Ebola: A Poem for the Living", which TWB provided translation for in 7 local languages, proved more successful in reaching an audience. The video was shown on national television in stadiums and churches and was viewed over a million times. Interpretation for radio and voice is likely to play a greater role in future crises.
5. Diffusion of successful innovations - TWB is using learning from Kenya and West Africa to refine the Words of Relief innovation and promote its wider use. In particular, at the time of writing, TWB had just secured funding from LinkedIn to digitise its community translator's training package. In addition, TWB is working to refine its support model, including exploring options for in-country TWB personnel, a 24-hour support network for the first 72 hours of crisis, and options for incentivising volunteers. TWB is taking an iterative approach to developing the Words of Relief model by using the experiences of the development and implementation stages to inform the next phase of the project. The success of diffusion will be dependent on convincing NGOs of the importance of reserving time and budgets for local language communication. In West Africa, agencies were not prioritising language needs, and TWB funded an advocacy video, a published impact study, and an infographic to create awareness of the need for translation as a critical part of agencies' disaster responses.

The research team used evidence collected for this case study to assess the success of the Words of Relief innovation process against 3 criteria. Overall, this process was highly successful in generating increased learning and evidence around the problem of lack of local language translation and resulted in a service that has demonstrated clear comparative benefits over the alternative of English/French-based communication. During the pilot stage, over 428,000 crisis messages were translated and made available on the Communicating with Disaster Affected Communities (CDAC) Network website. There is evidence of some uptake of the message library in Kenya: at its peak (outside of crisis), the message library received 450 views per month. Also, the innovation is unique in having been the subject of a comparative assessment. (Dr. Sharon O'Brien conducted an impact study of the project by comparing two posters on Ebola, one produced in Swahili and one in English. She assessed 197 multilingual people living in crisis-affected areas of Kenya and found the poster was not understood in Swahili by 8% of the group and was not understood in English by 76% of the group (see figure 1 opposite). It was also notable that 82% of those tested said they would prefer to receive crisis health messages in spoken format. While this study demonstrates comprehension of messages is greater when conveyed in a local language, it does not go so far as to measure what some might view as the "true" impact of local language translation: the changes in behaviour and practice such messages are intended to bring about. Moreover, TWB does not have a mechanism for assessing its reach: if, how, and how often partners are using translated material in their communications.) However, as noted above, wider diffusion remains a challenge owing to the lack of attention by aid agencies to issues of language in humanitarian response.

Next, the case study explores 6 factors generally held to be fundamental to successful innovation processes and the way in which each works in the context of Words of Relief. In brief:

• Managing relationships and setting common objectives - "A challenge for TWB is that NGO partnerships during the pilot phase were informal....NGOs and government agencies were slow in responding to requests for content....[I]n the heat of an emergency, NGOs are often focussed on programmatic issues, and individual aid workers do not see language as a priority."
• Dividing tasks and responsibilities - "Internally, the Words of Relief team is small and nimble. The organisation decided to recruit local staff because it was 'important that the people who are telling the story understand the problem'. Working as a remote team forces the organisation to clearly delegate responsibility and be rigorous in internal communication."
• Resourcing an innovation - "TWB is a virtual, digital network and does not have a physical presence in West Africa....The Ebola response taught TWB it would be necessary to have an in-country presence: in future, resources should be reserved for a presence in the field to allow face-to-face contact with partners, advocacy on the importance of language and assessing the uptake of translated materials."
• Flexibility of process - "There were three lessons relating to how organisations can build flexibility into the innovation process. First, in order to stay flexible, TWB controlled the rate of the project's growth....Second, the opportunity for flexible funding allowed TWB to access an additional small grant for implementation of Words of Relief in West Africa....Third, TWB has a small team that collectively built learning into its processes."
• Assessing and monitoring risk - "While TWB had a good understanding of the risks faced by the project, it lacked a strong mitigation strategy. As a result, several of the risks, including the challenge of providing written communications in societies with low literacy rates, came to fruition and had an arguable impact on comprehension of the translated messages..."
• Drawing on existing practice - "The Words of Relief project drew heavily on experience of its staff, the staff of Mission 4636 and the translation community more broadly to design the innovation. Existing practice influenced the choice of NGO partners, informed the type of messages chosen for the library and was critical in designing training for the spider network translators."

"An important theme arising from this innovation is the necessity, and difficulty, of demonstrating impact....TWB has established an evidence base for the importance of translation in crisis but evidence of the impact of the Words of Relief products is largely anecdotal." Several emerging lessons are identified for best practice in innovation, such as: "For technology-driven innovations, participative design and a good understanding of the user community is vital for achieving wider use."

This study is one in a series of 15 case studies, undertaken by Active Learning Network for Accountability and Performance in Humanitarian Action (ALNAP) in partnership with Enhancing Learning & Research for Humanitarian Assistance (ELRHA)'s Humanitarian Innovation Fund (HIF), exploring the dynamics of successful innovation processes in humanitarian action. They examine what good practice in humanitarian innovation looks like, what approaches and tools organisations have used to innovate in the humanitarian system, what the barriers to innovation are for individual organisations, and how they can be overcome.