The language lesson: what we’ve learned about communicating with Rohingya refugees

A Translators without Borders study found that access to information has improved in the Rohingya refugee response as a result of an increased humanitarian focus on communicating with communities. Yet language barriers still leave many Rohingya refugees without the critical and life-saving information they need. Prioritizing spoken communication in Rohingya and a mixed approach on formats and channels is key to effective communication.

Our assessment of comprehension and support needs among Rohingya refugees tested their comprehension of simple spoken, visual, and written information.

From the outset, language challenges have played a central role in the Rohingya refugee response. There are at least five languages — Rohingya, Bangla, Burmese, Chittagonian, and English — used in the response. Low literacy levels and limited access to media compound the situation.

To find out how humanitarians can effectively communicate with refugees, Translators without Borders assessed language comprehension and support needs among the refugees. We surveyed more than 400 Rohingya men and women living in the Kutupalong refugee camp in Cox’s Bazar, Bangladesh. We asked them what languages they spoke, how they preferred to receive information, and we tested their comprehension of simple spoken, visual, and written information.

Here is what we found.    

Communication has improved, but not all Rohingya refugees feel informed

Twenty-eight percent of refugees say they do not have enough information to make decisions for themselves and their family. Extrapolated to the whole camp population, this suggests that about 200,000 people feel that they lack the basis to make properly informed decisions.  Nevertheless, it is a marked improvement from a year ago when an assessment by Internews found that 79 percent of refugees did not have enough information.

Communication in spoken Rohingya is critical

Rohingya is the only spoken language that all refugees understand and prefer. Our study shows that 36 percent of refugees do not understand a simple sentence in Chittagonian. Women are less likely than men to understand spoken Bangla or Burmese. Refugees prefer to receive information in spoken Rohingya, either by word-of-mouth, loudspeaker, or phone call.

This preference for spoken Rohingya coincides with strong trust levels in imams, family, aid and medical professionals, and majhees (government-appointed community leaders) as sources of information. Radio, TV, and the internet are less trusted by and less familiar to women.

After spoken Rohingya, simple visual messaging is the most widely understood format. Comprehension rates for visual communication are high regardless of gender, age, or education level.

These Rohingya participants helped us assess language comprehension and support needs among the refugees living in the Kutupalong refugee camp in Cox’s Bazar, Bangladesh.
These Rohingya participants helped us assess language comprehension and support needs among the refugees living in the Kutupalong refugee camp in Cox’s Bazar, Bangladesh.

Burmese is the preferred written language, and is relatively well understood

After Rohingya, Burmese is the preferred language for written communication. Although two-thirds of refugees prefer written communication in Rohingya, the language lacks a universally accepted script. Refugees prefer written information to be given in brochure or leaflet form. This allows them to take information away with them and ask a friend or family member to help them understand it.

Sixty-six percent of refugees said that they cannot read or write in any language, and comprehension testing broadly confirmed this. When tested for reading comprehension, 36 percent understood Burmese, a similar rate to Bangla and English.

Investment in language will improve the response

These findings make it clear that there are varied language needs within the Rohingya community. They show that different people understand, prefer, and trust different formats and sources of information. Nonetheless, practical actions for effective humanitarian communication exist.

Using Rohingya for spoken communication, and Burmese for written information is important. Providing information in a mix of formats and channels to account for varied preferences and education levels will also help.

Investing in formal training for field workers and interpreters in the Rohingya language and in humanitarian interpretation techniques is key. Staff should be supported to communicate in the language understood and preferred by the whole community.

This enumerator is tests a Rohingya man’s comprehension of simple spoken information.

As time goes on, communication and language preferences may change. Ongoing assessments on information and language support needs should be coupled with further research to better understand communication issues affecting the Rohingya refugee response. Sustained coordination among humanitarian organizations can help ensure communication is consistent, appropriate, and addresses key community concerns.

View the research brief.

Read the full report.



This study is part of the Common Service for Community Engagement and Accountability. Funded by the United Kingdom’s Department for International Development (DFID) through the International Organization for Migration (IOM), and by European Union Civil Protection and Humanitarian Aid (ECHO). It was conducted in partnership with IOM Needs and Population Monitoring and REACH Initiative. Translators without Borders has been working in Bangladesh in support of the Rohingya refugee response since 2017, conducting research on language barriers and communication needs, advocating for local language and cross-cultural competence, providing translation and localization support, and training humanitarian staff on the Rohingya language and culture.

Written by Mahrukh 'Maya' Hasan, Evidence and Impact Consultant for the Rohingya refugee crisis response in Bangladesh.

Hefazot transforms to nirapotta; janela becomes kirkiri

One year into the Rohingya refugee response, a language evolves with its people.

Language is fluid. It is subject to environment, culture, and the whims of communities. It’s been one year since more than 700,000 Rohingya fled over the border from Myanmar into Bangladesh. And it is here in these cramped refugee camps that a language is shifting and evolving right in front of us.

The early days

In the early days of the response, the language challenges for the refugee community were immense. First responders struggled to communicate where and how to access lifesaving services, and to document individual accounts of trauma. The community struggled to explain its essential needs (According to one report, more than 60% of refugees said they could not communicate with aid workers), and dozens of untrained interpreters emerged overnight to fill the need for linguistic middlemen. Many of these amateur interpreters spoke the local Chittagonian; while somewhat similar, there are very distinct differences in the languages that create confusion, misinformation and miscommunication. Rohingya speakers estimate that there is around a 70% similarity between Chittagonian and Rohingya (Rohingya Zuban report). That might sound pretty good – but keep in mind that there is more than 80 percent similarity between Spanish and Italian, and no one would ever hire an Italian interpreter for a Spanish refugee!

“Only a few of our men knew Bangla or English. The locals were helping, but even they couldn’t fully understand us. We couldn’t explain to them why we were fleeing, what was being done to us across the river.”  

Woman in her mid-30s, living in Nayapara, an informal camp in the region.

For example, early in the response, the phrase ‘violence against women’ was frequently misinterpreted as ‘violent women’. Certain kinship terms, like husband (beda / zamai / shwami) and daughter (zer-fua / maya-fua / mela-fua), led to some families being separated when shelters were assigned. Then there was gaa lamani — in Rohingya it means diarrhea, but in Chittagonian, it literally translates as ‘body falling down.’ This certainly led to some confusing sessions with health workers.

Signs directing the community to health centers, food distribution sites and other essential services were written mostly in English (although less than 5 percent of the population is literate in English). The main avenue to complain or give feedback was the complaints box – a concept that not only requires a level of literacy, but is also culturally alien to the community.

Help desk sign

A new way forward

A year on, many organizations are creating innovative ways to communicate. For example, many are working with the community to develop image-based signage.The challenges in developing images that represent such seemingly simple concepts as ‘caution’ or ‘hospital’ give an insight into the complexities of communicating symbols amongst different languages and cultures.

“A white hand means clean hand. If you want to stay ‘stop’ or ‘caution’, use red. A red hand will stand out. It will tell us to stop.”

Middle-aged man, testing shelter signage

More than a million Rohingya refugees now live in camps spread across the southernmost tip of Bangladesh. Here, older refugee communities that arrived over the last 30 years live side by side with new arrivals and the host community. Throw into the language ecosystem the institutionalized jargon spoken by English speaking aid workers and you have a fascinating interplay of language and culture.

Language is influenced by its surroundings. For example, the Rohingya dialect spoken by the older arrivals now differs from the Rohingya spoken by the newer arrivals. Decades of living amongst a Bangladeshi host community has seen their mother tongue adopt a number of Bangla words. For example, a newly arrived refugee might use the word hefazot, to refer to ‘security’ or ‘safety’ while the more established refugee community now borrows from Bangla nirapotta. Older refugees might use the word janela (actually borrowed from the former Portuguese colonizers) meaning window, while newer refugees use kirkiri.

“When I go to the clinic, the doctor can’t understand when I explain what’s wrong using Rohingya language. The health interpreter sometimes teaches me the word for my condition in Bangla. This is helping me communicate better with the doctor.”

Woman, 54

When speaking to a newly arrived Rohingya refugee, you will notice the influence of Burmese, Arabic, and Farsi in their terminology. Serama (from siyama in Burmese, meaning ‘female teacher’), serang (‘to make a list’) and atwarta (‘documents’) show the Burmese and Rakhine influence on the language. While mosiboth (‘danger’) and izzot (‘honor’) come from Arabic, aramiyoth (‘health’), moroth (‘male’), and rong (‘color’) are Farsi words either borrowed directly or via Urdu.

“Sometimes it’s even difficult for us to understand the new Rohingyas, especially if they come from fuk-kool” (literally, ‘the east side’ of the mountain range). “Their accent is distinct, and they use words that many other Rohingyas don’t use. Maybe they use more Rakhine words.” 

Salim, Rohingya interpreter from Teknaf.

In the last year it has become clear that humanitarian responders are giving more than aid to the community. New English words are creeping into Rohingya dialogue every day.  For example, the Rohingya word for ‘toilet’, tatti is now commonly replaced by the word lettin (fromlatrine’) and modotgoroya, the word for ‘aid worker’, has become bolontiyar (fromvolunteer’) in everyday Rohingya vocabulary. While the registered Rohingya community uses the Bangla word shoronati, the newer arrivals have replaced the Burmese dokasi with the English word ‘refugee’ (pronounced rifuzi). Interestingly, even English words that they picked up while in Myanmar are now being replaced with “newer” English words, like the word for intravenous saline (deep from ‘drip’ in Myanmar; selain from ‘saline’ in Bangladesh).

“Most of us now say ‘hosfital’ for medical centers, but the older women still prefer to ‘dattahana.’”

Young woman, focus group discussion

The camp is full of different languages; Burmese rhymes compete with Arabic hymns and Hindi pop songs. The community is eager to learn new languages. Burmese is regularly cited as the most desirable language to learn, closely followed by English and Bangla (in that order). And while the teaching of Bangla is officially banned by the government, some Rohingya men – particularly the youth – study informally at night among themselves and with the older, registered Rohingya refugees.

This is what makes our work here so fascinating. It’s riveting watching language twist and turn to fit into its new environment like you would squeeze into a pair of new jeans. That’s why resources like our glossary, resources, and the training we provide to field workers in this response is so crucial. This ensures important information is delivered in the right language and that as their language needs shift and evolve, we are able to move with them. Over the next year we’re sure to see more change, as more children have access to learning centers that teach English and Burmese, and interactions between the community and aid workers from around the world increase. Listen carefully; language matters.

This blog post is based on dozens of conversations and focus groups held by TWB with the community over the last year.

Written by Irene Scott, TWB Program Director, Bangladesh, and AK Rahim, TWB Sociolinguistic Researcher.

Report from the Field

Our Board Chair visits Bangladesh, sees progress and challenges first hand

I recently visited Bangladesh with Ellie Kemp, our Head of Crisis Response, to see first hand the work of Translators without Borders (TWB) around the Rohingya crisis. Our trip included a visit to the “megacamp” at Kutupalong, the biggest refugee camp in the world, and meetings with our partner humanitarian response teams based in Cox’s Bazar. We also spent a few days in Dhaka meeting with donors and partner organizations.  

The crisis is an incredibly challenging one. One year ago this month, the Myanmar army escalated a long-standing campaign of persecution against the largely Muslim Rohingya to what the UN High Commissioner for Human Rights has characterized as ethnic cleansing. Thousands were killed in Myanmar, and over 700,000 fled over the Naf river to Bangladesh; at the peak, 20,000 refugees arrived per day. The Myanmar government refuses to recognize the Rohingya as citizens, claiming they are Bangladeshi migrants. The Bangladeshi government, although generously offering them sanctuary, is facing its own political challenges and does not recognize them as refugees either.

The Rohingya people therefore are denied the right to work and not allowed to register as residents, and cannot build more permanent homes than the bamboo and tarpaulin shelters they have been in for the last 12 months. Formal schooling is not allowed in the camps;  people in camps are even officially forbidden from owning mobile phones. On top of this the humanitarian response has been suffering from poor coordination due to unclear division of responsibilities. Unfortunately these challenges have been acutely felt in the way the responding organizations have communicated with the communities they are trying to help.

This response was supposed to be different.

There has been increasing awareness over the last few years of the need to improve programs for communicating with communities (CwC) and to build these programs into every response. This was one of the first major responses since the World Humanitarian Summit “Grand Bargain” signed two years ago in Istanbul, where the humanitarian community committed itself to doing this better. Unfortunately it has not quite played out that way. Too often, key roles in CwC are left vacant or not given the resources they need. Key initiatives, such as refugee registration (a sensitive topic for a systematically persecuted population), have been handled without proper planning of how to communicate. And while some major donors, such as the UK’s DFID and the EU’s ECHO now recognize the problem with inadequate funding for CwC, the funding provision still remains far below the need. Our task on advocacy around the need for mainstreaming CwC continues…

Andrew B,
Andrew Bredenkamp at Kutupalong camp, Cox’s Bazar, Bangladesh.

From a linguistic perspective the situation is complex too. Rohingya is not a written language, and the thousands of Rohingya in the camps who have received an education were taught in Burmese. Rohingya is related to Chittagonian, the local language spoken in Cox’s Bazar and more distantly Bangla, the national language of Bangladesh. Unfortunately this has sometimes led to responders assuming that Chittagonian and Rohingya are basically the same language. We heard the statement repeatedly that “Chittagonian and Rohingya are 70 percent the same, so we’re using Chittagonian speakers”. Bearing in mind Spanish and Italian are 80 percent similar and that no one would consider using Italian to communicate with a Spanish community, this highlights the need for continued awareness about language issues among responders.

This is not an academic discussion.

The words for “help,” “pain,” “pregnant,” and “menstruation,” even the phrase for the common cold, are all unrecognizably different in Rohingya and Chittagonian. There seem even to be differences between language used by male and female Rohingya speakers.    

Despite these difficulties TWB and our consortium partners, BBC Media Action and Internews, have been able to make a huge difference. Here are some of the highlights:

  • We have been providing language services across the response into both Bangla and Rohingya.
  • We have been providing training for staff and volunteers working for the response organizations, focusing on the differences between Chittagonian and Rohingya and developing glossaries around key topics for critical sectors, such as water, sanitation and hygiene (WASH), health, and protection, including work around gender-based violence.
  • We have been supporting the listening programs of our partner Internews and the content programs of BBC Media Action, helping to make the response more accountable to the refugees and host communities.
  • We have been continuing to advocate for all aspects of the response to take into account the need to communicate with the Rohingya community and the local host population.

Enormous credit has to go to the team, led by Ben Noble, our Country Director, and Irene Scott, Program Director. I have to also mention the amazing efforts of AK Rahim, our South Asian linguistic expert. AK is an amazing source of knowledge about the languages and cultures of the region and how they interact. He has been our secret weapon in winning the trust of the host Chittagonian population as well as the Rohingya community, and has led the research that enables us to provide practical advice for humanitarians on communicating more effectively with both.

We heard time and again from our humanitarian partners that our work was indispensable and extremely effective.

Our donor meetings were extremely encouraging, not just in terms of the desire to support our work in Cox’s Bazar, but also more strategically. There was explicit confirmation at the highest level that “the humanitarian community is still not doing enough on CwC.”

Overall another great testimony to the importance of our mission. There is a lot we can learn about this response for others we are and will be involved in. The need remains immense.

Donate to the Rohingya refugee response

Written by Andrew Bredenkamp, Chairman of Translators without Borders Board of Directors.

Bangladesh Program Update

Bridging language gaps empowers people to communicate in Cox’s Bazar refugee camps

Kutupalong makeshift camp, Cox's Bazar, Bangladesh.
Kutupalong makeshift camp, Cox’s Bazar.

Cox’s Bazar, Bangladesh, once famed for its beautiful 120km long beach, is now home to one of the largest refugee populations in the world. Between 900,000 and one million Rohingya women, men and children, depending on the estimates, now live in the area. Since August 2017, more than 670,000 Rohingya have fled across the border from Myanmar and settled in camps in and around Cox’s Bazar.

Translators without Borders (TWB) first came to Cox’s Bazar in October to assess the communication and information needs of the affected population.

Our team rapidly discovered that language was making communication between the affected communities, humanitarian organizations, and the host population extremely difficult. As reported by our partner organization Internews, more than 70 percent of the refugee population identified themselves as being totally illiterate in any language and more than 60 percent said they were unable to speak to humanitarian providers.

In Cox’s Bazar, Rohingya is often the only language spoken by those most in need. It is an oral language, with no commonly accepted written script.

One of the major communication problems in this humanitarian crisis is the lack of a common language. The humanitarian workers mostly speak English, local NGOs and government officials speak Bengali, many interpreters speak Chittagonian, and the refugees speak Rohingya.

The reality

Take a moment to imagine this in the context of a refugee camp. Signs are erected to identify health facilities and safe spaces for women in a language they do not understand. Information can become distorted as it is passed from person to person and humanitarian organizations rely on untrained interpreters to communicate life-saving information as part of their support to the refugees. As summarized by TWB’s sociologist,

Three interpreters (Hassan, Rafique, and Abdullah) in Kutupalong makeshift camp near Cox's Bazar, Bangladesh. Photo: Eric DeLuca / Translators without Borders
Three interpreters (Hassan, Rafique, and Abdullah) in Kutupalong makeshift camp near Cox’s Bazar, Bangladesh. 

“There’s just a lot of crucial information lost in this crisis.”

One of the most urgent needs is to find ways for the refugee population to fully express their needs to humanitarian responders.

With thorough research and community interaction, we are developing a professional training program and tools to help interpreters and humanitarian organizations understand some of the cultural and linguistic specificities of the refugee population.

Shades of meaning

TWB is developing a freely downloadable glossary of key humanitarian terms. This translates technical terminology in English into simple and clear Bengali, Rohingya, Chittagonian, and Burmese terms. The aim is to cover concepts relevant to a range of sectors, making the glossary useful across the humanitarian response.

‘We really deliberated on the meanings and context of the translations,’ says TWB’s sociolinguist. ‘Words can have shades of meaning, so the social and cultural context is important.’

Working as a consortium with Internews and BBC Media Action, TWB is contributing to a regular newsletter distributed to all humanitarian organizations in Cox’s Bazar. This newsletter, entitled What Matters? The Humanitarian Feedback Bulletin, specifically addresses communication and language issues. The first newsletter, distributed in February this year, highlighted the important differences in weather terms between Chittagonian, Bengali, and Rohingya. This is vital when distinguishing between a warning for strong winds or a cyclone, for instance. 

Ultimately, bridging these gaps is empowering people to communicate. When people can communicate they can assert their rights and humanitarians can deliver life-saving information.

With the cyclone and monsoon season starting soon, the need for simple and actionable information, in plain and clear language that the refugees can understand, is becoming even more acute. The United Nations has estimated that more than 100,000 refugees could be in grave danger when the rains begin in April. These are likely to cause major flooding and landslides in the steep hills and unstable terrain where the camps are located and contribute to the spread of disease.

‘This is where translating key Water, Sanitation and Hygiene (WASH) messages are critical,’ says our sociolinguist. ‘Community workers need to be able to explain the differences in severe weather systems between here and Myanmar, what services are there to help in a disaster, even how to help prevent the spread of disease. These are not messages you can afford to miscommunicate.’

Follow the progress of our work in Cox’s Bazar and consider a donation to support our work around the globe.

Written by TWB’s Program Director for Bangladesh

The Silver Lining – Education brings hope during a refugee crisis

There is a lot of despair and pain radiating from the refugee camps in and around the Cox’s Bazar district of Bangladesh. Physical pain from disease and injury, coupled with a lack of food, are constant issues for the Rohingya refugees in the camps. The mental anguish is much greater. Loss of family to violence, loss of homes and crops, and an ongoing feeling of degradation and violation of rights – this anguish lives with every refugee, every day.

And yet, while walking through camps, meeting with responders and activists throughout Cox’s Bazar, there was also a thin yet constant thread of hope. Would it be possible, now that the refugees are relatively safe, in camps run by Bangladesh and the international community, to truly educate hundreds of thousands of Rohingya children, giving them a future they could not have previously imagined?

Educating an illiterate population

The new influx of more than 600,000 refugees to Bangladesh includes a large number of school-age children. According to Save the Children over 60 percent of the new refugees are children. These numbers could increase as even more refugees are expected to cross the border by the end of the year. UNICEF estimates that more than 450,000 Rohingya children aged 4-18 years old are in need of education services. That includes those who have been in the camps for longer periods [source: Reliefweb].

Evidence indicates that a very large number of the children, as well as the adults, are illiterate. In fact, in a rapid survey conducted by the TWB team in October with Rohingya refugees, 73 percent of respondents self-reported to be illiterate. This illiteracy is limiting the children’s ability to be further educated and to demand their human rights.

Evidence also indicates that when education is made available, literacy rates increase. In fact, in the study TWB completed last month, it was clear that refugees who have been in Bangladesh longer show higher levels of literacy than those who had more recently-arrived.  While not easy to obtain, education is more readily available in the established camps than it was in Myanmar where twin restrictions against movement between villages and education above primary level severely limited access to education. When our team tested populations who have been in Cox’s Bazar since prior to August 2017, comprehension rates improved across the board.

Now, with hundreds of thousands of children together in the refugee camps in Bangladesh, is this an opportunity to offer them education and a future?

Unfortunately, though, the language of instruction will be unfamiliar to the Rohingya children. Currently, a number of organizations are looking to set up learning centers in the camps. The goal is to give the refugee children at least two hours of education a day, beginning in January. Yet the official curriculum that the government of Bangladesh has approved does not include mother tongue education, and the teachers who are being hired will teach in Bangla and Burmese, two languages that the children do not read or speak.

Why mother tongue education matters

A wealth of experience and evidence over the last 50 years has proven that children learn better when they are taught in their mother tongue language. We also know that countries do better when their children are educated well. Evidence from a project that Save the Children has implemented in Thailand focusing on mother tongue education for Rohingya children, shows that learning a second language, English or Thai, is difficult when children do not understand the language of instruction. This undermines children’s ability to participate and invest in their education, despite their motivation [source: Save the Children].

But the issues with mother tongue education for the Rohingya children are deeper, because their mother tongue, Rohingya Zuban, is largely oral. The illiterate community speaks it fluently but does not generally have a means for written communication, through their mother tongue. Interesting work is already being done to establish a written form of Rohingya Zuban. A script was developed decades ago, and has been taught within the established camps and throughout areas of Bangladesh and Malaysia. The teaching is generally ‘under the radar’ of even informal education centers, and the materials used are handwritten, as unicoding of the language, is not complete. But even so, there is a major desire among the children and adults in the established camps to learn the written form; estimates put the number who have learned some of it at 10,000. Even more encouraging is the excitement generated among the students when they do have the opportunity to learn it – there is a true sense of the empowerment and identity that learning to read their own language gives them.

This initial mother tongue education work is unknown to most international agencies setting up learning centers, and its potential is unexplored. TWB is working with these agencies, as well as local organizations, international organizations specializing in mother tongue education and hopefully, the Bangladesh government, to include mother tongue tools in the curriculum. Teaching aids in Rohingya Zuban, mobile and online tools in unicoded Rohingya Zuban, and printed Rohingya Zuban early readers would all make a difference.


Now, back home and separated from the daily grind of the response by miles and time, I have reflected on that seed of hope that is education, and started to figure out how TWB can contribute to its growth. I believe TWB can make the greatest impact, by including mother tongue teaching and learning aids into the education programs being developed for Rohingya children.

The Rohingya refugee crisis offers the potential to educate hundreds of thousands of illiterate children, eager to learn, in their mother tongue. I hope we can make it happen.

Please follow this link to support TWB’s response to the Rohingya refugee crisis

Rebecca PetrasBy Rebecca Petras, Deputy Director and Head of Innovation at Translators without Borders.

Bot Towan! #InterpretersMatter

Today was a bit grueling. We went to interview people who have newly arrived at Balukhali makeshift camp about cyclones and cyclone preparations. We did that, and in the process confirmed what I already knew: specific skills are needed to act as a translator or interpreter in a crisis. After today, I am more convinced than ever that language services - translating, interpreting, simplification and training - are an essential part of this crisis response. TWB has a vital role to play here

TWB's first Rohingya interpreter

Our very first semi-trained Rohingya interpreter accompanied me and the two co-leads of the Communicating with Communities Working Group (CWC WG) to run a focus group discussion with ‘model mothers’ (women trained by UNICEF to help people in the community with basic needs), and to interview various members of the community – young, old, and leaders. The day was hot and long, but manageable. What was difficult was talking about cyclones to traumatized people, many of whom told of horrible stories and cried as they recalled what they left behind. The threat of cyclone damage is very real in the camps, especially with the makeshift shelters, but on a sunny day with no wind, it felt trite when set against the horror of gunshot wounds, burnt homes and lost family.

Rohingya interpreter at work on Cox's Bazar
TWB's first Rohingya interpreter interpreting at a focus group discussion with ‘model mothers.’

Yet Rafique, the first Rohingya interpreter who has received some training, handled it all very well. Rafique is a long-term resident of Cox’s Bazar. He is Rohingya by birth, born in Myanmar, and very committed to helping the new arrivals. For years he has run the Rohingya Youth Association, an unofficial group in Cox’s Bazar that teaches long-term Rohingya camp residents some basic skills, especially reading and writing English and Bangla (the children in the camps are not officially allowed to go to school). A number of the kids whom he and his team have taught have gone on to universities around the world, and many of them will help us with our language work from afar.

Training Rohingya interpreters in Cox's Bazar

Rohingya is Rafique’s mother tongue. He had done some ad hoc interpreting for various journalists in town, but he had never been trained. Like many unskilled interpreters, he made classic mistakes. He summarized a person’s long explanation in just a few words, and he very often editorialized what the person said – adding his own explanation. He also would not always properly understand what the English person asked him to do, nodding that he understood when he actually was not quite sure.

Training interpreters like Rafique is one of Translators without Borders’ major goals in Cox’s Bazar. While locals will say that the new arrivals understand Chittagong, the local Bangla dialect, just fine, we keep finding that that is not the case, especially in areas of health. Today we found that is also not the case in simple explanations about cyclones.

Prior to going to the field, I worked with Rafique over several evenings, giving him basic training on how to interpret. We worked with videos of new arrivals talking about their harrowing trips to Bangladesh. He practiced interpreting their explanations, working on the full meaning, but only the meaning – not his additional thoughts. We also discussed the ethics of interpreting and did some basic work on how to operate in a humanitarian context, including how to speak directly to the person being interviewed and how to work with the international staff.

I also worked with the two international team members about the interpreter relationship. While humanitarians who work in the field intuitively know that the interpreter is a vital link that has the power to help the situation greatly, they are often under a lot of stress, working long hours, and possibly unaware of how to ‘get the most’ out of the interpreter relationship and role. This particular situation was a good place to start because the two CWC WG co-leaders are communicators themselves, so they were engaged and willing to learn, focusing on changing their instructions to accommodate the interpreter, asking him to work with the interviewee to give information in small chunks, and encouraging him to sit at the same level as the interviewee to build trust and engagement. The final preparation included giving Rafique all of the field questions in English and Bangla before the interviews. It is surprising how often those working with interpreters do not educate them beforehand on what they will be talking about. Rafique reviewed all of the questions ahead of time so he could practice in his head how to interpret to the interviewee and then could focus during the interview on providing the information back to the interviewer.

Rafique did a fabulous job. He worked really hard all day, as a team with the interviewers. There was very little misunderstanding, and when once or twice Rafique started to add information, I reminded him that that was no longer ‘interpreting’. He quickly corrected himself.

Why words matter

The real reward came toward the end of the day. Sitting around on a mat with the model mothers, we began discussing the Rohingya words for ‘cyclone.’ In helping the CWC WG evaluate best communications about cyclones, I want to make sure that communications are truly understood by the new arrivals, especially those who are illiterate (9 out of 10 of the model mothers were illiterate and did not understand basic Bangla or Burmese). In the back of my mind, I kept thinking about the miscommunications in the Philippines prior to Typhoon Haiyan in 2013. The English language radio stations reported a ‘big wave’ coming; to the Tagalog listeners, this did not seem threatening because it was not called ‘typhoon’ – as a result, many did not leave their homes and were lost once the storm hit.

Rafique asked the model mothers what a ‘big storm with wind and lots of rain’ would be in Rohingya, and they sang out, simultaneously, ‘BOT TOWAN!’, while a very large, stormy cyclone would be ‘boyar awla towan,’ and a lesser storm would be ‘towan.’ In Bangla, a cyclone is ‘tofan,’ which is not far from ‘towan.’ But a very large stormy cyclone is ‘boro dhoroner tofan’, which is significantly different.

Even more importantly, in Bangla, the word ‘Jhor’ denotes a storm with wind and is often used for a cyclone. In Rohingya, ‘jhor’ only means rain without being a real storm and without wind. Similar to the Philippines in 2013, that simple misunderstanding, if broadcast from Bangla weather and warning systems, could be the difference of life and death, especially in camps where word of mouth is the main mode of communication, and winds will blow off roofs and drop shallow-rooted trees.

Words matter. I am very proud of Rafique – it was particularly gratifying when the model mothers, through the one woman who could speak some English, told me that he was the best interpreter with whom they had worked. I think it had a lot to do with him being Rohingya and really listening to how they communicate. I am looking forward to more trainings in the coming days.


Follow the TWB team's journey as they respond to the Rohingya refugee crisis - TWB's most challenging response yet.

Rebecca PetrasBy Rebecca Petras, TWB's Deputy Director and Head of Innovation

On the ground in Bangladesh. So – how do we communicate?

A second report from Rebecca Petras who is heading up TWB’s response to the Rohingya refugee crisis.

The language complexity in the current Rohingya refugee crisis is deep. I had only a faint understanding of it when I landed a few days ago; I have a slightly better sense now. The Rohingya refugees come from Rakhine in Myanmar. They are Muslim; the other dominant population in Rakhine is Buddhist. The political issues between the two groups and the Rakhine as a whole and the government of Myanmar are extremely complicated, and not for my humble explanation. Suffice it to say, on 25 August 2017, a massive and violent event forced thousands of Rohingya to abandon their villages and flee to Bangladesh, through hills, unfriendly areas, and across water. There are still many thousands waiting to cross the river; in total, there are well over 700,000 new arrivals.

In and around Cox’s Bazar, a tourist town (with the world’s longest contiguous beach) in Chittagong division of Bangladesh, there are now official and unofficial camps, sprawling across hills. Because of decades of unrest in Rakhine, there were already approximately 200,000 Rohingya refugees living in either one of two official camps or within the host community, and many have lived there for two decades. They are now witnessing a massive and very uncertain influx from Myanmar, including thousands of orphans, thousands of traumatized and abused women, and many more who need medical attention.

'There really isn’t any communications happening yet, and no one really knows how to do it'

All of this makes for a very complicated language situation, with an amalgamation of spoken Rohingya from long-term refugees and new arrivals, spoken Chittagong from locals, written Bengali (or Bangla), and, possibly, written Burmese. Add layers of what is allowed by the government (still unclear which languages are being allowed), as well as how to translate complicated English terms into Rohingya, and we have a tricky communications issue. One of the main goals of Translators without Borders’ initial work here is to assess the language needs and then direct the numerous responding aid organizations, with accurate information on language. We will be testing assumptions and testing actual comprehension of material that is given to refugees.

We are beginning that assessment now – I will be working with community health workers and youth this coming week, and our research lead (Eric DeLuca) will be joining me in one week to test agency communications tools with new arrivals. But, at the same time, responding aid organizations want to start communicating right now. The community engagement leader of one of our main international partners said when I first met with him that there really isn’t any communications happening yet, and no one really knows how to do it. So while we try to put standards in place, train new interpreters, support interpreters with resources, and address the various language needs, we also need to just start communicating now. With seemingly endless rains and very little infrastructure in the camps, there is a very real danger of water-borne diseases, making communications urgent. What I need most at this time is more time in the day to get it all done.

Below are some suggestions of how you can support this response. Stay tuned for more updates this week.

Rebecca PetrasRebecca Petras, TWB Deputy Director and Head of Innovation

Taking action in the Rohingya crisis: TWB’s biggest language challenge yet

It is somewhere between 9pm and midnight, depending on where exactly my flight is right now. My rubber boots, rain gear, and TWB T-shirts are stowed in the hold; I am enjoying my second film. In a few short hours, we will arrive in Bangladesh, and the work will begin.

39,000 feet above the Earth, language is not an issue. International flight attendants and travelers basically speak the same language. We all understand ‘chicken with rice’ or ‘coffee or tea’ in the few international languages needed…English, German, French, maybe some occasional Arabic. And it is easy, seat back, chatting with seat mates with wildly different backgrounds, to feel comforted by the connection those few common languages bring us.

It is exactly that feeling – that connection and comfort – that language often gives us. I have lived for years in places where the native tongue was not my own: I know the sense of warmth when someone makes the effort to speak my language. Nelson Mandela had it right when he commented on the power of language: “Speak to a man in his language, and it goes to his heart.”

When in crisis, language does even more...

It helps on a very fundamental level, giving people in crisis the basic information they need to be safe, warm and fed. Yet millions of people, especially those who are refugees in foreign lands, must cross a language barrier every time they need basic information. They rely on others for the information they need, hoping that it is accurate and true, because they simply do not understand the language of those trying to help, or they are illiterate and cannot read whatever directive is provided.

How often I wonder how I would handle such a situation. I know that when I get important information in the language of the country where I currently live, the time to understand and then respond is at least doubled – the effort required is so much more. And that is when I’m sitting at my desk, well fed and not fearful for my life or that of my children.

The clear inequity of information that holds billions of people back is what motivates me. It is why I work hard with my colleagues every day to build an organization that uses language to jump over barriers. And it is what has motivated me to go to Bangladesh to set up language provision for the aid organizations trying to help the more than 700,000 Rohingya refugees who have arrived in the past six weeks.

Tomorrow I will be in Cox’s Bazar, a place I didn’t know existed until a few short weeks ago. I have a singular goal: To use language to bring a bit of comfort and help to those who have suffered too much already.

Language matters: I hope you will share this journey with me.


why TWB is responding to the Rohingya refugee crisis

Over 700,000 Rohingya people have fled Myanmar to Bangladesh (in and around the beach town of Cox’s Bazar) in the past two months, many of them entire families - families broken by violence. This is a complex political and humanitarian crisis, and one of the most difficult language contexts TWB has ever experienced.

The Rohingya population is highly vulnerable, having fled conflict and living in extremely difficult conditions. When we launched this response remotely in September, the goal was to find Rohingya translators to translate urgent materials that would help give practical but vital information to the thousands of refugees flooding across the border into a land where they did not speak the language. However, it became immediately apparent that there was very little translation capacity in Rohingya and, furthermore, that we would need to get audio and spoken Rohingya support because very few people write this language, and illiteracy levels are high. It was also too challenging to try to do this work remotely. Yet no situation we have encountered is more in need of our resources.

So we took a chance without solid funding and decided to activate Plan B, sending Rebecca to Bangladesh to try to get something set up to respond to the Rohingya refugee crisis. She will be in the country for four weeks, bringing together a community of translators and figuring out how best to enable them to provide the language link between responders and vulnerable people. She will also be working with aid organizations to ensure that language solutions are funded.  She will be joined by Eric DeLuca, TWB's Monitoring, Evaluation, and Learning Manager, who will conduct a comprehension study in some of the numerous camps, to assess the best ways to communicate with those who are affected by this crisis.

We will be following the team as they document their journey in Cox’s Bazar to set up this response for Communicating with Communities, and we'll be providing regular updates on how they are progressing over the coming weeks.

Rebecca PetrasRebecca Petras, TWB Deputy Director and Head of Innovation