Introduction
Between 2015 and 2024, the author and co-researcher Nafisa Shama Probha conducted testimony collection work before the Administrative Review Tribunal in Bangladesh, gathering accounts from approximately 400 to 500 survivors and witnesses of the 1971 genocide. This article reflects on what that work taught us about testimony — its power, its fragility, and the profound cost it extracts from those who tell it and those who receive it.
Nafisa Shama Probha died on 15 August 2024. She was unable to bear the accumulated weight of what we had heard and seen. This article is dedicated to her memory, and to everyone who has given us their testimony in trust.
The Witness and the Archive
A testimony is not simply information. It is an act — an act of courage, of trust, and of assertion. When a survivor chooses to speak, they are insisting that what happened to them happened, that it should be known, and that someone should hear it. The responsibility of the interviewer is to be worthy of that trust.
This places obligations on the archivist that go beyond methodology. It demands a particular quality of attention — what the philosopher Simone Weil called "waiting" and what trauma scholars call "witnessing." The interviewer must be present to what is being said without imposing, without rushing, without allowing their own discomfort to truncate the account.
Methodological Approach
Our methodology combined structured and semi-structured interviews. For each testimony, we prepared a framework of topics — the witness's location in 1971, their relationship to events, what they saw and experienced, what happened to family members — but we did not impose this framework rigidly. We allowed the testimony to find its own shape.
We recorded all interviews with explicit consent. Where witnesses requested anonymity, we recorded the testimony without name attribution, noting only the district and relationship to events. Where witnesses were comfortable being identified, we preserved their names with their permission.
We interviewed primarily in Bengali, though some interviews were conducted in regional dialects and later transcribed. We did not use interpreters between witness and interviewer where possible, as the presence of a third party changes the dynamic of testimony in ways that can restrict disclosure.
The Psychological Cost
Nothing in our training prepared us for what the testimonies would do to us.
After several hundred interviews, patterns began to appear in our own responses. Sleep disturbance. Intrusive recall. A difficulty distinguishing between what we had witnessed directly and what we had been told. The clinical literature on secondary traumatic stress describes this constellation of symptoms, but reading about it and experiencing it are different things.
We sought supervision and support. We took breaks. We talked to each other. None of it was sufficient. The accumulation of other people's suffering is not something that can be processed in real time. It settles in, slowly, and it changes you.
Nafisa experienced this more acutely than I did. I do not know why — perhaps temperament, perhaps proximity to certain testimonies that resonated with her own history. She was meticulous in her work and deeply compassionate in her manner with witnesses. The witnesses trusted her immediately. That trust, and what they placed within it, became unbearable.
Ethical Obligations
The testimony collector bears ethical obligations both to the witness and to the archive. To the witness: to receive their account without judgment, to protect their identity where requested, to represent their words accurately, and to use the testimony only in ways consistent with the purposes for which it was given. To the archive: to maintain the integrity of the record, to document sources and contexts, to distinguish between what was directly witnessed and what was reported, and to preserve the testimony for future use.
These obligations sometimes conflict. A witness may wish to make accusations that cannot be independently corroborated. A testimony may contain inconsistencies that, if noted, may appear to undermine the credibility of the account. The archivist must navigate these tensions with care, neither suppressing difficult material nor presenting it without appropriate context.
What the Testimonies Tell Us
After hundreds of interviews, certain things become clear that no single testimony can convey.
The killing was not random. It followed patterns — geographic, demographic, temporal — that reflect deliberate targeting. The targeting of Hindu communities, of Awami League supporters, of educated men, of young women, was consistent across regions and across time. This consistency is itself evidence of organised intent.
The use of sexual violence was systematic. Survivors across different regions, who could not have coordinated their accounts, described the same methods, the same contexts, the same patterns of selection. This consistency is not coincidence.
The involvement of local collaborators — the Razakars, the members of the Al-Badr and Al-Shams — was pervasive. Local knowledge was essential to the targeting. Pakistani forces could not have identified their victims in unfamiliar villages without local guidance.
Conclusion
Testimony collection is not neutral work. It changes the witnesses, who must revisit what they survived. It changes the collectors, who must receive what no human being should have to carry. And it changes the archive — transforming an abstraction called "the genocide of 1971" into something specific, named, and human.
We owe the witnesses who spoke to us more than we can repay. The least we can do is preserve what they said, accurately and with care, and ensure that it is heard.
This archive exists because of their willingness to speak. It is dedicated to their memory, and to Nafisa's.