It comes to the palm before the page, on a national day of remembrance related to the 1971 war of independence.
A youngster picks up their phone and starts scrolling. It starts with a black-and-white image of refugees, a serious tune, and a label reading “sacrifice.” Then there is a swipe followed by a brief video of a survivor talking. Then perhaps a branded tribute, a school parade, a comedy clip, an advertisement. What does this unique, modern-day barrage of tributes imply? The past does not disappear; it glimpses, subsides, re-emerges, and is swept down the stream that carries all the rest.
This is among the distinguishing features of public memory in our era: 1971 is no longer remembered mainly through family stories, textbooks, memorials, museums, and newspapers. It has become an element of feeds, clips, search results, recommendations, reposts, subtitles, memes, playlists, and AI-generated summaries. It goes more quickly, farther, and more democratically than ever. It also moves through structures that are not created to be historical but to be visible, engaging, and memorable.
This is why the question before us has changed. What becomes of a foundational national memory when it is put into the infrastructure of sorting attention?
This is the point of memory politics in the age of algorithms.
The politics of memory is never only about the past. It concerns who gets to define the nation through selective remembrance and selective silence, and which parts of the history are elevated, ritualised, or pushed aside.
The institutions that mediated this memory were relatively obvious over the decades. Survivors and witnesses were present. Some families kept their stories a secret, sometimes agonisingly. Political actors used the war to exercise moral authority. It had historians, journalists, textbook writers, filmmakers, and museums. These schools or points of reference did not generate a pure memory; rather it possessed their omissions, their fallacies, and their black holes. But at least theirs was an identifiable role.
Digitalisation has broadened that landscape. It has made fragile documents, testimonies, and photographs easier to preserve, search, and circulate. Students and researchers worldwide can now experience the pieces of 1971 without waiting for a gatekeeper to open the room.
The difference between an archive and a feed is not obvious. An archive is constructed with preservation in mind. It raises the question about what a document is, its origin, the manner of cataloguing it, and on what grounds it ought to be read. Meanwhile, a circulation-based feed is constructed with optimisation in mind. It poses the question of what will be clicked, watched, reacted to, shared, or recommended. The archive appreciates provenance; the digital platform is performance-oriented. The archive is sorted for retrieval; the feed is structured in a way that is easy to notice.
As soon as 1971 pops up on the feed, it no longer seems like history to the public—it appears as content. Its appearance is influenced by systems that favour speed, emotion, signal clarity, and the repeatability of engagement. This does not necessarily lead to falsehood, but it does strain the memory to make it more readable for the platform. Complicated histories are pushed to be succinct. Subtlety is a bad rival of certitude. Visual fragments are created out of long moral and political reasoning. Testimony is clipped. Grief is aestheticised. Patriotism is packaged.
It is easier to consume a war of unprecedented magnitude and cruelty rather than comprehend it. The genocide, the refugee exodus, the asymmetry of experience, and the post-traumatic afterlives may all be flattened into a series of emotionally resonant, highly viral symbols.
This marks a significant change in memory politics. Previously, one had to be concerned mainly with denial, censorship, and ideological rewriting. Such threats did not disappear, but today, the process of remembering is also mediated by commercial apparatus, the logic of which is neither archival nor pedagogical. There is no need to deny history any longer to render it weak; one can simply shred it, speed it up, seasonalise it, and devour it in the incessant race to attention.
That seasonalisation is nothing new. The digital public is flooded with memories on some dates, particularly in March and December. Speeches circulate, posters return, archival photographs reappear, slogans are revived. For a day or two, the nation appears to remember; then the traffic drops and the topic sinks within the feed. Memory comes in spikes. It comes as a fad and goes as a fad. The past thus is not forgotten—it is simply rendered intermittent.
This is important since the memory of a nation cannot endure as a periodical upsurge. People do not stay historically literate by becoming twice-yearly ceremonially expressive. But when 1971 is experienced largely in the form of commemorative traffic, commemoration risks being performative rather than educative.
Critique of the digital present must not be nostalgic; the analogue era was also afflicted by silence, isolation, and familial memory. Digital culture, too, has allowed overlooked testimonials, local histories, and diasporic rediscovery. It has given us new methods to access the past, despite the inaccuracies. The youth retitle old films, scan family documents, revitalise neighbourhood histories, and distribute repressed testimonies. The diaspora also encounters 1971 through digital titbits that inspire further research.
And thus, it is not a question of whether digital memory is good or bad. That is too blunt a question. The actual issue is whether democracy is accessible for commercial formatting. Is it possible to expand participation in memory preservation without giving memory over to the logic of virality?
AI compresses and counterfeits memory with cloned voices, synthetic testimony, and fabricated authority. A chatbot can settle a dispute or remove a picture’s provenance, which undermines evidence of a historical event as complicated as 1971. Histories of atrocity depend on confidence in testimony, records, images, documents, and the labour of verification. As soon as synthetic media is a commodity, misusing that trust becomes easier. False material may be added to the stream. Real content can be rejected as a counterfeit. The conditions of proof themselves are in a state of flux. This poses both an ethical and an epistemic dilemma. The witness is not crude. The words of the survivor are never just emotionally strong resources to cut-and-paste to get them interested. The corpses are not aesthetic material. Mass violence is a memory that must be recalled responsibly in context, with restraint and source fidelity. The more technologically reproducible the memory is, the more these ethics are required.
So what would a more severe digital memory politics entail in the case of Bangladesh?
First, it would see archives as infrastructure instead of decoration. Memorial sentiment varies across countries. It takes strong institutions to safeguard oral histories, newspapers, photos, local records, and other documents. These resources should be searchable by students and researchers. The latter will win when it is hard to find serious archives and easy to find shallow material.
Second, protecting context as a social good is important. It is crucial to know where the photo came from and the historical context of the testimony. A viral fragment should help you understand more. Exposure must result in understanding. The clip should not end the story.
Third, civic literacy should include knowing how to use digital history. People need to know about 1971 and how to judge what they see. They need to be able to trace an image, check a quote, tell the difference between archival and stylised tributes, and spot fake authority. In the age of algorithms, memoirs that are not checked are structurally weak.
Bangladesh must protect and discover its history simultaneously. More and more, search engines, social media sites, recommendation systems, and AI interfaces decide what people see first and what they don’t. Ranking is now a part of the politics of remembering. If a country gives its history to the market, it may lose control over how it is shared.
The fight over 1971 is taking a new shape, however. The old questions remain. Who is remembered? Who is left out? Who speaks for the nation? Who is supposed to define sacrifice? But the terrain has shifted. It is no longer threatened only by silence but also by noise, not just by erasure but also by endless circulation without depth.
It is a disturbing aspect of the present. A society may seem overwhelmed by memory and historically weakened. It can write all the time and yet know precious little of substance. It may confuse the visibility with the seriousness. It can absorb the 1971 symbols and still lose its attention discipline.
It is not to save 1971, then, out of the digital world. That would be inconceivable and, frankly, stupid. The challenge is greater: to construct historical forms of digital memory that don’t break history down into bite-size content. It can also lose it because the past is forever on the screen, circulating, beautifully edited, instantly shareable, and slowly losing weight.
Khairul Hassan Jahin is a journalist at The Daily Star. He can be reached at [email protected].
Views expressed in this article are the author's own.
Follow The Daily Star Opinion on Facebook for the latest opinions, commentaries, and analyses by experts and professionals. To contribute your article or letter to The Daily Star Opinion, see our guidelines for submission.
