The Empty Square Also Testifies: When the Record Comes Back Blank

The file arrived on my screen labelled 'chess'. That was the only word in it. Then seven fields, every one of them blank. Title: not applicable. Source: not applicable. Type: unclassified. One-sentence summary: an empty string. Author stance: not applicable. Purpose: not applicable. Information points: an empty list. And then the strangest line of all: 'Entities Involved — identify from the information points above.' A command to identify, issued over a list containing nothing. Beside it, a second instruction: 'Source Quality — judge from the source fields of the information points.' As a chess man, I know that sheet of paper. Imagine a scoresheet with everything written at the top — White, Black, event, board number, date, arbiter's name — and then every move-space below it empty to the final line. It still looks like a record. Thirty-five years of digging through the paperwork of this game have taught me that the dangerous kind of emptiness is never the visibly blank one; it is the one that has learned to sit inside the correct format. The board remembers what the camera missed, but when the scoresheet goes blank, even the board's memory stops somewhere. Chess is a strange game. Every move in it is mathematically inevitable; once a game has been played, it is in principle one hundred per cent recoverable. The phrase 'in principle' is the trap. The only raw material for that recovery is the scoresheet, and scoresheets are written by people — by hand, in pencil, sometimes in a tired club secretary's notebook. Cricket's ball-by-ball data is collected automatically, paid for by sponsors. Chess still collects its own memory largely through volunteer labour. The game that should have the easiest archive in world sport has the most fragile one. In 2026 I sat in a Dhaka press box and watched a teenager become a grandmaster — Niaz Murshed, the first in South Asia, a full year before Anand. My notebook from that day holds no moves. It holds damp paper and the noise after the last round; the moves had to be filled in later from somebody else's typing. That same year the two countries took two roads. India built a machine on that moment. We built a statue. A statue carries a date on its plaque. It does not carry a crosstable. Bangladesh has five grandmasters today and has been waiting close to seventeen years for a sixth. The federation runs on roughly Tk 11 lakh a year — smaller than one franchise cricketer's bat deal. School chess has gone quiet, and the calendar of home norm tournaments ends in 2026. In that setting an archive failure is not a luxury problem; it is another symptom of the same illness. Manon Reja Neer took the national title undefeated at fourteen; Niaz took his seventh at fifty-nine. Both are ends of one ledger, and nobody has paid the bill for the thirty years between them. So read this blank file inside that context. Three different conditions have collapsed into one, and they do not cost the same. First, the result never existed — the game was not played, the report was not written. Second, the result exists but nobody captured it — the game was played and the scoresheet is still lying on the table. Third, the expensive one: capture was attempted, the pipeline ran, and the output is zero. The middle condition is our national epidemic. We keep colliding with the third, because it is the only one where the system itself prints 'not applicable'. Something needs stating plainly here. Watch an engine evaluation bar. If the position is not on the board at all, the bar still renders 0.00 — calm, neutral, entirely self-assured. A spectator reads equality. The truth is that there is no position. The two fields being generated on top of an empty information list — entities, source quality — behave exactly like that 0.00. They are derived calculations over an empty set: they inherit the emptiness and then present themselves as findings. A norm certificate issued from zero games is not an error; it is a deception. The four suspected causes in the report translate into chess without much difficulty, and the translation reorders the diagnosis. An empty body returned means the game was never played — anti-scraping, JavaScript rendering, a paywall: the source did not fail, it never arrived. A selector mismatch means the scoresheet was on the table and the collector searched the wrong table. Lost chunks mean the middle of the transcription vanished — the slyest failure of all, because the opening and the result look intact and the game appears complete. Encoding and language drop means a notation nobody could read was quietly filed as 'no game took place'. Four causes, four different treatments; without the list we would operate on the wrong patient every time. In my experience the diagnostic order runs backwards. First ask whether the raw material arrived at all: the content-length of the response, the character count of the main text node, whether the chunk count is zero. Without those three answers, arguing about titles and conclusions is jumping into the middlegame before the board has been set. The failed report also says something it did not intend to say. The minimum it lists — title, source, publication date, at least three sourced information points, summary, stance — is a norm requirement. A FIDE title norm demands a set number of games, opposition of a certain strength, and a crosstable. A single claimed win is not a norm. Likewise, analysis built on undated quotations and unsourced points resembles a norm application with no crosstable: handsome paperwork, no certificate. Where three sourced points are missing, what I am holding is not a match report but fiction wearing the smell of a match. Here is where my real objection sits. Asked to explain failure, all of us point at the pipeline — scraping, selectors, chunks, encoding. The actual blind spot is upstream of the pipeline and it is economic, not technical. Nobody pays for capture. Newsrooms spend prestige on engine analysis, ratings on arguments, credit on bylines. The person typing scoresheets until midnight has no field named after them; our blank file did not name the scoresheet either. No line item in a federation budget says 'archive', just as none does in a cricket board's. The investment layer that gets cut first is the one that turns out to be the most valuable evidence when the accident happens. The second blind spot is psychological, and it is a trap. A blank field makes the hand itch; the urge is to fill it with something plausible. In the same way a fourteen-year-old champion gets drafted into a 'next Niaz' redemption arc after seventeen dry years, an empty information point gets filled with a fluent sentence. A filled field is not a captured fact. Neer is genuine evidence that a pipeline is possible; he is not the pipeline. Who funds him, who coaches him, and who answers for him at twenty-two if he chooses teaching — until those questions have answers, we are back at the statue. So the only thing I take from an empty result is this, and it is not technical. A game that was never written down never really ended; it returns, a little later each time. The scoresheets nobody types tonight will be the only honest answer to somebody's question in ten years about why we fell behind — except that by then the source will be gone. The next time a file comes back empty-handed under a chess label, the first question should be whether the board was set at all. The second is harder: who is writing the scoresheet tonight, and who is paying them?