The university had a archive room in the basement of the humanities building that most people did not know existed.
This was not an accident. The archive room was not listed in any campus directory. It did not appear on the building map posted beside the lift on each floor. You found out about it the way you found out about all genuinely useful things in academic institutions — through someone who already knew, who told you only after deciding you were the kind of person who could be trusted not to misuse the information. In Dev's case the someone was his thesis supervisor, Professor Iyer, a seventy one year old historian of folk traditions who had spent forty years documenting the things that official history preferred not to examine too closely and who had recognised in Dev, within approximately fifteen minutes of their first meeting, a person who shared this preference for the underside of things.
The archive room held accounts. Testimonies, mostly. Collected over decades by researchers who had gone to the places where things happened and talked to the people who had survived them or to the descendants of people who had not. The accounts covered a range of phenomena that official scholarship did not have comfortable categories for — disappearances, recurring patterns at specific locations, the kind of events that happened once and then happened again twenty years later and then again forty years after that with such precise consistency that coincidence became an increasingly inadequate explanation.
Dev had been working in that room for three years.
He came in early and stayed late. He read in the original languages when he could and in translation when he could not, making notes in the margins of his printed copies in the small precise handwriting that had not changed since he was eight years old sitting in the dormitory corner with a borrowed pen and whatever paper he could find. He followed threads. He cross-referenced. He built, slowly and with the patience of someone who understood that the architecture of a thing only becomes visible when you have enough of its pieces, a picture of something that had been operating for a very long time in a very specific place.
The place was Crescent Lake.
He had known this was where he was going when he chose his research focus, though he had told himself for the first year that he was approaching it academically, objectively, as a researcher rather than as one of the five boys who had grown up three kilometers from whatever it was. He had stopped telling himself this somewhere in the second year when the accounts started converging in ways that made the pretense of detachment feel not just dishonest but actively dangerous. You needed to know what you were dealing with. Pretending you were neutral about it was a good way to miss things you could not afford to miss.
What he had found, assembled carefully across three years of archive work and fieldwork and the kind of reading that leaves marks on a person — not just notes in margins but something deeper, something that changes the quality of your sleep — was this.
The entity at Crescent Lake was old. Older than the house. Older than the town. Old in the way that geological features are old — not measured in human generations but in something else entirely, something for which human language did not have a precise word. It had been there when the first people settled the valley. It had been there before them. It would be there after.
It did not hunt. This was the first thing that distinguished it from the creatures of folklore that most people reached for when they tried to understand it. It did not pursue. It did not seek out victims in the ordinary predatory sense. It waited. It prepared. And then, at intervals that Dev had managed to narrow down to approximately every twenty to twenty five years, it completed what he had come to think of as a cycle — a structured sequence of events involving a specific number of people, specific relationships between them, and a specific outcome that the entity required to sustain itself for the next interval.
The accounts went back two hundred years with reasonable reliability and suggested, through vaguer earlier records, considerably further than that.
Every cycle involved five people with particular qualities. Every cycle involved a location — always the house, always the lake. Every cycle involved a separation and a loss. The details varied. The structure did not.
Dev had read thirty seven accounts. Some were fragmentary. Some were second or third hand. But the pattern was consistent enough across all thirty seven that consistency itself had become the most frightening thing about it — the sense of something that had been doing this for so long that it had refined the process to a kind of terrible efficiency.
He had not told Milan everything. Not yet. He was still assembling the last pieces, still checking certain things against certain other things, still making sure that what he thought he knew was actually what he knew rather than what he feared. There was a difference and Dev had spent enough time in archive rooms to understand why that difference mattered.
But he was getting close to having enough.
And when he had enough he was going to call Milan and say: it is time.
---
He met Priya on a Tuesday in February, seven weeks before he made that call.
She had come to the university to interview Professor Iyer for a piece she was writing about the documentation of oral histories in disappearing communities — a topic that was, as she put it in the email requesting the interview, both genuinely important and genuinely undercovered, which was the combination she looked for when deciding what to spend her time on. Dev happened to be in Professor Iyer's office when she arrived, sitting in the chair beside the window where he always sat when they talked, a stack of printed accounts on his knee and a pen behind his ear and the particular expression he wore when he had been thinking hard about something difficult for a long time and had not yet arrived at a conclusion.
She walked in and looked at the printed accounts on his knee and said, before she had introduced herself or acknowledged Professor Iyer or done any of the things people conventionally do when entering a room: what are those.
Professor Iyer made a sound that might have been a laugh.
Dev looked at her. She was somewhere in her late twenties, with the quality about her of someone who moves through the world at a slightly faster pace than most people — not physically, though she moved efficiently, but perceptually. She was taking in the room and its contents with the kind of attention that suggested she was already several steps ahead in some internal conversation about what she was seeing and what it meant.
Accounts, he said.
Of what.
Disappearances. Primarily. Events at a specific location over approximately two centuries.
She looked at the stack. Then at him. The assessing look of a journalist deciding whether a source is reliable — not just honest but reliable in the deeper sense of knowing what they are talking about.
She said: can I see one.
He handed her the top account. She read it standing up, which he noted — most people would have sat down. She read quickly but with full attention, the kind of reading that actually processes rather than skims. When she finished she looked up.
How many of these do you have, she said.
Thirty seven, he said. Plus my own notes.
She handed the account back carefully. She said: I'm Priya. I'm here to interview Professor Iyer but I think I might also need to talk to you.
Dev said: I think that might be true.
Professor Iyer made the sound again. Definitely a laugh this time.
---
They talked for two hours after the interview. Then they moved to the canteen and talked for another hour over tea that went cold while they were talking because neither of them was the kind of person who noticed tea going cold when there was something more interesting happening.
She asked good questions. This was the first thing Dev noticed — not impressive questions, not questions designed to demonstrate knowledge, but questions that identified exactly the gap in the information she had been given and went directly at it. She listened to the answers completely before asking the next one. She did not fill silences with noise.
He told her more than he had intended to tell her. This surprised him. Dev was not a person who told people more than he intended to tell them. He had very precise control over what he shared and with whom and in what order. But something about the quality of her attention — the specific, non-performative seriousness of it — made the usual controls feel less necessary.
He told her about the cycles. About the pattern. About the accounts. He did not tell her about himself and the five boys and Crescent Lake specifically — that was not his to tell alone, and he was not ready, and there was a difference between what he would share with a journalist he had met two hours ago and what he would share with someone he trusted.
But he told her enough that she understood the shape of what he was working on. And she received it the way good journalists receive things — without drama, without the performance of being shocked or fascinated, just with a focused, careful attention that said: I understand the significance of what you are telling me and I am taking it seriously.
At the end she said: I have a friend. She works in urban development. She's been seeing someone — one of five boys who grew up near this lake you're talking about.
Dev went very still.
His name is Raju, she said. He told her about it. She told me.
Dev looked at her for a long moment. The particular quality of coincidence that is not coincidence — the sense of things arranging themselves, of the pattern completing another of its steps.
Small world, he said carefully.
She looked at him with the eyes of someone who does not believe in small worlds. Neither did he, actually. But it was the thing you said.
He picked up his cold tea. She picked up hers.
Outside the canteen window the February afternoon was going grey and still. Somewhere two states away a lake was dark and patient and waiting.
Dev thought: it is almost time.
He thought: I need three more weeks. Maybe four.
He thought: I hope three weeks is enough.
He drank his cold tea and looked at the woman across the table who did not believe in ghosts and who had a camera and who was already, he understood without quite knowing how he understood it, going to be part of what happened next.
Whether he told her or not.
Whether any of them chose it or not.
The pattern did not require their permission.
It only required their presence.
---
*End of Chapter Twelve*