Chương 63: People

He saw them from the ridge at 1104, which gave him the surveillance advantage and approximately zero clarity on what to do with it. Three players. The ridge put him eight meters of elevation above their position and a hundred and thirty meters north, in a dense broadleaf cluster that provided cover in both directions: he couldn't see them clearly, and they couldn't see him at all. He had been running this protocol since Day 14: find elevation, assess from distance, do not commit until the assessment is complete.

The protocol had kept him alive through a hundred and twenty-six days of a place that was specifically designed to make alive difficult. The protocol did not have an exception for people who needed help. He settled into the broadleaf cluster and watched. People. Those were people. Three of them, alive, vertical, moving with the specific imprecision of humans who were frightened and fighting and had not yet figured out how to be frightened and effective at the same time.

He had not seen another human being in a hundred and twenty-six days. He had rehearsed introductions four days ago like a man preparing for a job interview in a language he'd forgotten. He had spent three days before that agonizing over whether he could fight a person if he had to. He had spent six weeks monitoring usernames on a channel and constructing behavioral profiles from text patterns because text was the closest thing to human contact available to him.

And now there were three of them, a hundred and thirty meters away, fighting thorn-rats. The first living people he'd seen since Sacramento, since the office, since the last time he'd been in a room with another human being and had failed to appreciate that this was a thing that could be taken away, and they were losing a rat fight. Not the grand reconnection with civilization he'd been imagining. Not the careful, data-supported approach to Haven's Reach he'd been planning for weeks.

Three strangers, covered in thorn-rat bites, in a clearing he hadn't known existed, doing badly. A hundred and twenty-six days of solitude ends with: three people losing to rats. I practiced my introduction for this. I workshopped whether to lead with my name or my level. The universe has a sense of humor and it is specifically bad. He settled deeper into the broadleaf cluster and did what he always did.

He watched, and he took notes, and he did not examine the distance between what he was feeling and what he was writing down. They were dealing with a thorn-rat swarm, which was the Crimson Wilds' way of telling newcomers that even the low-threat classification required a plan. Thorn-rats were individually manageable: he'd been handling them since Week 2 without breaking stride. A swarm was a different category.

The swarm's damage was in the accumulation: twenty small hits, thirty, the Hunger meter spiking as the metabolic cost of sustained low-grade injury compounded, the specific exhausting arithmetic of fighting something that replenished faster than it died. He watched them fight it. The large one first. He was hard to miss. Broad-shouldered and thick through the chest in the way that said the Strength allocation had been building on something that was already there before AetherForge.

Maybe six-one, six-two, with the kind of dense, heavy frame that moved slowly and hit with intent. Dark hair cropped close enough that the forest hadn't had much to work with, brown skin weathered by what looked like weeks of outdoor exposure, and forearms that were currently bleeding from three separate thorn-rat bites he didn't appear to have noticed. Strength-heavy build, holding his ground at the swarm's center through sheer physical presence rather than any technique Finn could identify.

He was swinging wide and connecting on maybe forty percent of his strikes, which was not a good ratio but was producing enough force on contact that each connection counted. The kind of fighter who won through impact density rather than precision. Latino, Finn guessed from the build and coloring, though the guess was worth exactly as much as any other assessment made at a hundred and thirty meters through broadleaf cover, which was: noted, not confirmed. The young one next.

Younger than the other two by enough to be visible in how he moved, maybe nineteen or twenty, tall and lean in a way that hadn't filled out yet, dark skin and a shaved head that was growing back unevenly, as if the decision to maintain it had been made in a world with mirrors and razors and he'd been in this one long enough for that decision to become logistically impractical.

He was moving wrong: too much lateral drift, not enough ground-hold, burning Agility on repositioning that wasn't producing threat reduction. His face, from what Finn could read at distance, had the quality of someone who was processing more information than he knew what to do with, everything registering, nothing yet organized into a framework that could produce efficient action. But he wasn't stopping.

Every wrong movement was followed by another one, and the aggregate was: still in the fight, still functional, still not dead. Finn noted this separately from the wrong movements, because persistence and technique were different variables and conflating them was how you misread people. And then the smaller one.

She took him longer to read, not because she was harder to see but because what she was doing took a second to recognize, and then a second longer to accept, because recognizing it meant acknowledging something he hadn't expected to find in another person. She was maybe five-three, five-four, compact and sharp-featured in a way that read as economy rather than delicacy, everything about her built for function within a smaller frame.

Dark hair pulled back in something practical that had probably started as a ponytail and had devolved into whatever kept it out of her face during a rat fight. East Asian, Japanese or Korean from the bone structure and coloring.

She moved with an attentional quality that was different from the other two, not the big one's brute-force ground-holding, not the young one's frantic repositioning, something more measured, more deliberate, and harder to classify from a distance because what she was doing wasn't primarily about the rats. She was watching meters. Not her own. Theirs. It took him another thirty seconds to be sure.

She was managing their states: tracking the large one's injury accumulation, the young one's positioning, the swarm's center of mass, all simultaneously, and adjusting her own movement to compensate for deficits she was reading in real time. When the big one took a cluster of bites to his left side, she shifted to draw attention from his injured flank before he'd even registered the damage.

When the young one drifted too far lateral, she closed the gap he'd left, not by telling him, but by being where he should have been so the swarm couldn't exploit the opening. She was running support. Not the healing kind, she didn't have the toolkit for that at what he estimated as Level 5. The awareness kind.

The kind that came from an INT/WIL build pointed at interpersonal data rather than environmental data, from someone who had looked at three people in a fight and understood, on some level that was operating below conscious strategy, that the most useful thing she could do was see the whole battlefield and adjust for the people who couldn't. It wasn't perfect. The response time was too slow on two of the corrections, and she'd missed a positioning gap on the far side that the swarm had briefly exploited.

It was rough, unpolished, a proto-version of a skill that didn't have a name yet. But it was real. And the other two were beginning, visibly, incrementally, to lean into it, to move with slightly more confidence when she was near because something about her positioning told their hindbrain that the gap was covered even if they couldn't have explained why. He had been doing this for himself for four months. Solo.

Environmental threat data, processed and acted on in real time, the entire awareness loop closed inside one person because there was only one person. He had not expected to see it directed outward at other people by choice rather than necessity. He had not expected to recognize it so quickly, or for the recognition to feel like what it felt like, which was something adjacent to relief and something adjacent to loneliness and something he was absolutely not going to write on the bark-sheet.

Three people losing a rat fight and me up here cataloguing the tactical reasons I'm not helping, which is a thing I'm apparently doing now. Risk assessment. Unknown party. No behavioral baseline. Could be PKers. Could be the three most dangerous players in this sector running a lure I've seen work twice in the last month. The story has been taken without consent; if you see it on Amazon, report the incident. Could also just be three people losing a rat fight. I know which one I'm hoping for.

I'm not writing that down. The swarm retreated at 1131. The large one had landed three consecutive heavy strikes on the swarm's anchor cluster, the dense center where the behavioral coherence lived, and something in the swarm's collective calculation had apparently decided that the return on investment was insufficient. They pulled back.

The fight had lasted twenty-seven minutes, which was longer than he thought it should have been with three players against a standard-density swarm, and shorter than it would have been if the large one hadn't found the anchor cluster by what looked like instinct rather than strategy. They had won by a margin that made him uncomfortable.

He ran the automatic risk assessment while they were still catching their breath below him, because automatic meant it happened whether or not he had decided to do it, and he had built the automatic layer precisely so that his conscious decision-making could happen on top of a completed baseline rather than from scratch. Level range: three to six, estimated from combat performance and the Nexus-visible light around their ability use.

Behavioral markers for PK activity: absent in the initial forty-seven minutes of observation. No coordinate-solicitation pattern: he'd pulled the channel cross-reference while they were mid-fight, running the account handles he could partially read against his behavioral database. Nothing matched the movement patterns he'd catalogued for the primary cell. Assessment confidence: 41%. Insufficient. He needed more data. AETHERFORGE NEXUS 】 Surveillance Protocol — Active | Subjects: 3 | Est.

level range: 3–6 | PK markers: NONE detected (47 min.) | Assessment confidence: 41% The Watchers note you watched three people fight a thorn-rat swarm for twenty-seven minutes from a concealed position with full tactical advantage and did not intervene. They are not judging this. They are recording it, which is different from judging in ways they are not currently prepared to elaborate on. They would also like to note, for the record, that the three subjects are not PKers. They are telling you this freely.

You should trust them. You should go down there. There is no reason to wait. They are genuinely, verifiably, almost certainly not going to kill you. The Watchers understand if the phrasing of that last sentence did not land as reassuringly as intended. They would like to clarify that this is a limitation of the format rather than an indication of anything specific. Probably. Assessment confidence: 41%. The Watchers note this number could be higher if you used sources other than a bark-sheet and a ridge.

They are declining to say how much higher. They feel the uncertainty adds texture. He had not moved. This was accurate and the Watchers were welcome to find it informative. He had not moved because his protocol required an assessment and his assessment was at forty-one percent and he had a threshold and he was below it, and those facts together produced a man on a ridge watching three people recover from a thorn-rat fight instead of doing anything else.

In Sacramento, this would have been called standing at the edge of the office party pretending to check your phone until you'd scoped out the social dynamics enough to choose which conversation cluster to approach. The methodology translated across dimensions. The stakes had adjusted upward considerably.

The fundamental experience of a man observing a group of people from a safe distance and telling himself it was strategic rather than what it actually was, which was the behavior of someone who did not know how to walk up to strangers and say hello, that had remained remarkably stable. He was aware it was not the complete story.

He was choosing to finish the assessment before he examined that awareness, because examining it first was how the math stopped being the math and started being a feeling, and feelings were information but they were not the same information as forty-seven minutes of behavioral observation, and he needed both before he decided anything. They made camp at 1340, which told him something. That was early: an experienced solo player or a prepared group would have pushed another two hours of light.

That they stopped at 1340 meant either the swarm fight had cost more than he'd assessed or they were running on a conservative timing protocol he hadn't accounted for. He mapped the camp position on the bark-sheet and noted the watch cadence: the large one took first position, moving the perimeter in a wide irregular loop that was inefficient but thorough.

The small one was doing something near the fire: he couldn't see the detail from the ridge but the posture read as treatment, systematic, the same deliberate attentional quality he'd noted during the fight now directed at wound assessment rather than battlefield positioning. The young one sat with what looked like a bark-sheet equivalent and wrote in it for twenty minutes. He filed each data point in the correct column.

The young one's bark-sheet caught his attention in a way that was disproportionate to its tactical significance. It was the posture: hunched forward, writing with the concentration of someone organizing information rather than journaling, stopping periodically to reference something, making corrections. Finn recognized the body language because he'd been doing it every night for four months. The young one was building something on that sheet. A framework. A system.

It wasn't Finn's system, the approach would be different, the categories would be wrong in ways he could probably identify from ten minutes of reading, but the impulse underneath it, the conviction that writing things down was how you kept the chaos from eating you, that was the same. I am watching a man write in a notebook from a hundred and thirty meters away and having feelings about it. This is what my social life has come to. David Attenborough would narrate this with great sympathy.

"Here we observe the lone data analyst in his natural habitat, watching other humans interact from the safety of elevated cover. Note the bark-sheet in his lap and the carefully maintained expression of tactical neutrality. He has not spoken to another person in one hundred and twenty-six days. He will not speak to these ones today. He is taking notes." At the bottom of the column, where the classification header would go, he wrote: pending. He had not used that header before.

His classifications were threat, neutral, resource, or deferred. Pending was not a category he'd needed: things either went in a column or they didn't go on the bark-sheet at all, because the bark-sheet was for things he understood well enough to note. These three people were not yet something he understood well enough to note in the standard taxonomy. He wrote pending and left the header blank because the right word hadn't arrived yet.

The Exposure meter was climbing from the afternoon transit, manageable but present. Hunger at forty-four percent from the extra attentional load the surveillance had added on top of the day's travel. He ate half a ration without descending from the ridge, because descending from the ridge would have required making a decision he hadn't finished building the justification for. Forty-one percent. The threshold is eighty-five.

I have forty-four points to collect before I have to decide what I'm going to do with the answer, which is plenty of time. Plenty. I am choosing not to examine the phrasing of 'have to decide.'

【 AETHERFORGE NEXUS 】 Perception — Check | Novel application: interpersonal behavioral assessment | Cross-protocol transfer: environmental read → human subject The displacement-reading protocol he developed for thorn-serpents in broadleaf undergrowth has been successfully applied to three humans on a hillside, which is either an efficient use of existing calibration work or a data point about how he has come to categorize people. A folder has been opened. The Watchers are not naming the folder.

He dismissed the panel. The Nexus was correct: he was reading the three players the way he read terrain, which was either a sign that the skill had generalized usefully or a sign that he had been alone long enough to have nothing else to read people with. Probably both. He wasn't going to examine that tonight either. He stayed on the ridge until full dark.

The camp below him burned at a size that was slightly too large for efficient heat management, the large one's fire, he guessed, built for warmth rather than concealment. He noted this in the bark-sheet. Not a threat marker. Just data. The small one was still moving between the other two an hour after the fire was built, checking something Finn couldn't see from the ridge but could infer from the pattern: injury assessment, probably. The systematic quality was still there. She was running a protocol of her own.

It was not his protocol. It was not a protocol he'd seen anyone describe in the channel. It was hers, built from whatever her INT/WIL allocation was processing, and it was pointed at the people next to her rather than the environment around her, and he found he could not stop watching it. The young one had moved from the bark-sheet to helping the large one with something structural at the camp's edge, some kind of windbreak or lean-to assembly.

The construction was wrong in several ways Finn could identify from the ridge and one way that was actually clever, a cross-brace technique he hadn't seen before that compensated for the broadleaf's lower structural integrity. The young one had looked at a problem and solved the part of it he understood in a way that suggested his INT allocation was doing more work than his combat performance had indicated. Three people. A fighter who hit hard and didn't notice his own injuries.

A support player who watched everyone's meters but her own. A thinker who moved wrong in a fight and built things right in a camp. They were alive and functional and losing by margins that were going to get thinner if nobody helped them, and they were a hundred and thirty meters from a man who had everything they needed and a protocol that said not yet. He did not calculate how many minutes he'd watched them fail to die and succeed anyway. The number would have been exact and he had decided not to know it tonight.

He went to sleep with the pending -classification column open, feeling a little bit like the Watchers and slightly dirtier for it.