Monday arrived with the particular efficiency that Mondays had when you had given them enough structure to work with.

Yutong was at her desk by eight, which was later than she would have preferred but earlier than anyone else in the Whitmore lab, which meant the space was hers in the way she liked it: quiet, undemanding, requiring nothing except work. She had learned the rhythms of the lab in two weeks — who arrived when, who made noise, who left dishes in the sink and who cleaned them, the particular social topography of a shared research space that told you everything you needed to know about how people managed themselves when no one was formally watching.

She had learned, for instance, that Marcus arrived at nine-fifteen and immediately made tea and talked at whoever was nearby for ten minutes before settling into productive silence. That Aiko came in at eight-thirty, put her headphones on, and did not remove them until lunch. That the postdoc — whose name was Dr. James Hartley, and who had looked at her Sichuan credentials with that particular half-second of assessment — arrived at nine and spent the first thirty minutes of every day reading emails with the focused dissatisfaction of someone hoping for different news than they were receiving.

She had learned where she fit in this: early, quiet, the new variable the system was still adjusting to. She did not mind being a variable. Variables were manageable once you understood their parameters.

She made coffee. Opened the dataset she'd been running overnight. The numbers were cooperating, which she noted as a positive variable and did not celebrate, because the data that cooperated on Monday had a tendency to become complicated by Wednesday.

She had a meeting with Dr. Osei at ten. Research direction, the calendar invite said. She had prepared for it with the same thoroughness she brought to everything: current progress summary, three potential directions for the next phase, two questions she wanted answered about the lab's relationship with the department's computing cluster. She was, in every measurable way, ready.

She was not thinking about the seminar two weeks ago, or the corridor afterward, or the note in a shared document she hadn't yet received because it didn't yet exist. She was thinking about the dataset, and the coffee, and the meeting at ten.

This was, she told herself, a perfectly accurate description of her mental state.

---

Dr. Osei's office was organized with the specific controlled chaos of someone whose mind moved faster than their filing system. Papers in precise stacks that only made sense to their owner. A whiteboard covered in notation that had been partially erased and rewritten enough times that the earlier versions showed through like geological strata. Three mugs, two of which were empty and one of which appeared to contain tea from a previous meeting.

Yutong sat across from her and opened her notebook.

The first twenty minutes were straightforward: research progress, timeline, the computing cluster question which Dr. Osei answered with the brisk efficiency of someone who had navigated departmental bureaucracy long enough to know exactly who to ask for what. Yutong took notes and asked follow-up questions and felt the particular satisfaction of a conversation that was moving efficiently toward useful conclusions.

Then Dr. Osei said: "I want to talk about the joint seminar series."

"I saw the approval announcement," Yutong said. "Biweekly sessions starting next month."

"Yes. And connected to that—" Dr. Osei paused in the way that indicated what followed had been considered carefully. "Your work on protein interaction networks has a significant methodological overlap with what the Henderson lab is doing. Specifically the temporal variance modeling — the question you asked at Lin's seminar."

Yutong kept her expression neutral. "There's some overlap, yes."

"More than some. I've been looking at both datasets and I think there's a genuinely productive collaboration to be had, particularly with the conference deadline in March. If we could get the two frameworks talking to each other before then, it would strengthen both papers considerably." Dr. Osei looked at her with the direct attention of someone who had decided something and was now informing rather than asking. "I'd like you to work with Lin Sicheng on the shared methodology. He's the only person in Henderson's group with sufficient expertise in the temporal variance component — which is, not coincidentally, exactly the piece your work needs."

The room was quiet for a moment.

"Of course," Yutong said. Her voice was even. Her notebook was open. Her pen was positioned correctly.

Under the table, her fingers found the edge of her notebook and pressed, once, evenly.

She thought, briefly and without intending to, about the question she had asked at the seminar. About the way the question had emerged from genuine scientific interest and had been answered with a completeness she hadn't expected. About the corridor afterward and *another time* and the way she had walked away from both without resolving either.

She had asked the right question at the seminar. The right question had, with the characteristic irony of systems operating according to their own logic, produced this outcome.

"Excellent," said Dr. Osei, and moved on to the computing cluster allocation.

---

The email arrived at twelve-seventeen.

She saw it when she returned to her desk after lunch — a meal she had eaten efficiently and without particular attention to the food, which was her habit when she had other things to process. The subject line was: *Methodology alignment — proposed first meeting.*

She read it.

*Dear Song Yutong,*

*Following Dr. Osei's communication, I'd suggest we meet to align on methodology before beginning the collaborative work. Wednesday at 2pm in the Henderson lab conference room would work on my end, if that suits you. I've attached a preliminary framework document — review at your convenience before we meet.*

*Best,* *Lin Sicheng*

She read it again.

The email was exactly what a professional communication between two researchers beginning a collaboration should be. It was precise. It was efficient. It contained all necessary information and no unnecessary information. The attached document was thorough — she opened it and scanned the first three pages, which were well-organized and revealed a mind that thought in clear structural hierarchies.

There was nothing in the email that was personal. There was nothing that referenced the seminar, or the corridor, or the forty-seven seconds of Q&A, or *another time.* There was no acknowledgment of anything except the work.

This was, she recognized, exactly how she would have written it.

The recognition was not comfortable. It would have been easier if the email had been warmer than necessary, or cooler, or somehow wrong in a way she could use to create distance. Instead it was correct. It was the email she would have sent, and he had sent it, and she did not know what to do with the fact that after four years they were still capable of operating on the same frequency when the situation required it.

She typed: *Wednesday at 2pm works. I'll review the document beforehand.*

She sent it before she could deliberate further. Then she opened the preliminary framework document and began to read with the focused attention she brought to anything that required her to know what she was dealing with.

It was good work. She made notes in the margin of her own document — questions, points of disagreement, places where Whitmore's framework offered something Henderson's didn't account for and vice versa. By the time she reached the end, she had filled two pages and had a clearer picture of where the collaboration would be productive and where it would require careful navigation.

She did not think about Wednesday. She thought about the methodology.

This was almost entirely true.

---

The Henderson lab conference room was smaller than she had expected.

She had built a mental image from the building's general scale and the seminar room's dimensions, and had arrived at something larger — something with sufficient buffer space that proximity would be a matter of choice rather than architecture. The actual room offered a table that seated four comfortably and six if no one moved their elbows, a whiteboard along one wall, a window overlooking the courtyard, and approximately enough space between the two chairs they were using for her to be aware, with uncomfortable precision, of how much space there was.

He was already there when she arrived — three minutes early, which she had anticipated and had therefore arrived four minutes early to avoid the dynamic of entering a room where he was already in control of the space. He looked up when she came in. Neither of them remarked on the timing.

"I read the framework document," she said, setting her notebook on the table.

"I assumed you had." He moved his laptop slightly to create a more equitable division of table space. "What were your questions?"

She opened her notebook to the two pages of notes. He looked at them briefly — she caught the look, a quick assessment of the volume and organization of what she'd written — and something in his expression shifted in a way she filed without examining.

They worked.

The first thirty minutes were straightforward: terminology alignment, establishing which version of which metric they were each using, the small but significant differences in how the two labs operationalized concepts that sounded identical and weren't. This was the necessary groundwork of any collaboration and she had done it enough times to find it almost meditative — the careful mapping of one framework onto another, finding where they matched and where they diverged.

He was good at this. She had expected him to be good at this — she had read his papers and knew how he thought — but there was a difference between knowing someone's work and working with them, and the difference here was that he moved faster than she'd anticipated, adjusted his explanations in real time based on her responses, and twice caught an inconsistency in her own framework that she hadn't noticed.

She noted this. She did not say so.

What she noticed, and also did not say: working with him felt different from working with other researchers. Most collaborations required a significant investment of energy in translation — explaining not just what you meant but how you arrived at meaning, building a shared language from scratch. With him, the shared language was already there, somewhere underneath the four years and the professional distance, and the translation happened faster than it should have. She was not sure whether this was because they had worked adjacently before, in a different context, or because of something more specific to him.

She filed this observation in the same category as the eighteen inches and the look across the library and did not examine it further.

The disagreement arrived at the forty-minute mark, over the baseline stability parameter.

"You're modeling baseline stability as a fixed input," she said. "But in the systems I'm working with, baseline stability is itself a function of the interaction history. It's not a parameter, it's an output."

"In a closed system, yes." He didn't look up from his laptop. "Our systems aren't closed."

"Neither are mine."

"Then you already know the fixed input is an approximation." He looked up now. "A useful one."

"Useful for your framework. Less useful for mine."

"Which is why we're here." He said it without impatience — a simple statement of the situation. "The question is whether we can find an approach that doesn't require either framework to compromise its core assumptions."

She looked at him. "Can we?"

"I think so." He picked up the marker from the whiteboard tray. "May I?"

She gestured toward the whiteboard.

He wrote her argument first — her position, accurately represented, in three clean lines. She watched him do this and noted that he had understood it precisely, which was its own kind of unsettling. Then he wrote his counterargument beside it, and she stood to look at the whiteboard properly because the table angle wasn't sufficient, and they ended up standing side by side in front of the whiteboard with perhaps eighteen inches between them and the marker in his hand.

She read both arguments. He was right about the framework. She was right about the baseline. The two positions were not actually in conflict — they were operating at different levels of the same problem.

She picked up a second marker and drew a bracket connecting the two arguments. "This is where they interface," she said. "If we model baseline stability as a dynamic variable within your framework rather than a fixed input, you don't lose the framework's predictive power and I don't have to pretend my systems are simpler than they are."

He was quiet for a moment, reading what she'd drawn. She was aware, in the periphery of her attention, of the eighteen inches between them, and of the fact that neither of them had moved to increase it.

"That works," he said.

"I know."

He looked at her sideways — a brief, unguarded look, the kind that happened when someone was surprised into it. Then: "You're right about the baseline. I'm right about the framework. We need both."

"Yes," she said. "We do."

They stood in front of the whiteboard for a moment longer than was strictly necessary for the purpose of reviewing a methodology diagram. The window behind them showed the courtyard in the gray of a Wednesday afternoon, rain beginning again on the glass, the particular October light of Edinburgh that arrived sideways and left early. She was aware of all of it — the light, the rain, the whiteboard, the eighteen inches — with equal and involuntary precision.

Then she stepped back to the table, and he capped his marker, and they returned to the work.

---

The meeting ended at four-fifteen. They had accomplished more than she'd planned, which she noted as a positive variable and which she suspected he had also noted, from the particular quality of the silence as they packed up — not the silence of two people with nothing to say, but the silence of two people who had been doing something that worked and hadn't yet decided what to call it.

He held the conference room door open as they left. A small, unremarkable gesture — the automatic courtesy of someone raised to hold doors. She walked through without looking at him.

In the corridor, they went separate directions without discussion. She was three steps toward the Whitmore lab when she heard his footsteps stop.

She did not turn around. She continued walking.

At her desk, she opened the shared document they'd set up for the collaboration and transferred her notes from the meeting: agreed methodology, open questions, next steps. It took forty minutes and required enough concentration that she was able, for those forty minutes, to simply work.

Then she reached the end of her notes and saw that he had already added a section.

His additions were detailed and well-organized — she expected nothing less. She read through them efficiently, noting points of agreement and flagging two items for further discussion. Then she reached the last line.

*The phase transition argument from the seminar — if you still want to finish it, the methodology discussion Thursday would be a reasonable place.*

She read it once.

She read it again.

Then a third time, not because the sentence was ambiguous but because she needed a moment with the fact of it — the fact that he had put it here, in the shared document rather than an email, tucked at the end of a page of professional notes where it would not be visible to anyone who wasn't already reading carefully. Where she would find it, and he knew she would find it, and he had put it there anyway.

He was opening the door he had left closed in the hallway two weeks ago. He was doing it in the most defensible possible way — adjacent to work, deniable if necessary, phrased as a professional offer rather than anything else. And yet.

She closed the laptop.

Looked at the window. Edinburgh was doing what Edinburgh did — rain on the glass, the courtyard below gray and wet, the kind of evening that arrived early this far north in October and stayed until morning.

She thought about what it meant that he had written that sentence. She thought about what it would mean to show up to Thursday's methodology discussion and finish the argument. She thought about the whiteboard and the eighteen inches and the look he'd given her when she'd said *I know.*

She had spent four years building a version of herself that did not need to think about these things. The version was well-constructed. It had held up under conditions she had not anticipated surviving intact. She had not expected to be tested on it in a conference room in Edinburgh by a sentence at the end of a shared document.

She opened the laptop.

She did not reply to the note. She closed the document, opened her own dataset, and worked until seven-thirty, when the building had been quiet for an hour and the rain had not stopped.

On the way out, she passed the Henderson lab. The light was still on inside.

She did not stop walking. But she noted the light with the same precision she noted everything, and filed it in the category she had not yet named, and walked out into the Edinburgh rain thinking about Thursday.

The rain was heavier than it had been in the morning. She did not have an umbrella. She had not brought one because she had been focused on the meeting and not the weather, which was unlike her, and which she noted as a data point she did not yet know how to interpret.