Closes it.
"When you say it like that—"
"That is exactly what happened."
Footsteps pass in the hallway. We both freeze, and for a moment I am acutely aware of how small this kitchen is, howclose she is, how her cedar and citrus perfume registers. I lower my voice.
"If we walk back into that room and tell my mother this is a misunderstanding, she will never let me forget it."
Ever. I would receive commemorative anniversary reminders. She would reference it at Eleanor's eventual grandchildren's christenings.
Tessa groans softly, tipping her head back, and I watch the line of her throat and look away immediately.
This is the problem with uncontrolled variables.
My last relationship ended because Rachel had decided, without consulting me, exactly who I was supposed to become. She had a timeline. A vision board, literally, which I discovered six months in.
This situation has the same shape.
Someone else stepping into my life's narrative without my explicit consent.
And yet it doesn't produce the same cold tightness in my stomach. It produces something else entirely, which is worse.
I can't let my guard down just because this is Tessa.
"You weren't asking me to be your fake girlfriend?"
"No," I say.
Then I pause.
I exhale slowly. I am a data scientist. I work with incomplete datasets. I can adapt.
"We can correct it later."
"When?" she asks.
"After the wedding."
The words leave my mouth before I fully analyze them. I grimace—a rare, involuntary expression. "If you're still willing."
She shrugs, and it is the most disarming thing I have ever witnessed, because this entire situation is structurally disastrous and she is treating it like a minor scheduling inconvenience.
"Temporary," I say quickly. I need this on the record.
"Temporary," she agrees.
"Strictly logistical."
"Of course."
"No emotional complications."
"Obviously."
We look at each other for half a second too long. I notice the faint amusement at the corner of her mouth and decide not to examine what that does to my pulse rate.
I will also need to establish explicit ground rules. Terms. Parameters with defined endpoints. A controlled experiment rather than whatever organic disaster this has been so far. I tell myself this is simply another data problem—social input variables, a fixed timeline, a measurable conclusion. Clean. Manageable.
