The lens that asks: who did you leave out of the room?

You open the job offer at the kitchen table. Remote-first, better pay, the work you actually wanted. You already know the answer, so you phrase the question as a courtesy: my growth versus my boss's needs. Two people in the room. You and the one you feel slightly bad about.

That framing isn't humble. It's an edit.

The dilemma

Here's a composite — the kind we see often enough in this category that the shape is familiar, assembled from several real patterns, no real person behind it.

Call her Shruti. Eleven years into a career, mid-level, respected. A remote-first offer lands: 30% more, a function she's wanted to own, no more two-hour daily commute. On paper it is not close. She drafts her thinking as a clean two-body problem — I owe my growth to myself; I owe my manager Dinesh a decent handover. She asks whether leaving now, three weeks before a reorg, is defensible.

Then the ugly detail she didn't put in the first draft. Two of her reports, Farid and Anya, have promotion cases in motion — cases that were built on the headcount and scope she defended for them, cases that stall the second she walks, because the incoming reorg will fold her team under a manager who doesn't know their work. They didn't apply for her decision. They'll pay for it.

And one more, further out of frame. Her partner's mother relies on a weekday care arrangement that quietly runs on this household's current schedule — Shruti's predictable 9-to-5, the commute window that happens to cover a handoff. Remote-first changes her hours, her travel, the whole domestic clock. Nobody in that house has said it out loud. It's just assumed, the way load-bearing things usually are.

Shruti's question — me versus Dinesh — has four people in it. She named one.

The read

Walk it through the eight lenses and watch something strange happen: almost all of them wave her through, and one changes the shape of the go without touching the go itself.

Duty (svadharma): the lens that asks whether this is your work to do. It reads clean. This role fits her nature and her craft better than the one she's in. Staying out of borrowed obligation isn't loyalty; it's avoidance wearing loyalty's coat. Pull: go.

Honesty: the lens that asks whether the account you're giving matches the facts. Interesting here, because her first draft was honest about Dinesh and silent about the other three. A true sentence can still be a lie by omission. The lens doesn't catch her lying — it catches her cropping. Pull: tell the whole frame, starting with yourself.

Detachment: the lens that asks whether you're clinging to an outcome or a self-image. She's not clinging to the old job; she's mildly clinging to being the person who left cleanly, the hero of a tidy story. Pull: drop the need for the exit to look painless.

And now the one that bites. Non-harm (ahimsa): the lens that asks who absorbs the consequence of your action — and whether they got a say. This is the lens that refuses the two-body frame on principle. It doesn't ask is this good for you. It asks who is downstream of this, and are they in the room.

What it sees here: Farid and Anya's promotions aren't a side effect, they're a direct cost, transferred to people who built nothing wrong and get no vote. The care arrangement isn't Shruti's to unilaterally re-time by accepting a schedule change on a Tuesday. These aren't reasons to decline. They're people she owes a conversation to before the signature, not an apology after.

Here is the tension, named plainly: duty and detachment both say go now, cleanly, you've earned it. Non-harm says the clean exit you're picturing is clean for you because you deleted the three people it isn't clean for. The lenses don't disagree about the destination. They disagree about what counts as an honest departure.

Notice what non-harm is not doing. It isn't calling her selfish. It isn't demanding she martyr the offer. It's doing something narrower and harder: insisting that a decision affecting four people cannot be made by negotiating with one and informing the rest.

The verdict: take the job — and do not take it yet. Before you accept, have three conversations you were about to skip: tell Dinesh with enough runway to protect Farid and Anya's cases (or to help you protect them from the outside), ask your partner the care question out loud instead of assuming the schedule will absorb it, and only then sign. The move survives all four lenses. The clean exit does not — and it was never the thing you were owed.

Why this isn't a chatbot agreeing with you

Paste Shruti's first draft — my growth versus Dinesh's needs — into a general AI chatbot and watch it argue the frame you handed it. Lean toward going and it validates your growth; sulk about guilt and it reassures you Dinesh will be fine. These systems tend to optimize for your nod, so they mirror back the version you walked in with. What they almost never do is add people to your frame you didn't mention, because you didn't mention them. A KarmaLens reading asks for the facts that would change the answer first — including who is downstream — then makes one committed call and holds it. It won't renegotiate the verdict because you flinched at the three conversations. It lays the options beside the call with what each one costs and who carries it, and it quotes a verse only where one actually fits. The point isn't that we're neutral. The point is that we won't crop your dilemma to match the version you walked in wanting blessed.

The takeaway: the four-chair test

You can run this tonight, no account needed.

1. Write your dilemma in one sentence, the way you'd text it to a friend. Count the people in it. Most of us write two.
2. Now finish this line, out loud: "If I do this, the consequence lands on ___ — and did they get a vote?" Keep finishing it until you run out of names. Each name is a chair at the table you'd set.
3. For every name that wasn't in your original sentence, ask the single diagnostic question: do they owe me a reaction, or do I owe them a conversation?

A name that owes you nothing and gets a conversation before the decision — that's a cost you're carrying honestly. A name you were planning to inform after, or hoping wouldn't notice, is a cost you were quietly transferring. The test doesn't tell you to stay or go. It tells you whether your 'go' was shaped by the whole room or by the two people you were willing to look at.

The self-flattering account always has exactly the right number of characters to make you the reasonable one. Scripture's non-harm lens keeps asking the same unflattering question: who else is standing in this room, and why did you write them out?

Browse more worked dilemmas in the gallery, or read the rest of this category. When you set the table for your own decision — who did you leave out?