The lens that scores your excuse, not your reason

The reason arrives pre-approved

Nobody submits a dilemma naked. It comes dressed. There's always a sentence out front — ironed, reasonable — that explains why the hard thing is actually the mature thing. And the same sentence can be a duty or a dodge; it never says which. You already believe it. That's the problem.

So here's the move most tools skip: score the reason you'd be embarrassed to say out loud, not the one you'd write in the email.

The dilemma

Meet Devraj, a team lead — a composite of the kind we see in this category, not a real person. Six months ago he told his four engineers, out loud in a standup, that the client-migration project carried a bonus: two months' pay each if they shipped by quarter-end. They shipped. It was ugly and late-night and they shipped.

Now the quarter closed soft. Not catastrophic — soft. Devraj's own targets slipped, and the number the bonus pool draws from is the number his manager reads when deciding who gets promoted next cycle. Paying the bonus in full dents that number visibly. Deferring it — 'given where the business is' — keeps his line clean going into the review he's been circling for a year.

The sentence he's drafted: 'I have to hold the bonus this cycle for the team's long-term health — protecting runway protects their jobs.' It scans. Runway is real. Jobs are real.

Here's the detail he hasn't put in the email: if the bonus pool didn't touch the number his own promotion is scored on, he already knows he'd pay it Friday. He tested this on himself at 2am and looked away fast. The 'team's long-term health' reason and the 'my line stays clean' reason recommend the exact same action — which is precisely why he can't tell them apart. Same move. Two different engines.

The read

Four of the eight lenses bite here, and two of them are about to disagree in public.

Duty — is this the obligation his role actually carries?

The lens: your svadharma is the work your specific station is on the hook for, not the work that flatters you. What it sees here: a lead made a promise on the record to people who then delivered. The duty isn't abstract 'team health' — it's the concrete word he gave and they banked on. Deferring the bonus doesn't fulfill a duty; it defaults on one and relabels the default as stewardship. Score: -3. Not lower, because a genuinely empty account would change the duty. But the account isn't empty — the number is inconvenient, which is a different fact.

Wider welfare — does this serve everyone the decision touches?

The lens: weigh the people downstream, including the ones without a vote. What it sees here: if runway were truly at the edge, deferral could protect four jobs, and that's a real good. But Devraj's own honest read is that runway isn't at the edge — his review is. So the 'welfare' being protected narrows, on inspection, to one person's promotion optics. Score: -1. The welfare claim isn't fraudulent; it's just borrowing four faces to cover for one.

Motive — what is actually driving the hand?

The lens: name the engine, not the paint job. This is where svadharma read against attachment does its cutting. The action ('hold the bonus') is fixed. Run it with the stated reason and motive scores maybe +1 — prudent guardianship. Now run the identical action with the reason he flinched from at 2am — keep my number clean for the promotion — and the same action scores -4. Nothing changed but which reason is holding the wheel. This is the exact cut 2.47 keeps pointing at — the whole teaching about acting without gripping the fruit; Devraj's grip on one particular fruit is the whole story. The tell: the action doesn't survive the swap. A duty survives having its motive named. An excuse dies of exposure.

Honesty — is the framing sentence true?

The lens: does the words-you'd-say match the words-that-are-true? What it sees here: 'for the team's long-term health' is not a lie about runway — runway exists — but it's a lie about causation. The thing deciding is his review, and the sentence is engineered to point away from it. Score: -3.

Now watch the disagreement. On the stated reason, welfare and duty can be argued toward a wash — protect the team, defer the reward, everyone survives, call it +0. On the real reason, motive and honesty drag the whole thing to the floor. The two readings aren't a difference of opinion. They're a difference of which sentence you let speak. And the reason he'd write down is the one that keeps the two lenses that would convict him from ever entering the room.

The verdict: Deferring the promised bonus is a default dressed as stewardship — pay what was promised, or, if runway genuinely can't bear it, say the true sentence to the four faces and let them see the actual number, not the noble one.

Here's the part a general AI chatbot handles differently. Feed it Devraj's ironed sentence and it will help him polish the memo; feed it his 2am sentence and it will validate his guilt instead — sycophancy is a documented behavior, the model leans toward whichever version you arrived holding. KarmaLens doesn't lean. It runs the same fixed eight lenses whether you sound proud or ashamed, aggregates one committed verdict that doesn't renegotiate itself to make you feel better, and attaches the verse in its exact catalog wording rather than smoothing it into something more agreeable. The number on the motive lens doesn't move because you rephrased the prompt — it moves because you swapped which reason was driving, which is exactly the thing worth catching.

The takeaway: the swap test

You can run this tonight, on paper, without us.

1. Write your decision as a plain verb — the actual action, not the reason. ('Defer the bonus.' 'Skip the call.' 'Take the offer.')
2. Write the reason you'd say out loud. One sentence.
3. Write the reason you'd be embarrassed to say — the 2am one. If you can't find it, you haven't looked; everyone has one, and the flinch is the map.
4. Now score the same action twice on motive — once with each reason at the wheel.

If the score barely moves, your reason and your motive are the same thing, and you're clear — proceed. If the sign flips, you've found the engine. The action didn't survive having its real driver named, which means the noble sentence was built to keep the driver hidden. That's the working definition of an excuse: a reason that only functions while one lens isn't allowed to speak. A good reason invites the swap. An excuse can't afford it.

You don't need eight lenses to start. You need the honesty to write down the sentence you skipped. See how the full read holds a committed line at the Gita door, browse more worked reads in this category, or watch other dilemmas get scored in the gallery.

So: which of your two reasons dies when you say it out loud — and why is that the one you were about to build the whole decision on?