Disagreeing Experts

Disagreeing Experts Skill

Ask one professional and you get their profession's answer: the lawyer sees exposure, the accountant sees the tax year, the therapist sees what the fight is actually about — and each is right inside their frame while the frames point different directions. Ask an AI and you usually get something worse: all three blended into an even-handed paste where the conflicts, which were the information, have been smoothed away. This skill refuses the blend. It picks the three professionals most likely to disagree about your situation, has each read it cold and rule independently — verdict, reasoning, what they would ask first, what they would charge you to stop doing — and then puts the disagreements in a table, named and explained. Because where experts disagree is exactly where your real decision lives.

What This Skill Produces

• The panel selection — three professions chosen for maximum productive disagreement about this situation, with the reason each earned a seat
• Three independent readings — each expert's verdict and reasoning in their own professional register, written as if the others do not exist
• The disagreement table — every point of conflict: who says what, why their frames diverge, and what fact would settle it
• The convergence list — the short set all three agree on, which is as informative as the conflicts and usually actionable immediately
• The first-question set — what each expert would ask before advising, since the questions reveal each frame's blind spots in the others
• The decision map — which expert to weight for which sub-decision, replacing "who is right" with "right about what"

Required Inputs

Ask for these if not provided:

• The situation — in full, messy is fine; each expert will extract what their frame considers relevant, and the differences in extraction are part of the output
• The decision in play — what actually gets decided after this, so the readings aim at something
• The panel, if the requester wants to choose — otherwise the skill selects for disagreement and says why; a panel that will agree is a wasted panel
• The advice already received — whose frame has already been heard, so the panel can be weighted toward the unheard frames
• The constraints — anything already fixed (deadlines, budgets, relationships that must survive), which the experts must respect rather than advise away

Framework: Select for Conflict, Rule Independently, Surface the Fault Lines

1. Choose the panel for disagreement, not coverage. The useful panel holds frames in genuine tension for this case: risk versus growth, this-year versus decade, the relationship versus the transaction. Three experts who would nod at each other teach nothing; the selection step must be able to say what each pair will fight about before the readings are written.
2. Write each reading in isolation and in register. Each expert reads the situation cold — no references to the other readings, no pre-softening. The lawyer writes like a lawyer (exposure, worst-case, paper trail); the therapist writes like a therapist (patterns, needs, what is not being said). The registers are kept distinct on purpose: the vocabulary differences carry information.
3. Let each expert be professionally biased. The accountant is allowed to care about the tax consequence more than the marriage; that is what accountants are for. The bias is labelled, never corrected — correcting it re-blends the paste.
4. Build the disagreement table as the centrepiece. For each conflict: the positions, the frame-reason they diverge (different time horizons? different definitions of risk? different clients — the lawyer protects you, the therapist protects the relationship?), and the fact or value judgment that would settle it. Some rows settle with information; some are genuinely values — the table says which.
5. Extract the convergence honestly. When three hostile frames agree on something, that something is load-bearing. The convergence list is short, and each item on it can usually be acted on today regardless of the rest.
6. Map experts to sub-decisions. The verdict is never "the lawyer wins." It is: weight the lawyer on the signature, the accountant on the timing, the therapist on the conversation — with the requester left holding the values calls the table exposed, which is where they belonged.