Socrates' 2,400-Year-Old Meeting Technique: Six Questions That Beat Every Modern Framework

In 399 BC, an Athenian jury sentenced Socrates to death. The formal charges were impiety and corrupting the youth — and part of that second charge was the questioning itself: young Athenians had taken to imitating him, examining their elders in public, and the men so examined blamed the teacher. Politicians could not state the principles behind their policies. Generals could not define courage. Their authority rested on assumed expertise — and assumed expertise evaporates under systematic questioning. But the charge sheet could not say everything it meant. An amnesty passed in 403 BC barred prosecution for conduct under the Thirty Tyrants, and Socrates' closest associates had been Alcibiades, who defected to Sparta, and Critias, who led the Thirty — both named by the accuser, Xenophon reports. Impiety was the admissible charge for a grievance that was also political.

Athens killed the man; the method proved harder to execute. Twenty-four centuries later, its recognizable descendants run in Netflix's proposal process, Berkshire Hathaway's investment discipline, and every consulting firm's hypothesis tree — because the problem Socrates worked on is permanent: human beings are far better at producing confident assertions than at testing them.

Now apply the test where it hurts. In your last strategy meeting, how many claims were asserted — and how many were asked to show their justification? This piece covers what Socrates actually did (it is more precise than asking lots of questions), the six question types worth learning, the companies that turned questioning into procedure, and the honest answer to the objection every operator raises: who has the time?

What Socrates Actually Did

The common misreading is that the Socratic method means playing devil's advocate. The practice Socrates called elenchus is more surgical: the subject asserts a claim; the questioner draws out the premises the claim depends on; the questioning then shows those premises to be undefined, untested, or in contradiction — and the subject revises. The purpose is not to win. It is to replace confident ignorance with an accurate picture of what is actually known.

Plato's Apology preserves the founding joke of the method: told that the oracle at Delphi had called him the wisest of men, Socrates went looking for someone wiser and discovered that the experts he questioned did not know what they thought they knew — while he, at least, did not think he knew what he didn't. In a meeting, that maps to the most valuable sentence a decision-maker can hear: we do not actually know this yet. Every claim in a room divides into justified, testable, or merely asserted by someone with status — and without structured questioning, the third category wins by default.

The Six Question Types

The taxonomy most teachers use comes from critical-thinking scholar Richard Paul: six categories, each targeting a different reasoning weakness. Here is each with its business-meeting form:

Why Modern Meetings Select Against It

If the method is this effective, why is it rare? Three structural forces work against it. The HiPPO effect: status hierarchies price questions to authority as insubordination, so the highest-paid opinion travels unexamined. Meeting theater: many meetings ratify pre-made decisions, and genuine questioning delays the performance. And time pressure: questioning takes minutes a packed agenda doesn't grant — decisiveness is rewarded in the moment, decision quality only later.

Notice that all three are social and structural, not intellectual. Which is why the fix is structural too — the same conclusion the bias research reaches from the other direction: you don't get questioning by hiring braver people; you get it by making questioning the procedure.

The Companies That Made Questioning Procedural

But Who Has the Time?

The objection is fair, and the honest answer is: not every decision. Socratic questioning is deliberate-mode thinking, and running it on routine, reversible calls is how process becomes theater. The triage rule from the satisficing playbook applies directly — two-way-door decisions get speed; the one-way doors, the consequential and hard-to-reverse calls, are where an hour of structured questioning is the cheapest insurance available.

And done properly it is bounded, not endless. Elenchus has a terminus: it ends when the assumptions are on the table and each has been sorted into justified, needs testing, or unsupported. A time-boxed session with the two rules below regularly finishes in under an hour — the first ten minutes spent, Socratically, on establishing what question is actually being decided. Most meetings skip exactly that step and pay for it downstream.

Running It: The Facilitator and the Two Rules

The facilitator's role is what Socrates called midwifery — maieutic: drawing out the group's knowledge rather than inserting your own. The facilitator asks, does not assert, and cannot be a stakeholder in the outcome; a questioner with a preferred answer is an advocate in disguise.

Two rules carry the session. No assertion without evidence — any claim must name the data, case, or reasoning behind it; I think customers will love this is a hypothesis, not an argument. And no dismissal without engagement — you may not reject a position without first engaging its strongest form, which is the steelmanning discipline applied in the room. Together they change what it costs to speak: assertions get expensive, reasoned dissent gets cheap. That inversion is the whole method.

The 2025 Footnote: A Socratic Machine in the Room

The method's newest students are software. Researchers are now prototyping conversational agents designed to play the Socratic questioner inside group decision-making — challenging emerging consensus and prompting justification precisely because human participants are reluctant to (see, e.g., the 2025 preprint on conversational agents as catalysts for critical thinking, arXiv:2503.14263). Early days — but the design insight matches Socrates' own: the questioner is most useful when it has no stake and no fear.

You don't need to wait for that research to mature, because the structural version already works: make the questions part of the format decisions travel in, and the format asks them every time, of everyone, without a facilitator in the room.

Where Argumentree Fits

Argumentree is the Socratic structure as a data model. Every discussion begins with an explicit central question — the step meetings skip. Every argument declares its stance and attaches its evidence, so an unsupported assertion is visible as exactly that. The AI argument check screens each submission for duplication, toxicity and logical relation to the claim it addresses, and ratings let the group weigh arguments on their merits rather than their source.

The result is a persistent, stake-free questioner: the format itself asks what do you mean, what is your evidence, and what follows — in every discussion, without needing a Socrates on staff.

The Verdict, Reversed

Athens judged that systematic questioning was too disruptive to tolerate. Modern organizations quietly reach the same verdict every week — not with hemlock, but with agendas that leave no room for what do we actually know? The companies that reversed the verdict did it structurally: they made the questions procedure, so no one has to be brave to ask them.

Your next consequential decision will contain assertions wearing the costume of knowledge. The method for telling them apart has been available for twenty-four centuries. The only modern addition is making it nobody's job in particular — and therefore everyone's.