The evidence
How to ask what would settle it
The most useful question in the argument, four ways to phrase it badly, and the two phrasings that get answered.
Claims this part is about, and what each was answered with
New starters take longer to become productive remotely
Empirical→EvidenceThe one claim answered with evidence that measured it
Productivity fell when we went remote
Empirical→DataRemote workers put in longer hours than office workers
Empirical·Nothing; it was never answeredthree claims · three empirical · none of the answered ones was answered in the wrong kind · one was never answered at all
Three empirical claims with three different fates.
The question is simple and the phrasing decides whether it works.
What would show that?
Asked well it produces a measurement, a concession, or a useful admission that nothing could show it. Asked badly it produces a defensive restatement and costs you the next three exchanges.
Four phrasings that fail
"Do you have any evidence for that?" Heard as you are making things up. The person defends their credibility rather than examining the claim.
"That's not evidence-based." A verdict rather than a question. Nothing to answer except by disagreeing.
"Actually the research shows…" Not a question at all. Starts a citation exchange, which is the thing to avoid.
"How would you know?" Would be fine in writing and lands as contempt out loud, because of the stress pattern. Small differences in phrasing matter more here than anywhere else in this subject.
Two that work
"Is there anything we could look at?"
The we is doing most of it. The question becomes a joint enquiry rather than a challenge, the claim is treated as worth investigating, and the implied answer is that you would accept the result.
"What would change your mind about that?"
Slower and better. It is not about evidence at all on its face, which is why it gets a straight answer. Three outcomes, all useful:
| They answer | What it tells you |
|---|---|
| A specific finding | Genuinely empirical — go and look |
| "Nothing would" | It is a value — stop citing studies |
| They have not thought about it | Often the most honest, and often the start of the useful part |
When to ask it
Early, and about a claim you might actually be wrong about.
Asked for the first time at week six, at the opponent's strongest claim, it is a manoeuvre and is received as one. Asked in week one at a claim you half-believe yourself, it is a question, and it sets the register for everything after.
The strongest version is to ask it of your own claim, out loud, first:
I think people learn faster near each other. What would show that? Actually — time to first delivery by cohort would, and we have that. Let me look before I say it again.
Nobody can refuse to play after that, and it costs one claim you were going to have to defend anyway.
The answer to expect most often
On this record, the commonest honest answer to "what would show that" was some version of nothing we have.
That is not a failure of the question. It is the single most valuable thing the question produces, because it moves a claim out of the category where more argument helps and into the category where only a trade does. Four of the nine empirical claims here ended up there, and each one that does is an exchange that can stop.
The version for a document
In writing the question has a different and better form, because nothing is at stake in the moment and the person can think.
You mention that hybrid teams decide worse. Is there anything we could look at on that? I am not being difficult — if there is a measurement I would rather we had it than argue about it.
The second sentence is doing the work. In writing, a question about evidence reads colder than it was meant, and one clause of disclaimed intent fixes it.
What to do with the answer
Three answers, three next moves, and the third is the one people handle worst.
| They say | Do this |
|---|---|
| A specific measurement | Offer to produce it, by a date |
| "Nothing would change my mind" | Stop citing studies at them, permanently |
| "I am not sure" | Leave it; come back to it in writing later |
The second answer is a gift and is usually received as an admission. It is not: it is somebody telling you accurately what kind of claim they are making, which is the thing you asked for.
The right response is to thank them and change register, not to point out that they have just conceded something. On this record the one time somebody answered that way, the questioner treated it as a win, said so, and the conversation did not recover.