What to do
Three replies, one for each kind
Once a claim is sorted, the reply is nearly determined. Three sentences, with the variants for the awkward cases.
Claims this part is about, and what each was answered with
Showing up for each other is part of what a team is
Value→A statement of what mattersAnswered in its own kind, which is why it went somewhere
People should be free to choose where they live
Value→A principleNew starters take longer to become productive remotely
Empirical→Evidencethree claims · one empirical and two value · every one of them was answered in its own kind
The three claims on this record that were answered in their own kind.
Sorting is only useful because it tells you what to say. Three kinds, three replies, and each one is a sentence rather than a strategy.
To an empirical claim
Is there anything we could look at?
That is the whole reply. It accepts the claim as worth investigating, commits you to the result, and makes the next step joint rather than adversarial.
Three things can come back. A specific measurement, in which case go and get it. A vague gesture at research, in which case ask what unit it measured. Or nothing, which tells both of you the claim is unexamined and should be held more lightly from here.
Variant when you think they are wrong: ask anyway, and say what you expect. "I think it will go the other way, but let us look." Committing in advance is what makes the question credible.
To a value claim
That is a real difference between us. Here is what matters most to me.
Two clauses. The first acknowledges without conceding; the second states your own value rather than attacking theirs.
What this avoids is the study, which cannot work, and the dismissal — "that is just a preference" — which is accurate and useless.
Variant when you share the value and differ on weight: "I think that matters too. For me it comes second to being there at four." Most value disagreements in this argument are about ordering rather than about whether something counts, and saying so narrows things considerably.
To a cost claim
What is the figure, and what is mine?
Short, and it does two things: it asks for the quantity, which is where the content is, and it offers your own, which prevents the exchange becoming an audit of one side.
Variant when the cost is the organisation's: "That is worth having in the open. What is it, roughly, and does it change what is possible?" A constraint acknowledged is a constraint that can be designed around.
The table
| Kind | Reply | What it gets you |
|---|---|---|
| Empirical | "Is there anything we could look at?" | A measurement, or the knowledge that there is none |
| Value | "That is a real difference. Here is mine." | A disagreement that can be traded |
| Interest | "What is the figure, and what is mine?" | The thing that was actually deciding it |
The two failure modes
Sorting out loud. "Is that an empirical claim?" is insufferable. The sorting is silent; only the reply changes.
Using the reply as a move. All three sentences fail if delivered as a technique. "Is there anything we could look at" from somebody who has already decided is heard immediately, and burns the question for later.
The sentences work because they are sincere. If you would not accept the measurement, do not ask for it; if you do not want the figure, do not request it. There is no version of this that works as a manoeuvre, which is the main thing recommending it.
What to do when the reply does not work
None of the three succeeds every time. Three failure modes and what each one tells you.
"Is there anything we could look at?" and they produce nothing, but keep making the claim. The claim was never empirical. Switch to the value reply, without pointing out the switch.
"That is a real difference between us" and they come back with more evidence. They have not recognised their own claim as a value. One further attempt, more explicit — "I do not think a study decides what work is for" — and if that fails, stop. Some people will not accept the frame, and continuing turns the discussion into one about epistemology.
"What is the figure?" and they will not give one. Either they do not know, which is useful, or the cost is private, which is their right. Offer yours anyway. An unreciprocated figure still sits in the room.
The reply that covers all three
If you can only remember one sentence, this is the one that does least harm in every case:
Say more about what you mean by that.
It is not a sorting question and it does not advance anything by itself. What it does is buy four seconds, produce more material to sort, and signal that you are listening rather than loading — which on this record was the thing most often missing from exchanges that otherwise had everything else.