Kinds
Why the sorting is never done
Eleven weeks, twenty-four claims, four sessions, and nobody asked what kind of claim anything was. Four reasons, none of them about intelligence.
Claims this part is about, and what each was answered with
Teams that sit together ship faster
Empirical→A studyWork should be something you go to
Value≠A studyA statement about what work is, answered with a productivity paper
The lease runs to 2031 and the rent review falls next year
Interest→A statement about who paysSaid once, in a finance update
three claims · one empirical, one value and one interest · one was answered with the wrong kind of answer
One claim of each kind, from the twenty-four counted here.
The striking thing about the record is not that claims were answered in the wrong kind. It is that in eleven weeks, across a proposal document, two all-hands sessions, a written consultation and three team meetings, nobody once asked what kind of claim anything was.
These were not unsophisticated people. Several of them sort claims for a living in other contexts. Four reasons, and they compound.
One: the claim arrives inside a response
Claims in an argument almost never arrive alone. They arrive attached to a reply to something else, which means the listener is already holding a position when the new claim lands.
A person in the middle of defending something does not pause to classify the thing being said to them. They look for the quickest available answer, and the quickest available answer is whatever they already have — which is usually of the wrong kind, because what they already have is their own position restated.
Two: it feels like a delay
Sorting is a step before answering, and in a meeting a step before answering reads as not having an answer.
The social dynamics of a consultation reward fluency. Somebody who says "let me think about what kind of claim that is" is heard as hesitating, and somebody who says "well, the research is clear" is heard as prepared — even when the research is not about the claim at all.
Three: nobody owns the quality of the argument
Everyone in the room owns their position. Nobody owns whether the discussion is working.
A chair owns time, an agenda and minutes. No one owns the question of whether the exchange that just happened could possibly have resolved anything. So a crossed answer passes without comment, because commenting on it is not anybody's job and doing it unasked looks like point-scoring.
Four: the sorting is unflattering to both sides
This is the one that keeps it from happening even when somebody thinks of it.
Sorting honestly means conceding that some of your own best material is of the wrong kind. The manager has to admit that "a company is a place" is a value and not a finding. The employee has to admit that "nobody is less productive at home" is empirical and that they have not checked it either.
Nobody enters a consultation wanting to do that, and the first person to do it unilaterally gives something up while the other side gives up nothing.
What it would have taken
Not a facilitator and not a framework. One person with standing, once, near the beginning, saying something like:
Before we go further: some of what we are going to say is about what the evidence shows, some is about what we think matters, and some is about who pays for what. Those are three different conversations and they do not answer each other. Can we try to say which one we are having?
Forty seconds. On this record nobody said it, and the eleven weeks that followed produced twelve crossed answers, five claims that were never answered at all, and a decision that none of the stated reasons accounts for.
The one thing an individual can do
You cannot make a room sort its claims. You can sort your own before you speak, which costs nothing and nobody notices.
The effect is modest and real: your own contributions stop being crossed answers, and over a few meetings people start responding to you in kind, because that is what they are being handed.
The two moments when it is easiest
Not in the middle of an exchange. Two other moments, both of them quiet.
Before the discussion, in the document. A proposal that sorts its own claims is extremely hard to answer badly. "Here is what we think the evidence shows, and what we would look at; here is what we think matters, which is a view; here is what it costs and who bears it." A reply to that document has to engage with three separate things, and it is obvious when it has only engaged with one.
After the discussion, in the summary. A note that says "we disagreed about two things that evidence could settle and two things it cannot" is a more useful record than one listing what everybody said. It also makes the next session shorter, because the settled-by- evidence items can be taken away and actually checked.
Both of those are available to whoever writes the document or the summary, which is usually not the most senior person in the room.
Why the problem is not consultation design
It is tempting to conclude that the answer is a better process — structured facilitation, a framework, a form with three boxes.
On the evidence here that would not have helped much and might have hurt. The four reasons above are about what happens inside an exchange and what it costs socially to interrupt one. A form does not change either; it adds a step people fill in afterwards to match what they already said.
What changed things, in the one meeting on this record where an exchange visibly worked, was a single person responding in kind to a value claim instead of reaching for a study. Nothing was announced. The other person noticed, and the next three exchanges were better.
That is the whole mechanism available to an individual, and it is worth more than the framework because it does not require anybody's permission.