The language of criticising carefully
Eight things people say in this kind of discussion. Match each to the move it makes.
Two people criticising carefully
A story about a company, or a story about a regulator
A: My criticism is about framing rather than accuracy. Almost nothing in the coverage was false.
B: Which coverage?
A: Fair. The four national papers I read, over about ten days in March. Not broadcast, and not local, because I did not look at either.
B: Go on.
A: Every one of them treated it as a story about a company. It was a story about a regulator that had done nothing for six years — and the company is the part you can photograph.
B: That is a claim about editors, though. Are you saying editors chose that?
A: No, and this is where I have to be careful. I am saying that the story which is available is the one with a photograph in it, and that a regulator has no photograph. Nobody has to choose anything for that to happen.
B: Then the criticism is about the form rather than about the people.
A: It is. Which makes it less satisfying and, I think, considerably more likely to be true.
B: I would add one thing that damages your case slightly. The local papers you did not read ran it as a regulator story for a fortnight before anybody else — and one of them then closed. So it is not that the story could not be told.
A: No. It is that it could not be told profitably.
Three questions.
1. What does A do when asked which coverage?
2. What does A refuse to claim?
3. What does B’s last contribution do?
The four moves
1 · Name the class before anybody asks. Four national papers, ten days in March, no broadcast. A criticism with no class named is not yet a criticism.
2 · Refuse the version that needs people to have chosen. Nobody has to choose anything for that to happen. Structural claims are harder to dismiss and usually truer.
3 · Accept a scope condition, including one that damages you. Volunteering the fact that weakens your case is the move that makes the rest of it credible.
4 · Narrow to what survives. It could be told; it could not be told profitably.
Six lines of one criticism, out of order. Put them into the order that builds.
- So the pressure is structural, and it would apply to anybody making the same programme.
- All three have stopped naming the studies they describe, at some point in the last two years.
- Which leaves a narrow claim: this format appears to select against a habit that written journalism can afford.
- My criticism is about three podcasts I listen to weekly, and not about podcasts in general.
- A citation costs eight seconds of audio and buys nothing that a listener can hear.
- I am not saying anybody decided to stop; I doubt very much that anybody did.
Two minutes of assembly
One word per gap. Eight lines you will need in the next twenty-six minutes.
Is that a claim about people, about the form?
That was true in March; I have not looked .
Does the class have that property, or some members?
Nobody had to anything for that to happen.
It could be told; it could not be told .
I read four papers about ten days.
I am not claiming that editors it.
people now get their news from a feed rather than from a front page.
Three questions.
1. What has to be named before a criticism is a criticism?
2. Somebody asks whether your claim is about people or about the form. What is the stronger answer?
3. You know a fact that damages your own case. What do you do with it?
The criticism, under the rule
How the twenty-six minutes run. Five alone to prepare, fifteen in fours under the rule, six to write the claim register. The rule: anybody may say “which class?” at any point, and the speaker names the class and says whether the class has the property or only some of its members do.
Stage 1 · Alone · 5 minutes · Prepare
Something you actually consume and have a criticism of: a newspaper, a podcast, a channel, a platform, a genre of video, a kind of headline, a type of coverage.
Four lines. Line one is the one that will be challenged first.
- 1 · The class, and the sample. What exactly, over what period, and what you did not look at.
- 2 · The criticism, in one sentence, with the correct generic pattern for the class.
- 3 · The structural version: what would produce this even if nobody chose it.
- 4 · One fact that damages your own case. Write it down now.
Stage 2 · Fours · 15 minutes · The discussion
Four criticisms, about three minutes each, with the rule running throughout.
Speak, be stopped, name the class, and volunteer the damaging fact before anybody finds it.
- Speaker: open with the class and the sample. Then the criticism, then the structural version.
- Anybody: “which class?” — at any point, on any general statement. Answered in one sentence: what class, and does the class have the property or only some members?
- Also allowed: “people, or form?” — which separates a claim about decisions from a claim about structure.
- Every speaker must volunteer their damaging fact before their three minutes are up. A criticism that survives its own counter-evidence is worth four that have not met any.
Stage 3 · Fours · 6 minutes · The claim register
Write down the four criticisms that survived, in full sentences, with the right articles.
Ninety seconds each. Every generic in the correct pattern, and every class named.
- Write each surviving claim out. Name the class, and choose the pattern deliberately — zero plus plural, the plus singular for a genuine type, a plus singular only for a definition.
- Every person in the four has to accept the wording, including the article choices.
- Underneath each, one line: what would change our minds about this?
- Finally, circle any the plus singular you have written and check it once more: is it a type, or is it a lot of people?
What the rule caught
Two questions.
1. How often were you stopped with which class?
2. What happened when you volunteered the damaging fact?
The class, and the sample
Two questions. Twenty seconds.
1. What does which class? ask for?
2. Which version of a criticism is harder to dismiss?
Next. The final sustained treatment of the article system — the two habits that create feedback where there is none, and the compound-noun stress that gives the game away even when the articles are right.