The argument is usually lost on the page

I have read other people's arguments for a living for twenty-five years. Almost all of them were written by someone who believed the facts would do the work.

They rarely do. A case is won or lost on the page long before anyone stands up, and it goes wrong in small ways: a claim that reaches further than the evidence behind it, a paragraph written for a reader who already agrees, a number offered without the comparison that would make it mean anything. Nobody tells you the moment they stopped following. You find out at the end, in the ruling.

Environmental argument has this problem badly, and it has it at a point where the cost is high. The science has been sufficient for decades. What gets contested now is whether a particular company, project or policy does what it says it does, and that is a question about evidence and language rather than about physics. It happens to be the exact question I spent a career taking apart.

Two things follow from that, and they pull against each other.

The first is that overstatement is doing real damage to the side making it. A claim that outruns its evidence hands the other side a free win, because refuting one weak sentence is much easier than engaging with a strong body of work, and most onlookers never see the difference. People building genuinely good things give away ground they did not need to give, in a paragraph they wrote quickly.

Precision on its own persuades nobody either.

That is the second thing. Writing that hedges every clause reads as doubt, and a reader who senses doubt supplies their own conclusion. The discipline is to say the strongest thing the evidence will actually carry, and then to say it plainly, which is harder than overclaiming and harder than retreating.

The gap between those two failures is narrow, and almost all the interesting work sits inside it. That is what this page is for. Policy documents that say something other than what they are reported to say, public claims worth taking apart, arguments I think are being made badly by people who are right.

These are my own views and my own reading, written in a personal capacity. The Through Line, below, is the more disciplined monthly version, written for founders and the people who write for them.