Coming back to a conclusion
Why each piece says what would change our minds, and why we return to our conclusions once events have had their say.
A conclusion is a claim about the world, and sooner or later the world answers it. Research that never goes back to hear the answer cannot learn from it. So we write our conclusions in a form that can be checked, and then we check them.
Say what would prove it wrong
A conclusion is only useful if it could turn out to be wrong. “This could go either way” is never wrong, and it helps no one decide anything. So when a piece reaches a conclusion, it says so plainly, gives the date, and names what we would expect to see if it were mistaken.
Writing this down first matters more than it seems. Without it, almost any outcome can be explained after the fact, and every explanation sounds as if it had been the plan all along.
State the limits
Each piece also says what the evidence cannot tell us. The limits that come up most often are plain ones. The history may be too short to tell a pattern from luck. The record may have gaps, or may have changed its definitions along the way. Other explanations may fit the same facts as well. And two things moving together does not show that one causes the other.
Stating limits is not modesty for its own sake. It tells a reader how much weight a conclusion can bear before they lean on it.
Return on a date
When a conclusion depends on what happens next, we set a date to come back to it. On that date we compare what we wrote with what happened, and we note the result, including when we were wrong.
A wrong conclusion, looked at honestly, is often more useful than a right one. It shows which part of the reasoning failed, and that part can then be fixed. Over time, these returns show where our reasoning tends to go astray, which is how the work improves.
A conclusion that is never checked is only an opinion with footnotes.
