A good deal of what I was taught as a young researcher has quietly expired. Not been refined or extended; expired. Some of it was wrong at the time and we did not know. Some of it was correct and has been superseded. Some of it was correct and remains correct but no longer matters. The uncomfortable part is that none of this announced itself. Knowledge does not notify you when it stops being true.

Everyone's answer to this is that we must keep learning, which is correct and also the easy half. Learning is pleasant. It adds, it produces a sense of progress, and nobody has ever felt diminished by reading another paper. The hard half is discarding. To let go of something you know, you have to accept that a piece of your expertise is now worthless, which is a small wound to an identity built on knowing things. And so we tend to carry obsolete knowledge forward, applying rules whose original justification evaporated years ago, usually without noticing that we are doing it.

What makes this worse with seniority is that nobody checks you. A junior person who states something outdated gets corrected. A professor states the same thing and the room writes it down. The correction mechanism that operated on you at thirty has largely switched off by fifty, exactly when the accumulated stock of expired knowledge is largest. The only remaining defence is self-administered: periodically reconsider everything you think you know and whether the conditions that made it once true still hold. So, as a reflection for today: what do you still believe mainly because it was true when you learned it?