HUME'S CRITERIA · FREEDOM FROM PREJUDICE
Freedom from prejudice
Of Hume's five criteria, this is the one about contamination. A judge, he argued, must keep the mind "free from all prejudice" and let nothing into the verdict except the object itself — not the author's reputation, not the fashion of the moment, not loyalty, not rivalry. The judgment should belong to the work, and works don't have names until someone attaches one.
Hume was blunt about how rarely anyone manages this. Reputation arrives before the art does; by the time you press play on an acclaimed record, the acclaim has already voted. The striking thing is that in 1757 he described what is now a replicated experimental finding: attach a prestigious label to a work and evaluations move, even when the label is false. Wine tastes better wearing an expensive price tag; the same manuscript reads worse under an unknown byline.
Most people, asked whether they judge music by the name on it, say no. That answer is worthless — not because people lie, but because prejudice doesn't announce itself to the person having it. The only honest way to know is to be caught in the act.
That is the entire design brief of the Prestige Test: same clips, rated blind and then labeled, with some labels deliberately swapped. When your rating follows a false name, prejudice is the only suspect left in the room. The gap between your two passes is Hume's criterion turned into a number — and because you are your own control, the number never depends on anyone's opinion of what the "right" rating was.
Freedom from prejudice is the first criterion the gym measures, but it is one of five. The others — delicacy, practice, comparison, and good sense — each have a machine of their own.
Questions, answered straight
- What did Hume mean by freedom from prejudice?
- In 'Of the Standard of Taste' (1757), Hume argued a critic must set aside everything about the work except the work — reputation, fashion, friendship, rivalry — and judge only what is in front of them. A judgment moved by the author's name rather than the object is, in his account, corrupted.
- Can prestige bias be measured?
- Yes, with a within-subject design: the same person rates the same works with and without labels, and some labels are deliberately false. The rating shift attributable to the label is a measurable quantity. You serve as your own control, so no external ground truth about the music's quality is needed.