Is There a Standard of Taste?

Hume begins his essay of 1757 by giving the sceptic everything. Beauty, he writes, is no quality in things themselves; it exists merely in the mind which contemplates them, and each mind perceives a different beauty. To seek the real beauty is as fruitless as to seek the real sweet or the real bitter. Having conceded that, he spends the rest of the essay explaining why nobody can live with it.

The argument he makes is a simple one and it has never been improved on. Whoever asserted an equality of genius between a forgotten pamphleteer and Milton would be thought to defend no less an extravagance than if he had maintained a molehill to be as high as a mountain. Nobody shrugs and says that tastes differ. We judge such a person wrong, and Hume thinks we are right to. So there is a standard, and the theory has just denied that there is anywhere for it to live.

His solution is to move the question from the object to the judge. The standard of taste is the joint verdict of qualified critics, and he specifies the qualifications with some care. Delicacy: the ability to register small ingredients in a complex whole. Practice: you cannot judge a form you have met once. Comparison: without it you will overrate whatever you happened to encounter first, which is why nobody's favourite film is the tenth they saw. Freedom from prejudice: the author's nation, politics and personal conduct must be kept out of the verdict. And good sense, to catch the failures of design that feeling alone will not.

Delicacy gets the essay's best story, borrowed from Don Quixote. Two kinsmen are asked to judge a cask of wine. One detects a faint taste of leather, the other of iron. They are laughed at. The cask is emptied and at the bottom lies an old key with a leather thong attached. This is the strongest card Hume holds, because it shows that at least one component of taste is a discriminative skill, testable against something outside the judgement itself. Skills can be better and worse. If taste contains a skill, the verdicts of the skilled are not merely different from the verdicts of the unskilled.

The objection is circularity, and it is serious. How do we identify the qualified judge? Largely by their agreement with the works we already hold in esteem. And how do we know those works deserve esteem? Because the qualified judges say so. Hume's defenders answer that this circle is no worse than in any expertise: good doctors are identified partly by outcomes we ourselves assess and partly by the assessment of other doctors. The reply only works if something in art plays the role that recovery plays in medicine, and Hume does not claim that anything does.

He is honest about two residues he cannot dissolve. The first is temperament. The young prefer amorous poetry, the old prefer reflections on conduct, and he simply accepts that these variations are blameless and beyond the standard's reach. The second is morality. A work that asks you to admire cruelty is worse for it, he thinks, and here he will not accept variation. The trouble is that having granted one class of irreducible difference, he owes an account of why this particular prejudice is legitimate while others disqualify a judge, and he does not really supply one.

The whole essay is worth holding up against any modern aggregate. A streaming service ranks films by millions of ratings and calls the result the best of the year. Hume's standard is not a headcount; it is the concurring verdict of a small qualified set, and he says plainly that the many are often wrong and are corrected later. An aggregate treats the viewer who has seen four films and the viewer who has seen four thousand as equal witnesses, which is exactly the equality the essay denies. Whether Hume is right to deny it is a live question. That the platform has assumed the opposite without ever considering it is not.