- Unsupervised Learning
- Posts
- The Art Quality Tier List (AQTL)
The Art Quality Tier List (AQTL)
One model for thinking about, rating, and experiencing art
I think I finally figured out what art is. Took me decades.
Here’s my definition, which has two primary components:
Indirect expression of something that matters to humans.
So that’s two pieces:
Indirect expression
Of something that matters.
The indirect part is key. You can’t just clinically describe the scene of an accident like a forensic pathologist 1 . With art, you’re showing rather than telling. You’re presenting one thing, and what comes through is something else.
Art is showing rather than telling.
And the second piece is that whatever you’re trying to convey has to matter. To humans. To us. It needs to have valence. And the more it matters—and the deeper we feel it as the consumer of the art—the better.
So that’s the definition. As for the execution, I think three components cause us to move up in the tier list we see above.
The quality of the message
The quality of the expression
The authenticity of the exercise
The message is the thing being conveyed. Pain, happiness, existential dread, etc. And the more powerful, complex, or nuanced, that is—the better the art. Then there’s the execution of the transfer mechanism—so like the quality of the song, or painting, or whatever. Those are the two big ones.
And then there’s the authenticity. We humans often care about the story behind a piece of art because it adds (or detracts) from the meaning. For example, if you love a piece but find out it was made on a dare in 30 minutes to prove that one can create a painting without feeling anything, the emotional value of the piece goes to zero.
A quick note on AI art
This is instructive for looking at the pushback against AI art.
There’s a hidden assumption in art that the thing being communicated was actually being felt by the artist.
I think this explains why many are angry about AI art. As good as AI is right now, nobody's arguing that AI creating an image or video is feeling anything.2
AI art is hollow because the artist wasn’t expressing something they were feeling.
So that's a fair point against AI art: It can’t be conveying emotion because no emotion is being experienced by the artist.
But I think there's another looser way to define art that’s also captured in the Tier list, which is on the receiving end. In other words, if the viewer of the art feels a thing, and that thing matters to them as a human, then I think we can still argue it's art.
It’s still indirect communication, and it still conveyed something that matters—it’s just not authentic. And that’s how we end up with the tier list.
An art enjoyment methodology
I love turning knowledge into methodology, so here’s my recommended approach for beginners looking to enjoy art.
Start by feeling the piece. No thinking. No analyzing. Just take it in and experience how it makes you feel. Let that go on for a while, preferably in silence.
Now that you have a sense of how it affected you, analyze the message you believe it’s trying to convey. Imagine what this person might have been feeling that drove them to create this.
Now that you have the message—or a possible message—analyze how it was transmitted to you. Think about the techniques used, and how they might have contributed to the impact.
If you’re with a friend, do this in silence for however long that takes to do 1-3, and then discuss their results of doing the same methodology.
As mentioned, I’m very much an art amateur, so I look forward to feedback from people who’ve thought about this for a long time. 3
1 It’s quite possible, obviously, to describe something artistically—which is why I used the word “clinically” here.
2 I suppose it’s possible for someone to feel something and then use AI to try to convey it, but the disconnect between the artist and the expression arguably detracts from the authenticity. I think this could change depending on implementation and over time as well.
3 Thanks to Kelly Small and Saša Zdjelar for their wisdom and contributions on this. Saša because he’s the first person I did this enjoyment methodology with, and Kelly because she’s my sister and an artist herself.