The fall-back notion of aesthetic judgment is the idea of personal taste. But that is usually invoked when you are just tired of arguing. You just leave it at that. Individual differences do not explain areas of broad agreement. It can't be coincidental that large numbers of people have admired John Coltrane, each individually. Taste is social, not individual in the strict sense. This is not to deny individual differences, of course.
But how is it socially constructed? One idea is that hegemonic groups impose their taste by choosing texts or artifacts that belong to their own group, or that reinforce their own values. The dead white male theory of the canon. There's a lot of truth to that, but it doesn't explain why Lorca is canonical and Prados and Altolaguirre are not. Most writing by dead white males is not canonical, and most non-canonical writing is not non-canonical because it is by women or people of marginal groups.
It is true that canonical literature can be used to reinforce hegemonic values, but that is increasingly less the case. Are Lorca's values more hegemonic than those of Prados?
There is the idea that any difference of valuation reflects a belonging to a different community of readers. That the Barbara Hernstein Smith theory. I don't think that works very well, for reasons that John Guillory explains.
You could try to explain it through institutions. Schools, record companies, museums, governments, have a huge influence on taste. But then you start getting a circular reasoning. Why do the institutions favor what they favor? Do they favor the canonical because it is already canonical, or is it canonical because they promote it? Could they make the non-canonical canonical just by promoting it?
You could look at Bourdieu's theory, in which taste reflects the habitus of a particular social class. Now we are getting somewhere.
But the high-brow taste will only work as high-brow if it promotes objects of art and works of literature that really are high-brow. In other words, you couldn't just call a middle-brow work high-brow and get the high-brows to think it is great, by sheer fiat. If you canonized a bad poem by putting it in all the anthologies, it would still be bad. You couldn't get intelligent readers to think it is fine, just because it in the anthologies. Then people would say: this anthology contains some poems that are there for the traditional reasons, and others that are there to represent other kinds of writing that we have decided to favor.
So we come back to the idea that some poems just really are better than others. Where "better" means having the intrinsic character or satisfying the taste of those who know what's good. It has to be intrinsic, in some sense, because it doesn't work to just impose that judgment by fiat.
Take "Muerte a lo lejos," by J. Guillén:
Alguna vez me angustia una certeza,
Y ante mí se estremece mi futuro.
Acechándolo está de pronto un muro
Del arrabal final en que tropieza
La luz del campo. ¿Mas habrá tristeza
Si la desnuda el sol? No, no hay apuro
Todavía. Lo urgente es el maduro
Fruto. La mano ya lo descorteza.
...Y un día entre los días el más triste
Será. Tenderse deberá la mano
Sin afán. Y acatando el inminente
Poder diré sin lágrimas: embiste,
Justa fatalidad. El muro cano
Va a imponerme su ley, no su accidente.
Guillén is not santo de mi devoción, exactly. It is not my personal taste that is at issue. But reading this poem you can see why he is a deservedly well-known poet. And when you read his many bad poems, you know exactly how and why they are bad, compared to this one. Getting back to Kant, you know that this is a subjective judgment, but it has to be more than subjective (individual) or it doesn't really make sense. If I prefer Lorca to Prados it cannot be simple (1) My individual quirkiness (2) The imposition of a canon by a hegemonic group or a set of institutions.