Next week in my Philosophy of Literature module I’ll be talking about pareidolia and theories of how and and why it works. How and why pretty much any closed loop with three dots in it is a face, because it ‘looks like’ one. The occasion for burdening my students with this is discussion of overly-linguistifying (in my view) theories of how literature ‘works’ and, more grandly, linguistifying theories of what Aristotle called mimesis, a.k.a. that whole ‘poetics’ ball of wax. I posted some of my thoughts about pictures and pictoriality before: it’s important to realize that even though a smiley face is an utterly conventional icon, it doesn’t follow that it works by convention.

Anyway, I thought it was a nice coincidence that Andrew Sullivan linked to this today, for his Faces of the Day thing.

Also, I just stumbled on a real sparklepop/powerfolk earworm of a tune by Vetiver, “Wonder Why”, which turns out to have a a pareidolia-based video. Great track. Get it free from Amazon.

The maps and the video are good examples for me because they preemptively emphasize something that is often raised as an objection to efforts to ‘naturalize’ the pictorial function: namely, it’s a learned process. By the end of the map series, and the video, you are more sensitized to faces and figures in maps, mailboxes and trashcans than you were at the start. To that extent your responses are ‘conventional’, in the sense of learned (when you could perfectly well have been learning something else, so the result is somewhat ‘arbitrary’). Fine, fine. But the point still stands. From the fact that a result is path-dependent, it may follow that it is conventional (in a perfectly good sense of that word). But, again, it does not follow from the fact that something is conventional in that sense that it ‘works by’ convention in some other senses that tend to be carelessly bundled in. The mechanism by which we recognize things as faces is cognitively distinct from the mechanism by which we recognize that ‘faces’ denotes faces. My target here is Nelson Goodmanian thinking, which tries to explain pictorial resemblance and representation on the model of linguistic denotation. He doesn’t say it works exactly the same, all the way up and down; that would be pretty obviously crazy. But he pushes the line that, in order to theorize how pictures work, you have to build on a kind of denotational foundation. I think the opposite: theories of linguistic denotation need to rest on a foundational theory of pictoriality. But enough about me. Enjoy the video and the song. Great song, I think.