Before I went to bed last night I put a physical post-it on my monitor listing three intellectual tasks for the next day (that is, today). I wake up early, cruise the web, and now I’ve got a different list of priorities. The new list reflects what I saw on the web, but also the effects of “sleeping on it.”
Trump as national mood regulator
The BIG THING seems to be the idea that President Donald Trump is in the business of regulating the national mood. Various actions he has been taking have certainly affected the national mood – the tariffs have affected the business community, the ICE raids have affected his base in one way, many of the rest of us in a different way, the same with his assault on higher education, this Epstein business, and so on. That’s one thing. What I’m suggesting, what I’m thinking about, is that he does those things, and others, in order to affect the national mood. That’s different.
Of course, DJT is certainly in the business of regulating his own mood, and his mood is directly tied to national events, to the national mood. The scope of his power as President is such that his mood regulation activity has an effect on the nation. Under what circumstances does mood regulation itself become the goal, though perhaps not consciously so, rather than the various ends toward which Trump’s actions are directed?
I think that’s a real question. By was of comparison, see my post from 2018, Trumposaurus Rex @ 3QD – Toward a cybernetic interpretation, and correlated 3QD article: Feed Me Donald! – Trump, Musk, the Internet, and Monsters from the Id.
The naturalist study of literature
To the extent that I’ve got an intellectual home base, it’s the study of literature. Back in 2006 I published a long article in PsyArt: An Online Journal for the Psychological Study of the Arts: Literary Morphology: Nine Propositions in a Naturalist Theory of Form. That’s the closest thing I’ve come to a systematic account of how I’ve come to think about literary study in the 21st century. It’s still reasonably accurate, though if I were to revise it today I’d say something about digital humanities and cultural analytics.
Back in 2010 I posted an informal piece in The Valve, which I then republished here at New Savanna: “NATURALIST” criticism, NOT “cognitive” NOT “Darwinian” – A Quasi-Manifesto. (You can download it here.) Think of it as an extended advertisement for the morphology piece with links to some practical criticism and a nod toward emerging in the digital humanities.
But why do it at all, curiosity aside? What’s it for? Ecological validity, that’s what. And what, pray tell, is ecological validity? The term originated in the behavioral sciences, mostly psychology I believe, and (quoting Wikipedia)
is often used to refer to the judgment of whether a given study's variables and conclusions (often collected in lab) are sufficiently relevant to its population (e.g. the "real world" context). Psychological studies are usually conducted in laboratories though the goal of these studies is to understand human behavior in the real-world. Ideally, an experiment would have generalizable results that predict behavior outside of the lab, thus having more ecological validity.
What’s that have to do with studying literature? Nothing and everything.
Ultimately we want to understand how the human mind works, all of its facilities and capacities, working together as we live our lives. How do we get that, all of it, into the laboratory? We can’t. But literature draws on a rich range of our mental facilities working in concert. If we could understand how the mind works while we read a book, listen to a poem, attendant a play, or for that matter, watch a movie or a TV show, that would show us the whole mind operating in an ecologically valid context. For those are all things that humans do.
That’s why we need the naturalistic study of literature, as opposed to the standard humanistic/interpretive study. Interpretation tells us little or nothing about the mind/brain. But naturalistic study, that’s what it’s about. As far as I know the only place I’ve made that particular argument is in some blog posts which I’ve put together into a working paper: The Brain, the Teleome, and the Movies. Here's the abstract:
Mark Changizi has argued that we will understand the mind/brain only when we have an accurate description and inventory of the tasks it must perform. He calls this the teleome. Until we know what a mechanism is built to do, we have no way of understanding the functioning of its parts. The same is true of the mind/brain. I extend Changizi's argument by noting that the appreciation of works of art calls on a full range of human capacities and is thus a rich source of insight into neuro-mental mechanisms. Moreover we have every reason to believe that we can develop sophisticated ways of describing works of art, verbal art and films are my particular focus and interest. Those descriptions will be invaluable for interpreting observations about the brain activity supporting those aesthetic objects.
Art in the morning
I’ve got a number of posts involving Art Club, an activity where I get together with some friends and we color line drawings (coloring books for adults). Well, I took art lessons for over half a dozen years as a child and make paintings now and then into my fifties. So I decided to return to it. I purchased some colored markers, a pad of paper, and went to it. I’ve done one post on that activity, Dot Paintings (after Jamie Bérubé), but I’ve done some completely different work as well, totally freeform. And I’m now working on some stuff that combines the mechanistic order and precision of the dob paintings with the more freeform. That’s what I’m going to do once I’ve posted this and then showered and had breakfast.
I’ve done 54 paintings since I did the first dot painting on July 16. The freeforms only took twenty to thirty minutes, but the dot paintings took up to an hour. But these recent ones, they take longer.
We shall see.
Other stuff: From my post-it
The big thing for this weekend is to work on my book: Play: How to Stay Human in the AI Revolution. I want to review the material I’ve done so far and then rough out my chapter outline.
But I also want to do some work on my series on the Greatest Literary Critics, NOT. I’ve got one more post to do, on Harold Bloom, and I can wrap it up and turn it into a working paper.
Finally, I’ve been doing some work with ChatGPT-5 on ring-from analysis of literary texts. I want to turn that into a working paper as well.
And, one more thing, the cosmos
I want to make a quick extension to my post, Notes on the Metaphysical Structure of the Cosmos: The recursive nature of the cosmos implies that the cosmos is never complete. It’s always evolving, along with its metaphysical structure. The recursiveness isn’t external to the cosmos. It’s internal. There’s not end to the structure of the cosmos.
Not to mention intelligence
The AI folks seem to think of intelligence as something that’s easily scalable, the intellectual equivalent of horsepower. And to be sure the human mind has that aspect. But there’s more to it. There’s task-specific architecture. That’s important as well, and it’s not scalable.
Why have GPTs had so much difficulty with ordinary arithmetic? They’re got plenty of “horsepower.” That’s not the problem. Architecture is. The single-pass architecture of LLMs is not suited to arithmetic. Guess what? It’s not good at sketching out a semantic network suitable for a given next, it can’t do that either. Nor, it turns out, ring-form analysis. Neither of those tasks is rocket science. But they require a specific orchestration of tasks, one that LLMs can’t support.
More later.
No comments:
Post a Comment