Can computers do narrative?

The question has been debated in three recent articles in Narrative:

Angus Fletcher, Why Computers Will Never Read (or Write) Literature: A Logical Proof and a Narrative, Narrative, Volume 29, Number 1, January 2021, pp. 1-28 DOI: https://doi.org/10.1353/nar.2021.0000

Jon Chun, Katherine Elkins, What the Rise of AI Means for Narrative Studies: A Response to “Why Computers Will Never Read (or Write) Literature” by Angus Fletcher, Narrative, Volume 30, Number 1, January 2022, pp. 104-113 DOI: https://doi.org/10.1353/nar.2022.0005

Angus Fletcher, Why Computer AI Will Never Do What We Imagine It Can, Narrative, Volume 30, Number 1, January 2022, pp. 114-137 DOI: https://doi.org/10.1353/nar.2022.0006

I’ve blitzed through all three articles, skipping some passages, reading others, some more, some less carefully, and highlighting some. My reaction:

NO! NO! NO!
STOP IT!
I don’t care whether or not computers can do narrative.

You’re making my eyes bleed and my ears turn to mush. I don’t see that these kinds of debates are getting any of us anywhere.

As I’ve explained many times, way back in the ancient days David Hays and I reviewed the field of computational linguistics for a journal that was then called Computers and the Humanities (it has since changed its name):

William Benzon and David Hays, “Computational Linguistics and the Humanist”, Computers and the Humanities, Vol. 10. 1976, pp. 265-274. https://www.jstor.org/stable/30199826 Ungated, https://www.academia.edu/1334653/Computational_Linguistics_and_the_Humanist

After standard review fare, we proposed, as a thought experiment, the idea of a computational system capable of reading a Shakespeare play, in some interesting, but unspecified, sense of the word ‘reading.’ We called it Prospero and set no date on when Prospero would be operational, but in my mind I figured we’d have it in 20 years or so. The article appeared in 1976. Add 20 to that and we have 1996. Was anything like Prospero available then? No. Not only that, but the symbolic computing that was at the center of our review, and of Prospero, was being pushed into the background by statistical methods.

It’s now 2016, 40 years after that paper. We don’t have anything like Prospero now – though I believe the late Patrick Henry Winston had been using the Macbeth story (but not Shakespeare's play) in an investigation of story comprehension – and I see few prospects for Prospero in the near future (and, yes, I know about GPT-3 and am even sympathetic). These developments make sense to me. I suppose that should put me in the Fletcher’s camp, but, no, not really. Nor can I sign on with Chun and Elkins.

There’s a lot of specific commentary I would make on these articles (some nit-picky, some substantive). Tempted as I am to offer a quick comment or two, I’d better not. If I started, I’d be tempted to go on, if only to clarify those quick comments, to fend off misleading suppositions, and so forth. And pretty soon I’d be launched into who knows what, much of which I’ve already said on way or another. No, for now it’s best to leave it alone.

We are far from understanding narrative either at the neuronal level (Fletcher’s hobby horse) or computational (all three). This debate seems pointless. We need to get beyond these boundary drawing debates and into some real work.

Maybe later.