Allison Parrish is a poet and a programmer who intermingles these two worlds brilliantly across her many projects. You can find her full portfolio here. I wanted to explore her work and reflect on it as some of what I’m doing currently with this project approaches some of her ideas.
While looking through her work, I found myself coming back to her Reconstructions site. The project is (to quote her) “an infinite computer-generated poem whose output conforms to the literary figure of chiasmus.” Lines are pulled from poems in the Gutenberg Poetry corpus and randomly selected to be conformed to an ABBA scheme. It continues infinitely until the page is closed.
The idea of a poem that is literally never-ending, going on forever if you let it, is so completely fascinating to me. The words are nice, sometimes a particularly poignant combination of lines is created, but obviously the significance of a work like this is its infinity. A wonderful detail of the site is that as new lines are added, older ones are raised to the edges of the screen, made translucent and faded, but they never disappear. They're memory is preserved on the page, always a part of the full poem, even as it grows larger and larger.
I'd love to be able to do something like this -- create a poem that's core concept is rooted in its construction rather than just its words. I'm not sure if that's quite possible for this project, but it's something I will always have in the back of my mind.
Another work from Parrish that fills this role is her piece Compasses, a collection of very unique poems that are all made of 4 main words, one a each "direction" like a compass, with words generated by an AI that sort of bridge the gap phonetically between the 4 directional terms ("north" -> "earthe" -> "east"). The poems utilize shape as much as they do their terms and AI component, which I found thought really added an extra dimension to the work. It’s fun, a work that doesn’t take itself too seriously, but does influence you to ponder the nature of space within a written work. I want so badly to implement something as clever into my own work, if I can.
One more interesting entry in Parrish’s repertoire is her Corpus-driven Narrative Generation notebook. It’s built in Python, meant to pull grammatical and named entity data from a large body of text using natural language processing. From this, one can play around with the data they extract and create some really funky pieces of work. She’s set it up in a way wherein you can write out whole scenes from randomly generated text. Take a look at this bit of the code:
rules = {
"subject": [w.text for w in subjects],
"object": [w.text for w in objects],
"verb": [w.text for w in past_tense_verbs if w.text not in ('was', 'were', 'went')], # exclude common irregular verbs
"adj": [w.text for w in adjs],
"people": [w.text for w in people],
"loc": [w.text for w in locations],
"time": [w.text for w in times],
"origin": "#scene#\n\n[charA:#subject#][charB:#subject#][prop:#object#]#sentences#",
"scene": "SCENE: #loc#, #time.lowercase#",
"sentences": [
"#sentence#\n#sentence#",
"#sentence#\n#sentence#\n#sentence#",
"#sentence#\n#sentence#\n#sentence#\n#sentence#"
],
"sentence": [
"#charA.capitalize# #verb# #prop#.",
"#charB.capitalize# #verb# #prop#.",
"#prop.capitalize# became #adj#.",
"#charA.capitalize# and #charB# greeted each other.",
"'Did you hear about #object.lowercase#?' said #charA#.",
"'#subject.capitalize# is #adj#,' said #charB#.",
"#charA.capitalize# and #charB# #verb# #object#.",
"#charA.capitalize# and #charB# looked at each other.",
"#sentence#\n#sentence#"
]
}
This is setting up an array filled with this random data, and constructing a little set-piece with said data. This is expounded upon further down the notebook, using other modules to weave together more complex thoughts, but the level of control here is what I enjoy the most. There’s an insane amount of potential in this, and given some previously-mentioned, potential limitations in my word choice and subject matter, this could be an amazing tool to mix things up and generate unique ideas!