Blair Simmons: A Search for Computer-generated Literature

wall covered in ivy with bushes in front
A photo from my trip to MIT.

Hello! Welcome to the world of computer-generated literature. I have been spending my summer searching far and wide for instances of computer-generated poetry, books, and plays (i.e. essentially any instance of language-related computer and human collaboration). I am just back from a trip to MIT where Nick Montfort, a professor of digital media, has a lab called the Trope Tank. He has written many influential media theory, software studies and critical code texts such as Racing the Beam and 10 Print. At the Trope Tank, Professor Montfort has an archive of computer-generated literature. There I spent 3 days reading, skimming and jotting down notes about every single one of them.

My favorite of them all, and probably the most well known is a computer program named Racter, which was created by Bill Chamberlain. In Racter’s book of prose, The Policeman’s Beard is Half Constructed, the preface states, “Racter (short for raconteur) is the most highly developed artificial writer in the field of prose synthesis today [and is] fundamentally different from artificial intelligence programming, which tried to replicate human thinking.” Once Racter is running, Racter needs no outside input. Here is an excerpt from one of Racter’s meanderings:

At all events, my own essays and dissertations about love and its endless pain and perpetual pleasure will be known and understood by all of you who read this and talk or sing or chant about it to your worried friends or nervous enemies. Love is the question and the subject of this essay. We will commence with a question: does steak love lettuce? This question is implacably hard and inevitably difficult to answer.

Here is a passage from The Policeman’s Beard

This meandering goes on for a bit more. The entire book is filled with humorous and seemingly wise and worldly insights. In my own programming, I have found Racter’s language structure inspiring.

However, one thing to note is this program’s lack of transparency. It has been incredibly controversial in academic circles. While there was an interactive version of Racter released to the public, the complete source code was never published. Many people have, or have attempted to, debunk Racter’s validity as a thinking computer. While of course Racter is not actually thinking, I think the programs structural choices are still worth investigating.

I am looking forward to continuing to update you all with more of my findings!