“BillBlakeBot (@autoblake)” by Roger Whitson

 BillBlakeBot @autoblake  I Tweet in the infernal method, by corrosives, melting away apparent sentences & Displaying the Infinite which was hid.  Golgonooza & Beulah & London &
Open “BillBlakeBot (@autoblake)” by Roger Whitson

Every three hours, this bot tweets approximately 100 characters (about 20 words) of language written by William Blake, but not exactly. The tweets are recognizably Blake’s, but there’s something odd about them, as if he was performing some kind of automatic writing or Surrealist automatism to compose those texts. He wasn’t, but in a way this bot is doing it for him, to show us some of the underlying structures William Blake’s poetry and prose.

André Breton defined Surrealism as “pure psychic automatism, by which an attempt is made to express, either verbally, in writing or in any other manner, the true functioning of thought. The dictation of thought, in the absence of all control by the reason, excluding any aesthetic or moral preoccupation.” The Surrealists were interested in finding an artistic expression that revealed a higher reality than our bourgeois consciousness would allow.

“BillBlakeBot” uses a markov chain generator that analyzes the David Erdman revised edition of The Complete Poetry and Prose of William Blake (1988) to determine the statistical likelihood that a word will follow another. It then uses that lexicon and probabilities to generate phrases and sentences that mimic Blake’s style, even if the results don’t make sense. Roger Whitson discusses his bot and markov chains in great detail in this ICR 2013 presentation, concluding with some provocative questions:

I think it adds another level to what it means to “read” Blake. Can we read these tweets as having relevance to Blake’s work? Is it simply nonsense? Literally, it might be, but I feel algorithms like this one complicate some of those questions in fascinating ways.

When we read Blake’s texts, we read the results of hours of carefully crafted language, revised and edited exhaustively before etching them in copper plates to produce his famous illuminated books. These texts are the result of very conscious and deliberate creative process applied to his mystical visions and other sources of inspiration. To use a markov chain generator to cut across his life’s oeuvre reveals patterns of style that Blake was probably unaware of. Beyond its utility as form of distant reading, its poetic output is worth reading closely to analyze the textures of Blake’s language beyond semantics. @autoblake gives us access to a surreal Blake, of interest to scholars and enthusiasts alike.

Coda: Interested in creating a Twitter bot from your favorite author’s work? Roger Whitson published the source code in GitHub.

Featured in: Genre: Bot

“@tonightiate” “@MassageMcLuhan” by Matt Schneider

These two bots generate short template based sentences and publish them on Twitter every 10 minutes. With them Schneider demonstrates some of the versatility of the same kind of device when applied to different topics.

#gifandcircumstance by Allison Parrish

Screen capture from #gifandcircumstance by Adam Parrish. Team Illuminate's glow in the dark dance from America's Got Talent. Text: "If I were your man..."
Open #gifandcircumstance by Allison Parrish

This bot mines the Twitter stream for phrases starting with “when,” extracts the clauses, and joins each phrase with a randomly selected animated GIF in a Tumblr. Here’s a more detailed description from Parrish’s blog:

A “#whatshouldwecallme-style tumblr” is one in which animated GIFs are paired with a title expressing a circumstance or mood—usually a clause beginning with “when.” I wrote a Python script to make these kinds of posts automatically. Here’s what it does:

(1) Search Twitter for tweets containing the word “when.”
(2) Extract the “when” clause from such tweets.
(3) Use Pattern to identify “when” clauses with suitable syntax (i.e., clauses in which a subject directly follows “when”; plus some other heuristic fudging)
(4) Post the “when” clause as the title of a tumblr post, along with an animated GIF randomly chosen from the imgur gallery.

Read more#gifandcircumstance by Allison Parrish

“@everyword” by Allison Parrish

Screen capture of the "@everyword" twitter account created by Adam Parrish. Text "Twittering every word in the English language. Task will complete in 2013. Tweet 1: skirling. Tweet 2: skirl"
Open “@everyword” by Allison Parrish

This bot has been on deceptively simple mission since it was launched in 2007: it is tweeting the English language, one word every 30 minutes, in alphabetical order. This work of conceptual poetry is delightfully absurd because it claims to be “twittering every word” and even offers a termination date in which such a project would be complete— when even the concept of what constitutes the English language is subject to debate, even if it wasn’t changing on a daily basis. To make such a feat even possible (unless you’re Wowbagger The Infinitely Prolonged) requires setting constraints—such as a choice of dictionary— though it is to Parrish’s credit that she doesn’t disclose the source, because it enhances the project’s conceptual claim.

Read more“@everyword” by Allison Parrish

“Tweet Haikus” by Brandon Wood