The Apocryphon of Tom Hater Oct 8, 04:24 AM
The excerpt is pulled from a text of six hundred sixty-six sections generated from the combination of Samuel Pepys’ diary entry for September 2, 1666 (the first day of the Great Fire of London ) and Chapter 13 of the book of Revelation.
The term apocrypha , which comes from Greek – ‘secret’ or ‘hidden’, is usually used to refer to texts of questionable authorship and/or texts contemporary to but excluded from the Biblical canon. Important apocryphal texts include The Book of Enoch and The Gospel of Thomas . The relationship between apocrypha and canon is usually turbulent in that the apocrypha expands original accounts, reinterprets events, and many times contradicts canonical texts.
In 1945, in Nag Hammadi, a town in central Egypt, local peasants uncovered a vast amount of Gnostic literature which is now known as the Nag Hammadi Library . Many of these texts are considered apocryphal (it contains the aforementioned Gospel of Thomas). The texts were hidden by monks 1600 years previous because they were declared heretical. The Nag Hammadi codices were written in Coptic, a late Egyptian language. The written form of Coptic uses the Greek alphabet.
I chose my source texts for their involvement with destruction and secret meanings, as well as a numerical correspondence between the year of the Great Fire of London (1666) and the number of the beast (666). The aim of the project is to see what varieties of augmentation occur when two texts that share certain linkages are combined, re-ordered, and infected by one another. I want to cause a distrubance in perception similar to what occurs when texts apocryphal to the Bible are compared to the traditional canon. However, I’m approaching the disturbance on a textual level where new narratives and perceptual states can rise from simultaneous reading of reorganized texts.
The combination process is largely automated by computer. I wrote the combination program in the Python programming language. The program uses Markov chain analysis to randomize the two texts based on probability. This makes the output text more readable and produces more evocative re-combination than pure randomization. This page explains it well: http://www.eblong.com/zarf/markov/ . I then pulled all the sections that feature “hater” as Mr. Hater or Tom Hater.
Code (Python):
TextGen Library
The script that ran the actual generation