Supercut texts are made by first collecting whatever is at focus (examples from Tom Comitta include first sentences of stories, last sentences of sci-fi novels, or nature descriptions in novels). This becomes a dataset and then each datapoint is arranged with others by pattern to create new text.

Scissors and tape are a significant part of it. Maybe we’d call them the middle part. Basically, for any of these supercut texts (the subset of citational fiction that I work in), I first spend a lot of time collecting whatever I’m focusing on, be it first sentences of New Yorker stories or last sentences of sci-fi novels or, in the case of The Nature Book, nature descriptions in novels. In a way, I’m collecting a dataset. Sometimes that set might be more comprehensive. For the piece made entirely out of sentences from New Yorker stories, I gathered and printed out the first line of every short story published by them during the past twenty years. (View Highlight)