1.618 THINGS I KNOW ABOUT HER

She might have reminded you of Juliana Pastrana, the «Non-Descript» – paraded under the name «the Bearded Lady» – who by her own account lived and died happy. But in fact she doesn’t resemble Pastrana at all, because like others of her kind, she is constitutionally elusive, «socially distanced». There is also a degree of uncertainty about whether she was happy at the time she was caught on film walking through the woods of Northern California in autumn of 1967. The anomalous existence of the footage testifies to the fact that her species, sapient but neither pre- nor post-human, does not answer the call to be recognized in and by «our» human social order. This withdrawal, even indifference – in short, this «retraction» – offers no reason to romanticize or exoticize the mystery. It instead shows up a fundamental blind spot in the anthropological regime. The common rejection of the footage as a hoax presents but the trivial, symptomatic side of this motivated blindness. The elusive character of her kind – even if a complete fiction – exposes what is more intractably at stake: not the transiency of our current era or age in deep time but the very coherence of the concept of the anthropocene.

1.618 Things I Know About Her mobilizes the predictive logic of text generation machine learning to regenerate scan-line by scan-line the 1967 Patterson-Gimlin film (stabilized by and provided courtesy of Bill Munns), purportedly documenting a Sasquatch near Bluff Creek, California. Set up by programmer Marc Padró, our neural network trained for roughly ten days before the machine ran without interruption for a month to generate the 3,175 frames of the entirely new film. The phi proportion (1:1.618) – the predictable standard for aesthetic recognition – was utilized to define the generative (reference/fill) ratio within each scan line. Each unit entailed a two-pass process that fabricated each frame anew via the neural network: first, the reference segment from the original footage (1.0) was used as the basis for predicting the fill portion (1.618), and second, the filled portion was then in turn used as the reference for predicting and filling the portion that was originally the reference. The segment length within scan lines was incrementally increased in size over the progression of frames in the original footage during the process of producing the new film sequence while simultaneously preserving, all along the way, the fundamental phi proportion. The escalation in segment length accounts for specific visible variations in the generated results over time.

Peter Freund with Eloi Puig
Spain, 2020, 2:11
Programming: Marc Padró

< VIDEO PREVIEW >