New Archive InterpretationsNew Archive Interpretations

With the project Dark Archives, Erica Scourti activates an automatic, constantly updated archive to foreground the algorithmic outsourcing of personal media organisation and its implications for networked, distributed memory.

Throughout the exhibition period of Temporary Fashion Museum, she uploads her entire fifteen-year personal media archive, consisting of daily photos, videos and screenshots, to platform monopoly services like Google photo. The collection is parsed by auto-editing, classification and tagging software, allowing hundreds of videos, collages and animations to be automatically generated as a result, which are again shared online. While based on the artist's 'real' archive, and intentionally collapsing temporal distinctions within it, Scourti offers up her past and present for a new and constantly updated collection of videos as a live, performed archive. The collection is parsed by auto-editing, classification and tagging software, allowing hundreds of videos, collages and animations to be automatically generated as a result, which are again shared online. While based on the artist's 'real' archive, and intentionally collapsing temporal distinctions within it, Scourti offers up her past and present for a new and constantly updated collection of videos as a live, performed archive.

_'Videos created by searching my whole current photo archive, uploaded to Google photos, with different keywords. __These semi-automated videos make up part of a commission at Het Nieuwe Instituut in which a group of writers have been invited to imagine and caption the speculative 'missing media' in my archive, with the aim of using these captions to create a new set of videos.' (YouTube) _With texts written by Jessica Bunch, Christina Chalmers, Sandra Huber, Linette Voller and Joanna Walsh.

Foregrounding the machine-reading of digital imagery and the associated metadata detected in all digital content, Dark Archives implicitly draws not only on the artist's individual archive but the millions of other user media that the algorithms use as deep-learning training sets. Next to incorporating other people's data that derive from her own archive, Scourti also involves elements of staging, scripting and fictionalizing, by asking a group of writers to speculate on and caption what they imagine to be the missing set of media that somehow evaded classification within the archive: the false negatives, the misclassifications, the media that fell outside of Google's definition for that search term. These captions are used to create a new series of videos, to feed into the final work which visitors can access on their smartphones, at The New Institute. Questioning notions of data collecting and shared authorship, Scourti also challenges notions of individual memory.

Dark Archives explores the implications for ownership and access of mass storage and classification of personal media, the project also points to the issues that arise when image archives can be parsed, and potentially monetised, in the excavation of intimate data once in the hands of corporate platforms.