This seminar constructs a mathematically informed interpretation of a classically romantic literary-theoretic thesis: that a work of art can aesthetically communicate an ineffably complex holistic understanding of the real world, which we might call the work’s ‘aesthetic meaning.’ Drawing on a generalization of ‘deep learning’ (“artiﬁcial intuition”) systems and on elementary algorithmic information theory, we describe a kind or aspect of aesthetic meaning—‘ambient meaning’—that may have a special resonance with Modernist and avant-garde approaches to aesthetic meaning, as well as with the concepts of aesthetically sophisticated cultural-materialist thought of the kind that theorist like Sianne Ngai, Jacques Rancière, or Raymond Williams practice.
‘Ambient Meaning,’ as we’ll call it, is the sensible (‘sensate’) representation of a virtual, diffused, immanent structure — the representation, by a work of art, of some systemic real-world structure akin to a mood (Heidegger), cultural logic (Jameson), sensorium (Rancière) or even ideology (Althusser/Ngai). In part I of this seminar, we will familiarize ourselves with the fundamentals of deep learning theory and information theory, and offer the hypothesis that the ‘ambient meaning’ of a work of art is, mathematically, the lower-dimensional manifold structure of the work’s imaginative landscape in intersubjective input-space.
(The seminar requires no prior mathematics or computer science background, but patience for deductive reasoning is recommended.)’
Image: Endogen Depression 2013 (The Box, La) Installation View Photo: Fredrik Nilsen