Alexis Wood on Erosion as Method(ology) at NACIS 23
Alexis Wood (Department of Geography & BCNM DE) presented her paper, “Erosion As Method(ology): Theorizing and Mapping Structures of Feeling” at the at the North American Cartographic Information Society (NACIS) Annual Conference in Pittsburgh on October 13th, 2023. NACIS is a cartographic society – bringing together academics, government agencies and commercial organizations – in appreciation, development, and explorations in the fields of cartography and GIS. The research and writing of this paper was supported by the Berkeley Center for New Media Summer Research Grant.
Given in the session Rethinking Map Conventions, this paper asks: How might we map the coalescence of ruralities, climate change, and digitalities in northern California's state secessionist movements? Using the theoretical foundations of Raymond Williams' concept “structures of feeling”, this paper argues for a critical cartographic representation of physical and sociopolitical processes through the methodological frame of erosion and deposition. This experimental approach to theorizing and mapping this region allows both material and metaphorical investigation of the transfigurations of the physical, social, and digital, not only within themselves but between one another through common processes across spatiotemporal scales. To this end, this paper calls for a return to ‘space as process’ in cartography to reorient the map towards agencies and relationships which are often invisibilized.
In addition, Clancy Wilmott (Department of Geography & BCNM) and Alexis Wood displayed a draft of their map titled “Before You Are Here: An Indigenous Cartography of the Ohlone Bay Area”, created in collaboration with the Sogorea Te’ Land Trust.
Berkeley Center for New Media was also represented by Clancy Wilmott in the session “Cartographic Research”, where she presented the paper, “Cartotopia: An Atlas of Artificial Intelligence Maps”, a snapshot of preliminary findings from Cartotopia, an ongoing atlas project which uses historical and contemporary cartography and imagery to interrogate the power of both cartography and visual computation through a series of artificial intelligence maps. This paper argues that the historical conditions of cartography—from production to analytic potential—are folded into machine learning through both cartographic and computational abstraction and generalization, revealing ongoing politics that linger in the dataset regardless of permutations.