University of Southampton OCS (beta), CAA 2012

Font Size: 
Everything flows: a computational approach to fluid landscapes
Dimitrij Mlekuž

Last modified: 2011-12-15


People, animals, things, substances, and many other entities exhibit flowing patterns of behaviour, leaving traces in the archaeological record. World is full of material features that channel and are made by these flows. Paths, hollow-ways,  roads, processional routes, streets all channel material flows of one kind or another, one of these flows being the movement of archaeologists themselves.
People, things, substances …  move into new positional and relational contexts with other things, and create new material encounters, allowing new and different flows to emerge.  Thus movement is continuously generative, and if we want to understand it archaeologically, we need to develop approaches that can cope with the fluid conditions of generative practices.
Movement cannot be pinned down because it is liquid, it is constantly shaping itself, stretching and extending in surprising ways.  In GIS,  movement is  usually represented in static form, as image, and is usually limited to the mapping movement as a series of points, in terms of mechanics of locomotion.  This is problematic, as it does not acknowledge the emergent aspects of flows, as well  as not focusing on performance of movement and its embodied knowledge and cognition.
In our paper we discuss tools we have developed to cope with the fluidity of flows, exploring their emergent properties in space and time. Our tools are developed upon the idea of "potential path fields" (Mlekuz in press), enabling us to explore landscape of flows that move with different speeds, scales and viscosies.
We approach flows as morphogenetic figures forged in time, and as field of potentialities and actualizations. Thus movement is both virtual (in Deleuzean sense; in capacity to bring about actualisation), but also real, transformed as it happens.
Another issue we touch is the active role of landscape which is not just a background, or  simple "friction surface", but becomes a flow itself, changing assemblage of individualized points, inhomogeneities, trajectories, complex relations ... being assembled and re-assembled by actualisation of flows themselves.
We want to stress that our tools are not producing non-problematic  representations of flows, as an end points that record and communicate information about movement. Even more, we believe that those representations would ultimately be misleading and wrong.  Instead, these tools should be seen as explorative activities, that do much more than resemble, they take the places of the original situation and thus enable us to intertwine our practice of modelling with the flows in the past.