PhD Dissertation · Virginia Tech · 2026

Adaptive Epistemologies and Neo-Wilds

Practice-Based PhD · Committee: Marcelo Stamm (chair), Paul Kelsch, Mintai Kim, Shaun Rosier

This dissertation develops a theory of adaptive epistemology for landscape architecture operating at territorial scales under conditions of irreducible uncertainty. Drawing on twenty years of practice engaging ecological complexity, embedded sensing, and generational robotics, it proposes that design practice is itself a mode of knowledge production, conducted through iterative, feedback-based, in-situ inquiry rather than predictive control. Three contributions organize the argument: adaptive epistemology names the framework, the cultivant names the disposition, and refraction names the method through which both became legible.

Core Contributions

Adaptive Epistemology

The framework. Design propositions as hypotheses tested in situ. An Orphic project conducted with Promethean tools — attentive to what systems reveal rather than determined to enforce what models predict.

The Cultivant

The disposition. Ongoing negotiation between designed intention and biological agency. Maintenance as epistemological act. Plurality as constitutive of the design relationship from the outset.

Refraction

The method. Systematic retelling of practice from vantage points its instrumental contexts never demanded. Developed through the Practice Research Symposia and continued across the dissertation.

Six Frameworks

Multiple Intelligences

Human, machine, and biological cognition operating simultaneously as irreducibly different modes of knowing.

Technogeographies of Sensing

Sensing as design. The instrument constitutes what can be known, and what can be known determines what can be governed.

Wetware

Biology coupled with computation as a knowledge-producing medium. Living systems generating responses that exceed what any sensor was calibrated to detect.

Generational Robotics

Learning loops across timescales exceeding human institutional memory. Knowledge distributed into infrastructure rather than held in the cognition of practitioners.

Coupled Ecologies

The territorial condition. Biology, computation, and infrastructure operating as a single entangled system producing a new category of landscape.

Reflexive Stewardship

The ethical orientation. Provisional, adaptive practice that remains answerable to what landscapes are becoming and to whom the costs and benefits accrue.

Chapter Structure

01
Territory: A Formed Condition
The territory as infrastructural hybrid, the failure of prediction, sensing politics, and the stakes
02
Adaptive Epistemologies: A Framework
The epistemological framework, six elements, Orphic/Promethean, Simondon, Kauffman, Bach
03
Refractions: Practice-Based Research
Refraction as method, PRS process, research-through-design methodology
04
Ecology of Practice
Communities of practice, collaborative networks, disciplinary hybridity
05
Tools: Practices, Manifests, and Speculations
Twenty years of practice across four phases: seeing, touching, coding, letting go
06
Models: A History and Treatise
Fluvial modeling history and the departure from prediction
07
Technogeographies of Sensing and Neo-Wilds
The politics of sensing, highly maintained wildness, landscapes in the gaps
08
Landscape as Medium: The Model and the Site
The landscape as computational medium, embodied and distributed knowledge
09
A Shifting Model of Interactions
Adaptive management enacted through wetware and coupled ecologies
10
Generational Robots: Persistent Autonomy
Slowness, generational knowledge, the wildness creator
11
Co-Creation: Multi-Species Communications and Authorship
Distributed intelligence, multi-species authorship, the Third Intelligence
12
Synoptic View: Relations and Trajectories
Synthesis, adaptive management vs. adaptive epistemology, contribution to the field
13
Vectors
The manifesto. Bold claims about where the discipline goes from here.

Virginia Polytechnic Institute and State University