Over the last year I’ve been doing some work to piece together my thoughts on research in landscape architecture. The impetus for this is primarily my new role as a professor at a research institution where I get to work alongside some excellent graduate students [landscape logics] and colleagues, both of which demand a robust and compelling approach to landscape research. This reflects a general trend both within the discipline and society at large over the last twenty years. And this trend seems to be reaching a new level now and dredges up some of the questions and ideas that have come up through my own practice and study over the last fifteen years.
Whenever considering this question, I am always interested in what I would describe as authentic landscape research. That is the methods, theories, procedures, and propositions that are specific (although not necessarily exclusive) to landscape. Rather than merely adopting conceptual tools and techniques from other disciplines and projecting them onto that unwieldy and nebulous class of objects known as landscapes, I think this could be an exciting time to try and flesh out our own body of knowledge as our discipline expands. Our tendency has been toward the former. For instance, history in landscape architecture has been dominated by the art-historical approach, with dashes of environmental history concepts added here and there, but any authentically landscape history should draw heavily from history of technology as well (more on this in the next post). This syncretism is fundamental to landscape itself, and the tendency to privilege the Eurocentric art-historical approach has lead to much futzing and academic hair-splitting over meaning, symbols, and style (or language).
There are several facets to this effort that should be polished and examined (and I’m working on a piece with a colleague that does just that which I hope to publish in an accessible outlet in a few months). However, the one I wanted to consider today has been on my mind a lot over the last couple of years. The concept of abduction comes from that great philosopher of science and American pragmatist who died penniless in his rural Pennsylvania house, Charles Peirce. Controversially, Peirce saw abduction as one of the three great forms of logic, alongside the commonly accepted deduction and induction. While inductive and deductive logic have taken their place as pillars in the philosophy of science, the notion of abduction remains an outcast. However, I think the ideas Peirce explored through his theory of abduction has a lot to teach landscape architecture about the methods and logic of speculative propositions. And this could be extremely important as the discipline continues to expand beyond a practical endeavor that merely draws from or synthesizes other forms of knowledge to develop its own approaches and generate landscape-specific knowledge through research (not just practice).
Inductive logic focuses on the process of inferring a general law or principle from the observation of particular instances whereas deduction is inference by reasoning from a general theory or model to particular instances. Peirce, however, was very interested in exactly how these inferences in the inductive process are created, selected, or otherwise arrived at. How does one select or generate a hypothesis that explains a set of facts created through observation (this hypothesis is then subjected to experimentation in order to test its veracity)? Usually, such as in the work of Popper, Reichenbach, and Braithwaite, this process of hypothesis generation is understood historically (the scientist has read something or learned of the hypothesis elsewhere), contextually (the circumstances guide the scientist- perhaps an idea their lab mate has had, or the general theories of the day circulating in popular publications) or psychologically explained as a ‘flash of insight’.
What seems at first a minor issue of logic is actually incredibly important if we try to understand these concepts related to the design process. Peirce believed that “abduction is the process of forming explanatory hypotheses. It is the only logical operation which introduces any new idea” and that it entailed “all the operations by which theories and conceptions are engendered”. If we understand an early design proposal as an experimental hypothesis (they are both speculative propositions, after all, albeit in different form) that is then put to the test using design methods (drawing, modeling, mockups, field work, historical research, to name a few broad ones) then the question of how these are generated becomes incredibly important for our own work, especially if it endeavors to function as design research.It is around here where abduction can begin to fade to black a little bit. It feels like an incredibly important and overlooked insight, yet it is unclear just what it is, or how to employ it. The best I’ve come to thus far is this:
Generating (or selecting, as one could and often does start from direct precedent) the speculative proposition is at once a technical and conceptual operation. The idea and mode of thinking or representing that idea (whether with language, or drawing, collage, or something else) are intimately related. For instance, a whisper of an concept might occur to you which you then choose to sketch out in section. This might become a fundamentally different hypothesis than if you had that same whisper and attempted to capture some of it by going on a walk and photographing objects and then pinning them up in a grid. This is a question of methods. Equally, to conjure forth the idea, you might push your mind to consider projects you’ve seen before, you might attempt to clear your mind and begin free associating, or you might go outside for a few long pulls on your electronic cigarette, or a walk with your dog. This is also a methodological operation.
Abduction, then, is the method of assembling methods. It is a technical-conceptual undertaking that squares with Whitehead’s famous dictum that “the greatest invention of the 19th century was the invention of a method of invention”. Every influential development in landscape design has had a methodology strongly associated with it; a specific way information is gathered or created, and ideas are generated. From Ian McHarg’s layering to Randy Hester’s community engagement projects to P-Rex’s modeling and mapping, methods are selected or created, eventually forming a more-or-less coherent methodology that can be applied when working in situations with similar issues and landscape types.
As landscape architecture continues to expand beyond practice into research the ability to create new methods and intentionally design appropriate generative and analytical methodologies that are landscape-specific is becoming extremely important. For instance, the late 00’s headlong rush to use parametric software as a form generator is a wildly inappropriate application for landscape most of the time. The emphasis on surface (rhino is a nurbs surface modeler, after all) and divorce of the resultant geometries from the material reality of actual landscape show this to the considerate landscape maker. But as an analytical tool? Well, now, there is something there. I could imagine twenty PhD’s that could come out of that question.
[thoughts, comments, or criticisms welcome!]