Towards a General Methodology for Second-Order Science 1

In recent years the field of cybernetics has been described as consisting of two bodies of work created in two time periods: first order cybernetics from the late 1940s until about 1975, and second order cybernetics from the mid 1970s to the present. Each period lasted about 25 years. What comes next? I shall describe here what I think comes next and how the new point of view emerged, at least in my own thinking.

“Second-order Logic” in Anderson, C.A. and Zeleny, M., Eds. Logic, Meaning, and Computation: Essays in Memory of Alonzo Church. Dordrecht: Kluwer, 2001. Pp. 61–76. The original session on this article expired before most views and downloads took place. It has had over 10,000 views and over 500 downloads. The new readers deserve a new session. Abstract. This expository article focuses on the fundamental differences between second- order logic and first-order logic. It is written entirely in ordinary English without logical symbols. It employs second-order propositions and second-order reasoning in a natural way to illustrate the fact that second-order logic is actually a familiar part of our traditional intuitive logical framework and that it is not an artificial formalism created by specialists for technical purposes. To illustrate some of the main relationships between second-order logic and first-order logic, this paper introduces basic logic, a kind of zero-order logic, which is more rudimentary than first-order and which is transcended by first-order in the same way that first-order is transcended by second-order. The heuristic effectiveness and the historical importance of second-order logic are reviewed in the context of the contemporary debate over the legitimacy of second-order logic. Rejection of second-order logic is viewed as radical: an incipient paradigm shift involving radical repudiation of a part of our scientific tradition, a tradition that is defended by classical logicians. But it is also viewed as reactionary: as being analogous to the reactionary repudiation of symbolic logic by supporters of “Aristotelian” traditional logic. But even if “genuine” logic comes to be regarded as excluding second-order reasoning, which seems less likely today than fifty years ago, its effectiveness as a heuristic instrument will remain and its importance for understanding the history of logic and mathematics will not be diminished. Second-order logic may someday be gone, but it will never be forgotten. Technical formalisms have been avoided entirely in an effort to reach a wide audience, but every effort has been made to limit the inevitable sacrifice of rigor. People who do not know second-order logic cannot understand the modern debate over its legitimacy and they are cut-off from the heuristic advantages of second-order logic. And, what may be worse, they are cut-off from an understanding of the history of logic and thus are constrained to have distorted views of the nature of the subject. As Aristotle first said, we do not understand a discipline until we have seen its development. It is a truism that a person's conceptions of what a discipline is and of what it can become are predicated on their conception of what it has been.

Proceedings of the 59th Annual Meeting of the ISSS - 2015 Berlin, Germany

Since around 1980, Ranulph Glanville has put forward the idea that rather than seeing research in design as one form of science, we instead see scientific research as a specific form of design. This argument, based on the way that scientific research inevitably involves design activity but not vice versa, and others like it around that time consolidate a shift during the 1970s in thinking about design, from a concern with the scientific method to the idea that design has its own epistemological foundations as a discipline. The attempt to base design on a linear version of the scientific method failed for reasons that have been pointed out by Horst Rittel amongst others: because design involves the creation of the new, design questions cannot be exhaustively formulated in advance. This has marked something of a parting of the ways between design and science as being incompatible in terms of method. Given Glanville’s argument this is not what we might expect: if science is a limited form of design, shouldn’t scientific approaches be commensurable with design even if they are not a basis for it? This apparent disjunction is only the case if we follow the changes in how design was thought about during this period without also following the comparable changes regarding science. Both broadly parallel each other, moving from a concern with method in the 1960s through a critique of this in the 1970s to new foundations from the 1980s onwards, focusing on what designers and scientists actually do in practice. Indeed the key critiques of method advanced by Feyerabend and Rittel, in science and design respectively, have similar structures and, so, what seems at first sight to be a rupture can also be read as a parallel journey. Using this account as a basis, and in the light of recent discussions regarding the idea of second order science, I suggest that we can understand contemporary design research as one example of second order research practice, as is indicated by its continuity with cybernetics. More speculatively, and with reference to the Fun Palace project of Joan Littlewood and Cedric Price, to which Gordon Pask also contributed, I suggest that architecture can itself sometimes be thought of as facilitating such a reflective and participatory enquiry.

Constructivist Foundations, 2014

> Context • The problems that are most in need of interdisciplinary collaboration are " wicked problems, " such as food crises, climate change mitigation, and sustainable development, with many relevant aspects, disagreement on what the problem is, and contradicting solutions. Such complex problems both require and challenge interdisciplinarity. > Problem • The conventional methods of interdisciplinary research fall short in the case of wicked problems because they remain first-order science. Our aim is to present workable methods and research designs for doing second-order science in domains where there are many different scientific knowledges on any complex problem. > Method • We synthesize and elaborate a framework for second-order science in interdisciplinary research based on a number of earlier publications, experiences from large interdisciplinary research projects, and a perspectivist theory of science. > Results • The second-order polyocular framework for interdisciplinary research is characterized by five principles. Second-order science of interdisciplinary research must: 1. draw on the observations of first-order perspectives, 2. address a shared dynamical object, 3. establish a shared problem, 4. rely on first-order perspectives to see themselves as perspectives, and 5. be based on other rules than first-order research. > Implications • The perspectivist insights of second-order science provide a new way of understanding interdisciplinary research that leads to new polyocular methods and research designs. It also points to more reflexive ways of dealing with scientific expertise in democratic processes. The main challenge is that this is a paradigmatic shift, which demands that the involved disciplines, at least to some degree, subscribe to a perspectivist view. > Constructivist content • Our perspectivist approach to science is based on the second-order cybernetics and systems theories of von Foerster, Maruyama, Maturana & Varela, and Luh-mann, coupled with embodied theories of cognition and semiotics as a general theory of meaning from von Uexküll and Peirce. >