Network

1.1 Network Theories and Actor-Network Theory

Many theories and models have been developed over the concept of 'network', which is used with its own terminology and from different perspectives in varying disciplines from sociology to psychology, arts, humanities, literature, science and mathematics. It is inevitable that this concept, which has a significant place in social and cultural studies, is used in the field of architecture. In recent studies on architecture and urban space, the concept of network is mostly examined through Actor - Network Theory (ANT). Developed as a result of different studies of Bruno Latour, John Law and Michel Callon, ANT created its own terminology in the 1980s. Within ANT, there are a number of borrowed concepts along with inspirations from the works of the sociologist / philosopher / thinker Foucault, Deleuze and Guattari (rhizome). As a material-semiotic tool, ANT is a method (rather than a theory) of analysis that everything in the social and natural worlds is formed by the continuous creation of the network of relationships they are in (Law, 2009). The ‘actor’ expression in ANT's name is important. The conventional use of actor in the Anglo-Saxon tradition has become synonymous with human. However, in ANT, 'actor', or 'actant' as the authors use, is a semiotic definition. As Latour (1996) defines "An actant can literally be anything provided it is granted to be the source of an action". Symmetry, which Law (2009) refers to as "principle of symmetry" and Callon (cited in Law, 2009) as "generalized symmetry" is a key important concept for ANT. According to ANT, all kinds of actors in the world are equal and symmetrical in this context. ANT is just one of the methods by which networks are created to investigate the relations and interactions in architecture and city, which contain different relationalities due to its multi-actor structures.

1.2 Nodes, Edges and Diagrams

Networks consist of nodes (or vertices) and edges (or links). Nodes -referring ANT- can be anything. The important issue here is that the created nodes must have the power to affect others in different scopes (Giddens, 1986). Thus, the nodes can refer to actors (in Giddens' words agents) as well as actants as defined in ANT like individuals, institutions, places, objects, materials and so on. Each point in the created network has an equivalent value. One point is not superior to another (ANT and the principle of symmetry). Edges, on the other hand produce and reproduce the relationships between things in networks. It is very important to define the relations in networks. Different forms of relationships can be constructed within a network. With the relationships established through the edges, certain points in the network form some assemblages (Deleuzian term) or bundles (Giddens’ term).

In the network research and analysis on architecture and urban spaces, the architectural product itself and the locales (in Giddens’ terms) where interactions take place are important nodes besides the actor-individuals. As stated, some nodes turn into assemblages / bundles with the increase in the relations established through the edges. Using the geographer Hägerstrand’s (and later Giddens, 1986) terminology, the assemblages / bundles formed in this way can be defined as ‘stations’.

As a result of the connection of nodes and edges, a tangled mesh is formed. Today, there are applications and interfaces with various algorithms that create these tangled meshes. Using these media, it is possible to create complex networks with thousands of points and edges. It should be reminded that these productions, which are referred to as network maps in terminology, should be viewed as diagrams. Sometimes a network diagram become so complex that it becomes difficult to read and analyze the relationships between the points. In such cases, the diagrams may need to be reconstructed, schematized or even mapped based on the things to be analyzed.

1.3Networks (and Diagrams Formed by Networks)

Networks and diagrams can be constructed in many different ways. Networking is a tool for building relationships and making different analysis. Networks do not produce a theory. They are descriptive and tell stories about ‘how’ relationships were established or not.

Network is a representation. It expresses the researcher’s way of seeing. Therefore, a representation cannot be expected to analyze everything.

Networks are ambiguous. Several networks can be built, interpreted and experienced on the same subject with different definitions, ways of looking and methods. In this context, the network built on a particular subject is not absolute, but each network is unique.

Networks offer instant reality. Each network produces different analyses in its own context and in its own reality.

Networks are endless and not final products. Each on these networks presents the possibilities of different operations, and different studies reproduce the network each time. Certain nodes of the network can be considered as focal and investigated further. In the reality of new foci, other networks can be created so that the network expands and multiplies. Networks are dynamic.

[i] Some of these are: the 'chance sociogram' model of Jennings and Moreno in the field of psychology in the 1930s; social network analysis developed by mathematical psychologist Rapaport; Erdős-Rényi model in the field of mathematics in the 1960s; the Barabási-Albert (BA) Model created in the 2000s by physicists Barabasi and Albert.

[ii] Deleuze and Guattari uses ‘rhizome’ to describe multiple, non-hierarchical theory and research in data presentation and interpretation. FFI, Deleuze, G., and Guattari, F. (1987).