ANT can be said to be proposing network geographies. There are two aspects of this proposal that need mentioning: first are the basic characteristics of the geographical imagination advocated by ANT, and second is the methodological implication of this perspective for geographical research.
Space and time are central to ANT. Actor-networks necessarily depend on nonhuman actors for extension across space and durability in time. To trace the practices of actor-networks conversely involves shifting in time space; space and time are entwined and emerge out of the enactment of heterogeneous relations of actor-networks. In other words, ANT follows the networks and associations which configure and shape the sociospatial landscape of the world. Attention to time and space has often been an implicit feature of ANT studies although it has been central to the philosophies that ANT is based on, such as in the work of Serres. Questions of space and time are currently being increasingly addressed in ANT studies that express both an interest by practitioners of ANT in geography and by geographers in potentials of ANT.
Central to ANT is the process of translation, that is, transformative practices that describe the making of connections, assemblages, or associations. This is the productive force of actor-networks. When translation is studied, it is evident that it enfolds combinations and recombinations of diverse objects or actors. This is a process that plays with – and goes beyond – the relation between proximity and distance. Actor-networks thereby render distant things seem proximate and conversely the inclusion or exclusion from actor-networks creates new distances between actors.
These moves disturb the conventional geometric conceptualization of space. The world is no longer a gridlike surface on which it is possible to draw stable and enduring lines of proximities and distance. This kind of Euclidian geometry and the cartographic geography that builds on it is rather seen by ANT as one instance of 'spacing' and 'timing'. Euclidian space is a particular form of time space that does not have monopoly on spatial imagination. There are possibilities, other ways of spacing and timing through which the sociospatial landscape emerges.
ANT suggests a turn to topological ways of thinking. Topologies can describe the relational ordering of spaces for which geometry cannot give a good indication of proximity and distance. Topology is not dependant on linear time. It can deal with time spaces that come across as fluid, mutable, or even flickering as fire. These are spatialities that emerge through enactment of complex and heterogeneous relations. Topology grasps sociospatial realities that are beyond measurement but which are all based on relations.
A topological world calls for topological geography – a geography that can deal with complexity of relations and networks. Topological world is a world of multiple spaces. The Euclidian space is not written off but only thought of as one possible way to order spatial configuration of relations enacted through actor-networks. Increasingly, the proponents of ANT and geographers alike are pursuing such a multiple approach that recognizes the interference between forms of time spaces. That demands an apprehension of how any object or any actor is part of and co producer of many time spaces. In other words, an actor is basically an intersection of multiple relations or to cite one of the old mantras of ANT: Every actor is a network.
ANT, as a study policy, is characterized by emphasis on description, where stress is put on following the relational practices of actor-networks. In line with ethnomethodology, the researcher should trace the footsteps of the actors under study and describe their ways of raising the world. It follows that one should trust the actors so as not to render them subject to external explanations. The explanations coming from the study should be internal, emerging out of the field of practice under study.
In this regard, it is important to highlight the relational worldview of ANT. The field is seen as always a relational achievement and many of the practices going on in every field revolve around its maintenance, or the continual enactment and re enactment of the field. This is a key point if the study is not to turn out to be a functionalistic exercise in filling up a blank field with descriptions of networks. It follows that the relations in and out of the field in question are as important as what goes on within it, as these constitute the field to a large extent and hence, the boundary of a field is primarily a practical achievement.
This relational understanding of the field has come as an inspiration for geographers grappling with the geographical dimensions of the field and fieldwork. The common sense notion of the field in human geography has been that of a bounded space, or at least a space that can be metrically defined and located in Euclidian space. Thinking of the field as an actor-network that draws on and enacts diverse time spaces opens up new pathways for fieldwork. It becomes possible to move the focus from a description of the content of a field – often predefined by particular actors – toward the work and the contingencies involved in constructing the field. ANT is thus able to sensitize fieldworkers to their own role in constructing the field they are describing. This underlines that an everpresent feature of fieldwork is that it partly creates the field it describes as it carves out situated knowledges of it.
In sum, network geographies acknowledge the ontology of relational practices as a starting point of research. They also acknowledge that the implication of that is a slight confusion of ontology and epistemology, that is, what the world is like and what can be known about that world. Practitioners of network geography partly assemble the time spaces they describe, through their research practice. The aim of such practice is not to find a fit between an account of a field and its reality but rather to find ways to move on, keep on course, or change it and thereby not least to reflect on alternative routes of travel.
ANT has attained a place in the toolbox of contemporary human geography. It has spread deeply into different subdisciplines often as a part of more general relational turn in the field. Taking the aforementioned critiques into consideration it seems however, imperative not to link ANT too tightly to other network approaches, simply because it is about much more than networks in the conventional sense of the word. ANT is thereby an important part of the inspiration to a broader new tradition of network geographies as it seeks to keep the concept of network, fluid and mutable.