Submerging ourselves in the question of rural areas and the digital divide, how the broadband strategy of Burgenland gets the energy network involved in broadband deployment, as well as the intricacies of fibre optics networks, I think we should now turn to something which is very obvious:(1) namely, that infrastructures such as telecommunications networks do not simply appear out of the blue, and that they are not very responsive to what any single person wants.

This second aspect is very intriguing because it runs counter to what most of us think of when we think of demand: if I as a consumer want to buy something, or in other words, I express demand for something, I can usually buy it. For telecommunications networks, we cannot apply the same notion: no matter how much I want to buy a service, chances are that my personal preferences matter less than the circumstances I am in. These circumstances look vastly different at the NIG in Vienna than they do at the outskirts of a rural village somewhere in Southern or Central Burgenland. And yet, the literature on broadband deployment stresses that demand is a key factor for deploying broadband networks. What do they then mean by demand?

What demand means at the supermarket looks very different from what demand means in the context of a network. We could now argue that comparing these two situations is not fair, because there are more factors which need to be taken into consideration. We could also claim that these contradictions mean that the entire concept of demand is useless rubbish. I think that neither of these positions is very useful for us because at the end of the day, people tend to consider both of them as instances of demand. From the perspective of a social science, and especially STS, this is a very interesting phenomenon – it tells us that ‘demand’ is not one thing. Instead, demand is ontologically multiple.

Demand as a multiple matter

We like using the concept of ontologies in STS (2) because it helps us grasp that there can be large differences between what something is in the eyes of two people. Ontologies are effectively about understanding things as being a certain way. Like with our bodies: of course, we think, a body is a body. But when we look at it a little closer, we can actually see that what a body is has undergone significant change over the course of time. Even more to the point, when you ask an athlete, a tailor and a doctor what a body is, they will most likely have very different approaches in their answers. None of them are wrong: the body is indeed an instrument, a scaffold for textile expression, an object to be mended or healed. But the body is all these things and more at the same time.

Ontologies are closely associated with practices – which allows us to observe how these ontologies are enacted. We can observe how tailors enact the body as they understand it and the instruments they use to do this: the cloth, the scissors, the puppet, but also the seams and darts. In the same vein, we can observe the practices of athletes which enact their version of the body – and what differences we can observe there between different kinds of sports! –, and we can also observe doctors when they enact the diseased, young, gendered, raced, or dead body through different questionnaires, tests, devices, treatments, or even systems of knowledge. The crucial thing about ontologies is that depending on how we see the body, we will do different things with it.

This brings us to a very interesting question, which is what I am trying to understand with my research. The literature on network deployment agrees that demand is important for network deployment. Knowing that ‘demand’ can mean many things, and knowing that networks go where there is ‘demand’, I began to ask myself – how is demand enacted in the context of network deployment?

Multiplying demand in the deployment context

Once we stop taking for granted that ‘demand’ is something out there, and instead look at the practices through which it is enacted, we see that demand is indeed multiple. My first stop was a recent study on investment decisions of operators (Curram et al., 2018). Reading this study, I began to scrutinise the different ways in which demand is mobilised in various parts of the text. To make it a little clearer what I mean, I am going to walk you through this analysis.

First of all, demand is enacted as an object which can be observed and monitored – for instance, as a subscriber base in a particular region. Importantly, this object can be transported into different contexts where it functions differently: demand enacted as a track record in securing subscribers is a different sort of demand than demand enacted as the subscription rates to the networks of other operators.

But demand can also be enacted through the types of revenues which an operator aims to generate. A wholesale operator who offers other telecommunications providers access to their network enacts demand differently from a retail operator because they are trying to do something different: their demand ends with a retail operator using their network, while a retail operator’s demand involves end users.

Operators may also enact demand as pertaining to different types of networks and network technologies – which is to say, as demand for fibre optics networks versus demand for cable networks. In the deployment context, this could mean that operators decide to take their fibre optics networks into a region where only networks which offer significantly lower speeds exist. Or they could do the opposite and decide to deploy a fibre optics network where others have recently built one, because their monitoring of demand (as an object) shows that demand (related to technologies) is sufficient to build a viable network.

In a similar manner, demand can be enacted as the intensity of potential usage of different types of networks. We can think of this as a strain put on the technical network capacity: how much bandwidth will be available to each household, how much simultaneous usage can the network tolerate. This form of demand is something quite different from the number of subscribers, because in fixed networks, one household usually amounts to one subscriber – but it might amount to two, five or even ten simultaneous users.

My analysis here is not conclusive, neither does it seek to be. Instead, my aim is to highlight the ontological multiplicity of what ‘demand’ is. This helps us ask more specific questions about which kinds of demand are understood to be lacking in rural areas. More positively speaking, it helps us ask questions about which kinds of demand those actors who deploy networks in these rural areas see, which instruments they use to see this demand, how they frame the issue of demand, and how this affects the ways in which the network is built.

I am trying to understand these questions better with my research. I have chosen a very new angle to understanding infrastructures, but I think it is worth exploring because it allows us to stop taking demand for granted, and hopefully opens us up to the many practices which contribute to enacting demand in particular ways.

What next?

Our journey started in a very concrete place that helped us feel the differences on two different sides of the digital divide. We then took a closer look at what energy networks are doing in this research project focused on fibre optics networks, before diving into what fibre optics networks are and why they are different from other networks. In this post, we approached the matter of demand.

Taking a closer look at demand through the lens of ontological multiplicity gives us a better grasp of how operators enact demand in the context of network deployment. In this context, ‘demand’ might be enacted in relation to subscribers, or to technologies, or to an intensity of usage. As ‘demand’ plays a key role network deployment, we can now begin to take a closer look at how ways of enacting demand through particular practices leads to specific kinds of networks. This will be our next stop, where we foray further into my empirical work. Stay tuned!

1 My master’s thesis is in the field of Science-Technology-Society Studies at the University of Vienna. I also work for the Austrian telecommunications regulator (RTR). Although the topic of my thesis falls into the field of telecommunications, this study is my own academic work and is not connected to the regulatory activities of RTR or TKK; in particular, the views expressed here are my own and do not prejudge the decisions taken by these regulatory bodies.

2 Annemarie Mol could be considered the doyenne of ontological multiplicity in STS. I draw on her work, in particular her seminal book The Body Multiple: Ontology in Medical Practice (2002), for my approach to demand in the context of infrastructures.