Brokering digital futures

Tuukka Lehtiniemi reflects on the role of the broker – mediator of knowledge, connections, and ideas – that researchers find themselves stepping into when working on tech-related projects. 

Researchers of the digital society who work on tech-related projects are easily turned into brokers mediating information and connections. The tech field seems to create conditions for brokers, and sometimes we might unwittingly fit the bill. As brokering almost certainly means having effects, researchers should at the very least be conscious of doing it. But brokering might also constitute a purposeful mode of involvement, a means to affect how the future of the digital society unfolds.

When someone acts as a broker, the suggestion is that people, things or interests need to be connected, and for whatever reason, this cannot happen without help. The broker acts in between, mediates, and helps make the connections. Various actors can be considered as brokers in this sense. Perhaps most typically, brokers would mediate financial transactions, helping connect buyers and sellers.

The notions of brokers and brokering have been employed more theoretically in, for example, the anthropological study of development to understand actors operating at the interfaces of social worlds and world-views, negotiating and mediating relationships, channeling resources and so on (Mosse & Lewis, 2006; Koster & van Leynseele, 2018). Also here, brokers create connections that were not there before, bridging gaps and filling absences, or replacing existing connections with new ones. For example activists mediating between local legal systems and broader institutions, or middle managers acting between different levels of hierarchies in organizations are brokers in this sense (see, Koster & van Leynseele, 2018). 

Scholars note that brokers do some sort of assembling too, helping make things like systems of exchange hang together. Brokering unavoidably has effects: as brokers mediate information of connections, they affect both what they mediate and also the connections that are forged as a result. In other words, by helping things hang together, brokers participate in shaping them.

Many mediators and potential brokers are present in tech: entrepreneurial figures, for example, are brokers of information, innovation, or business practices. Even technologies like digital platforms can mediate connections and transactions, perhaps they could be brokers too. What is suggested if we focus on researchers as brokers?

Researchers as brokers

I began thinking of brokering after doing research on MyData, a data activism initiative which sought to empower people as active participants of digital society by putting them in control of personal data. With Minna Ruckenstein and Jesse Haapoja, we participated in MyData in an inside-outsider role. We identified a gap between two strands of thinking that drove data activists (Lehtiniemi & Haapoja, 2020). The majority of them were focused on making individuals the control points of data flows. They essentially trusted that favorable market development would emerge if only data were to flow under individual discretion. The minority were oriented at more collective thinking. They focused on building data collectives and thinking in terms of citizenship rather than individual choices or market-making.

We considered this latter strand promising, and hoped to make it more visible. With some of the activists, we began to use the concept OurData, trying to bridge the gap with a twist of both terminology and thinking (Lehtiniemi & Ruckenstein, 2019). We were acting at the interface of the world of critical thinking informed by social scientific research and the data studies literature, and the more business-inclined and engineering-oriented world of technology-building. This, it seems to me, might be usefully thought of as brokering, where the broker introduces ideas and knowledge from one world to another, with potential consequences for how data activism develops.

Brokering between worlds – although in a different sense – is also present in our research on data work in Finnish prisons (see, Lehtiniemi & Ruckenstein, 2022 and this blog post). Fieldwork in the prisons, consisting of observing inmates as they do data work, and interviewing them and the prison staff, appears to set up a unique position for the researcher. Prisons are disconnected places, as the world of the prison is hidden from sight by literal and metaphorical walls. Prisons are also organizationally separated from one another, enclaves that have limited connection with one another.

Us researchers are among the few people who have observed prison data work at all, let alone in several prisons. Our observations and interviews in prisons and interactions with different parties, including the prison authorities and the startup company that outsources data work to prisons, have provided us with unique visibility to prison data work. But it is clear that information does not flow easily from one of the prison’s enclaves to another, or from the prison to the outside world. When engaging in those very interactions with different actors, we begin to broker information and our own interpretations about data work, and this would be hard to avoid even if we wanted to. As we bridge the obvious information gaps, our brokering might easily end up affecting the whole data work project.

Unwitting brokers

Scholarship on brokering shows that brokers generally thrive in tenuous or unstable situations and dynamic settings. Radically changing markets would be the prime example. There is an absence to fill in, a need for someone – the broker – to help with making connections (Koster & van Leynseele, 2018). Similar suitable conditions for brokers exist in the tech domain. While regulators are currently aiming to stabilize for example AI and digital markets, things are in constant flux. In the changing tech sector, there likely are absences that need filling and gaps that need bridging. In other words, there are unattended positions and potential interfaces that can be attended to by a broker.

Tech actors undoubtedly benefit from the constantly changing environment (see Sanders & Schneier, 2024). Despite this, there is also a desire for stability, if only to have some certainty about the effects of actions. Brokers that mediate knowledge and expertise – or even a promise of these – can therefore claim a central role. A visit to LinkedIn or a trade conference reveals various candidates promising to help with mediating, bridging and assembling: consultants, keynote speakers, and self-proclaimed AI experts are happy to act as brokers.

Perhaps this sets up a scene where also the researcher often becomes a broker. Perhaps the availability of gaps to bridge, absences to fill and connections to mediate can mean that sometimes we are sort of pulled in, and we start bridging, filling and mediating even without meaning to do so. We might become unwitting brokers: the available broker position can be thought of as a vacuum that really wants to be filled, and the work researchers do is suitable to be interpreted as brokering. It might be difficult to avoid being pulled into the role. 

Brokering intentionally

Why is our work prone to be interpreted as brokering? An important part of the research we do in the tech domain is to identify and make visible acts, actors, practices, events, and connections (or lack thereof). By making something visible, we begin to affect that thing in different ways, for example simply by making it actionable for others. Making visible inevitably involves conceptual work, and choices of inclusion and exclusion. While our analysis might not be aimed at imposing a certain order on things, we make sense of things to understand them, which in itself is by necessity a way of ordering too. Research also creates connections: an integral part of this work is to bring diverging perspectives together, because that is where the interesting stuff tends to happen.

Bruno Latour famously distinguished between intermediaries and mediators. An intermediary leaves things untouched, it “transports meaning or force without transformation”. Mediators, on the other hand, “transform, translate, distort, and modify the meaning of the elements they are supposed to carry” (2005, p. 39). Researcher-brokers certainly transform rather than transport, and the broker is almost guaranteed to influence things. If brokering inevitably shapes things, even unwitting brokers do this too.

Understood in this sense, brokering doesn’t fit everyone’s research ideals. For some, it is desirable to avoid influencing things as far as possible. But should a researcher of the digital society in fact be a willing and witting broker, precisely because of the influence the position affords? Tech developers proudly tell us that their gadgets and services actively shape the digital society. Shouldn’t a researcher aim to do that too?

At the very least, we should be conscious about brokering and its effects, and the possibility of being pulled into the role. We should actively consider the effects of mediation, and broker by choice rather than by accident. Even better, we can make brokering a question of involvement. Clearly brokering is a means to create societal impacts – after all, the last couple of decades make clear that those that can affect how the digital society develops can have a great deal of impact indeed.

Readings

Koster, M., & van Leynseele, Y. (2018). Brokers as Assemblers: Studying Development Through the Lens of Brokerage. Ethnos, 83(5), 803-813. 

Latour, B. (2005). Reassembling the social. An introduction to actor network theory. Oxford: Oxford University Press. 

Lehtiniemi, T., & Haapoja, J. (2020). Data agency at stake: MyData activism and alternative frames of equal participation. New Media & Society, 22, 87-104.

Lehtiniemi, T., & Ruckenstein, M. (2019). The social imaginaries of data activism. Big Data & Society, 6(1), 1-12.

Lehtiniemi, T., & Ruckenstein, M. (2022). Prisoners training AI. Ghosts, humans and values in data labour. In S. Pink, M. Berg, D. Lupton, & M. Ruckenstein (Eds.), Everyday Automation. Experiencing and Anticipating Emerging Technologies (pp. 184-196). London: Routledge.

Mosse, D., & Lewis, D. (2006). Theoretical approaches to brokerage and translation in development. In Development brokers and translators: The ethnography of aid and agencies (pp. 1-26). Kumarian Press Bloomfield, CT.

Sanders, N., & Schneier, B. (2024). How the “Frontier” Became the Slogan of Uncontrolled AI. Jacobin, February 27th.