Your more recent work on Twitter has deployed the concept of a gift economy, building off some of the ideas in our original white paper on Spreadible Media. How are you defining gift economy? Why is this appropriate for talking about digital media? How do contemporary forms of gift economies in the context of capitalism differ from more classical understandings of this context?
It is less the ‘economy’ in gift economy than the ‘gift’ that interests us. The gift as a ‘public recognition’. And this initial public recognition with the intention of more exchanges in the future, is a key aspect in gift economies as Boris Malinowski has pointed out. Together with my colleagues Johannes Paßmann and Thomas Boeschoten we looked into Twitter data to retrieve patterns of communication (The Gift of the Gab. Retweet Cartels and Gift Economies on Twitter). When investigating two samples, the MP’s of Dutch parliament and the German top Twitter accounts, we noticed clusters of users who were retweeting each other frequently, so called retweet-cartells, similar to citation-cartels in academia. We argue that the retweet equals a ‘public recognition’ and it can serve as an ‘opening gift’ with the intention to receive retweets in return.
What does the notion of the gift economy help us to see when we look at patterns in how content travels through Twitter?
We explicitly refer to your recent work on spreadable media where you employ the notion of gift economy to explain spreadability. We agree with you that this concept provides more plausible explanations for the distribution of online content than the notion of ‘viral distribution’. The retweet, the repin, the favourite are intrinsically related to attention. However, they are ‘cheap’ gifts as they are abundant. But such a gift can gain more value through the status of the user retweeting a less popular account and hence drawing attention to it. Therefore it is unsurprising that we find politicians mostly retweeting their own party members. Members of the Favstar scene frequently retweet accounts that are equally popular. They form a retweet cartell, very similar to academic citation-cartels. However, when we look at the @replies within our sample of Dutch MP’s we can see that they do not limit their communication to their own party members but with colleagues from all parties. Therefore, we conclude that if attention is drawn to messages through retweeting, users become selective in whom to award the ‘gift’ of a retweet.
I do not know how Paßmann and Boeschoten feel about it, but I would not necessarily stick to the strict economic understanding of the ‘gift economy’. I think it will prove even more useful to adapt the term. It is most likely a feature of stimulating communication and connection. With communication, I mean ephemeral communication, not conversations. The ‘gift’ is important to fuel initial contact making. Features as the retweet, the favourite, the repin, the +1 etc. are the grease of initial social interaction on large platforms. They facilitate low threshold communication; communication is the wrong word, and even contact is not covering it. It is something between a mere ping, recognition and contact. But it is crucial to enable interaction of users and spreadability of content in social media.
Your research is interesting for the ways that it combines large-scale/quantiative “sentiment analysis” tools with more qualitative use of cultural theory. Does this reflect different skill sets within the team of researchers? Are there any insights you’d like to share about mixed methods research growing out of this project?
I’m teaching at a media studies department within the Utrecht University humanities faculty, where usually qualitative research methods are paramount. But researching new media where any user activity produces data that can be analyzed stimulates to employ those data for research. These digital methods -as Richard Rogers has dubbed them- are invaluable expansion of our tool set. In the meantime many applications are available and many more are underway. Commercial platforms provide tools, but also the two main pioneering groups in this area, Manovich’s Cultural Analytics and Roger’s Digital Methods Initiative provide handy tools on their websites. For our Utrecht Data School we teamed up with Buzzcapture as a technological partner that supports our research actively with tools for data aggregation and social media data analysis. We conduct research concerning specific questions for our partners from public administrations, NGO’s and corporations. However, we take the liberty of asking different questions than the partners posed, or approach things from different angle.
I can see that student teams quickly develop a sort of division of labor, where scraping of data, working in spreadsheets, visualizing data and networks are carried out by different members of a team. We try to prevent this as far as possible, because we want all students to be involved in the entire process of the research project from scraping the data, cleaning up the data and preparing them for analysis and visualization to interpretation and contextualization. However, this is not easy, as there are indeed many specific tasks that require specialized knowledge and skills.
This work is inherently interdisciplinary. Software developers, computer scientists, data scientists, statisticians and also data journalists are great to team up with for different research projects. We frequently invite colleagues from very different areas to participate in the Utrecht Data School, either through directly contributing to a project or to teach students.
To the humanities researcher this development is exciting for two reasons: data analysis and visualization produces new insights in the online phenomena we are investigating. But through conducting these tools and methods we learn also about their role in epistemic processes. Our knowledge society increasingly thrives on computed results and automated information processing. The computer generated infographics appear persuasively convincing. It is therefore important to develop literacy that allows us to use the tools but also to be informed about their limitations and their persuasive effect. In view of your concerns about techno-determinism -which I share- I want to emphasize that we deliberately want them to develop critical understanding for the role of information technology in our epistemic processes.
We also want them to experience how unstable, how experimental and exploratory our research activities are. Although we think the results are often compelling, we want to keep up a healthy skepticism and remain open for doing things differently. We are also aware that we are in a data-rich environment, but that unfortunately research can appear analysis-poor. And I think it is necessary for the emerging ‘digital humanities’ to make this skepticism an inherent part of their use of information technology.