Data Power Conference 2015
Panel Session 5a): Data Subjects (Chair: Evelyn Ruppert)
Data subjects are not only those that work the data but those on whom data works: caught, as Foucault would have it, between subjectivation and subjection. This panel will address the production of various digital data subjects including the ‘datapreneur’, the ‘armchair auditor’, the self-quantifier, and the self-disciplining, self-exploitative, quantified academic. It will ask: how does digital data affect the experience of subjectivity, agency and power?
Data Literacy, Agency and Power
Jennifer Pybus, University of the Arts London
It is paradoxical that questions of agency arise in relation to big data considering that collectively we are a core site of its generation. Yet, given the highly proprietary nature of the devices, platforms and apps through which we generate ‘big social data’, critical questions are raised. Despite this intensive and extensive recursivity, the public imaginary lacks a clear understanding of their data outside of the platforms and apps in which it is largely generated.
This paper will therefore consider how users can reclaim agency within a digital landscape (Couldry & Turow; Andrejevic & Gates). Foucault once invited his interlocutors to ‘know oneself’ to actualize our subjective becoming. And yet, what he envisioned throughout his body of work was never subjected to the added dimensionality of the digital. This paper will therefore consider this question of ‘knowing oneself’ within our new datascape by sketching out what a preliminary framework for a more interdisciplinary approach to data literacy might look like.
Data literacy is an emergent field that is aiming to develop a set of competencies and knowledge to empower people to critically understand the dynamic flows, processes and economies related to our steadily growing digital footprint. My discussion will focus on the “Our Data Ourselves”, AHRC project at King’s College London, to consider the ways our co-researchers have helped us identify, visualize and more actively engage with the data that they have already collectively generated about themselves, to consider the agentic possibilities of the data subject.
The New Data Subject: Between Transparency and Secrecy in the Digital Age
Clare Birchall, King's College London
In the guise of transparency, digital data promises agency. But accompanying access to more data – our own, other people’s and that of the state – is a demand to act upon it. For example, we are called upon to engage with open government data as ‘datapreneurs’ in ways that will contribute to the data economy, as ‘armchair auditors’ to monitor the granular transactions of the state, and as consumers of apps based on open government data to enable us to make informed choices. This paper argues that agency is delimited by open government data as much as covert dataveillance and analyses the ‘data subject’ caught between transparency and secrecy. What are the implications of both open and covert approaches to data for citizenship and politics? What counts as a political intervention in this construction of data power?
The Quantified Academic
Gary Hall, Coventry University
The origin of the word data is as the plural of the Latin word for datum, which means a proposition that is assumed, given or taken for granted, often in order to construct a theoretical framework or draw conclusions. In engineering the datum point is the place from which measurements are taken. The datum point itself, however, is not checked or questioned: as the position from which measurements are made it is precisely a given. This paper addresses some of the datum points that are assumed and taken for granted when critical questions are raised about data’s power.
For example, a number of critics have commented recently on the way corporate social media (Facebook, YouTube, Twitter etc.) are contributing to a process of neoliberal academic subjectivation. It is a process of self-forming through the adoption of self-presentation techniques originating in the culture of Silicon Valley, including self-quantification, that can be linked to what has been termed the ‘“metricisation” of the academy’: the way academics are now exposed to a swathe of techniques for monitoring, measuring and assessing their teaching loads, journal citations, grant income, research outputs and impact, many of them enacted automatically through the algorithmic analysis of the associated data. The focus in critiques of data power of this kind, however, is almost invariably on the new, self-governing and self-exploitative data subjects academics are transitioning into. Rather less concern tends to be given over to the particular configuration of academic subjectivity they are changing from, which is often at root a liberal humanist subjectivity. By focusing on the latter datum point, this paper will show how both of these models of subjectivity – the self-disciplining neoliberal model on which the data works, and the liberal humanist model (complete with its enactment of taken for granted ideas of authorship, originality, the book and copyright) which works on the data to construct a theoretical framework and draw conclusions about its power – are involved in the subordination of academic agency and consciousness to the pre-programmed, controllable patterns of the cultural industries.
'Please Wait a Moment While We Refresh Your Assets': The Promise of Cognitive Computing
Adrian Mackenzine, Lancaster University
This paper will critically analyse a contemporary data assemblage, IBM Corporation's 'Watson.' IBM refers to 'Watson' as a 'cognitive computing' platform. The platform first became in prominent in 2011 as a winning contestant in the US quiz show 'Jeopardy.' Since that time, Watson has grown into a global assemblage, staffed by several thousand people, distributed across national and global data centres and funded by more than \$1.5 billion (USD). The paper will examine several aspects of this growth. The platform has been aligned and linked with high profile scientific institutions and grand scientific challenges (cancer, diabetes, etc.). It has been positioned in popular culture initially through 'Jeopardy' and latterly through a Youtube channel, podcasts, and then in more playful forms such as Chef Watson. At the same time, Watson has been promoted as a solution to institutional problems of managing large numbers of individuals in hospitals and universities, insurance and retail. Finally, through its the Watson Developer Cloud, the platform functions as the 'cognitive' end of apps in increasing numbers. Across these globally distributed and diverse settings, Watson claims to 'scale and democratise expertise.' Testing this claim of democratisation, the paper will suggest that Watson could be better seen as a diagram of contemporary power relations in which key techniques -- machine learning and data visualization -- are put together in making new models of local truth and establishing new relations between forces. Making sense of such diagrams could be useful in understanding the many problems associated with contemporary data economies and cultures.