Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Big Data from the bottom up:

Big Data from the bottom up: This short article argues that an adequate response to the implications for governance raised by ‘Big Data’ requires much more attention to agency and reflexivity than theories of ‘algorithmic power’ have so far allowed. It develops this through two contrasting examples: the sociological study of social actors used of analytics to meet their own social ends (for example, by community organisations) and the study of actors’ attempts to build an economy of information more open to civic intervention than the existing one (for example, in the environmental sphere). The article concludes with a consideration of the broader norms that might contextualise these empirical studies, and proposes that they can be understood in terms of the notion of voice, although the practical implementation of voice as a norm means that voice must sometimes be considered via the notion of transparency. Keywords Agency, reflexivity, analytics, political economy, voice, transparency changes. Without a doubt, the information types that Introduction management and governance take as their starting- We are living through a transformation of governance – point have changed: it is digital infrastructures of col- both its mechanisms and reference-points – which is lection, transmission, analysis and presentation that likely to have profound implications for practical pro- have made possible continuous data-mining. cesses of government and everyday understandings of Compared to representative sampling, such new the social world. A shift is under way from discrete approaches to data collection are totalising; they are forms of intervention in social space based on intermit- also characterised by the aggregation of multiple data tent and/or specific information-gathering to continu- sets through the use of calculation algorithms. This ous processes of management based on total and seemingly increased role for algorithms has led some unremitting surveillance (Ruppert, 2011). Both man- commentators to focus on the dominance of ‘algorith- agement and government increasingly are becoming mic power’ (Lash, 2007), an approach that leaves no predicated upon the continuous gathering and analysis room for agency or reflexivity on the part of ‘smaller’ of dynamically collected, individual-level data about actors. We posit that emerging cultures of data collec- what people are, do and say (‘Big Data’). However mis- tion deserve to be examined in a way that foregrounds leading or mythical some narratives around Big Data the agency and reflexivity of individual actors as well as (Boyd and Crawford, 2011; Couldry, 2013), the actual the variable ways in which power and participation are processes of data-gathering, data-processing and constructed and enacted. organisational adjustment associated with such narra- tives are not mythical; they constitute an important, if highly contested, ‘fact’ with which all social actors must London School of Economics, London, UK deal. This article will offer a social approach to the construction and use of such data and related analytics. Corresponding author: The possibility of such a social approach to Big Data Nick Couldry, Department of Media and Communications, London has, until now, been obscured by unnecessarily general- School of Economics, Houghton St., London WC2A 2AE, UK. ised readings of the consequences of these broad Email: N.Couldry@lse.ac.uk Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http:// www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (http://www.uk.sagepub.com/aboutus/open- access.htm). 2 Big Data & Society This more agent-focused inquiry into the conse- Doing social analytics quences of algorithmic calculation’s deep embedding in everyday life has been foreshadowed in some earlier Our first example of a more agent-focused account of debates, notably Beer’s (2009) response to Lash’s (2007) Big Data is what has been called ‘social analytics’ (see argument that ‘algorithmic power’ has changed the Couldry et al., forthcoming, for a much more detailed nature of hegemony. As Beer (2009: 999) noted, soci- account). A social analytics approach is an explicitly ology must also ‘focus ... on those who engage with the sociological treatment of how analytics get used by a software in their everyday lives’. Such a focus does not range of social actors. Such an approach aims to cap- come naturally within Lash’s broadly philosophical for- ture how particular actors reflect upon, and adjust, mulations of issues in social theory which foreground ‘a their online presence and the actions that feed into it, collapse of ontology and epistemology’ (Lash, 2006: through the use of ‘analytics’. ‘Analytics’ here is used 581), and a new power-laden regime of ‘facticity’ broadly to cover both basic analytics (the automated (Lash, 2007: 56) in which ‘there is no time, nor space measurement and counting installed within the oper- ... for reflection’ (Lash, 2002: 18). If that were right, ation of digital platforms and associated websites, why pay close attention to what actors say when they apps and tools) and the adjustments made by actors ‘reflect’ on their position in the social world? But this themselves in response to such measurement and count- analytic closure is unhelpful. ing operations. Platforms that count and sort online Needed instead is a more open enquiry into what data, such as Google and Facebook, work automatic- actual social actors, and groups of actors, are doing ally via algorithms, often allowing users only limited under these conditions in a variety of places and set- degrees of manual adjustment (van Dijck, 2013). tings. Without denying of course the ‘generative’ Other adjustments around those operations may take importance of algorithms (Lash, 2007: 71) when direct digital form (a website redesign) or organisa- embedded in modes of calculation, processing and tional form (an adjustment in an organisation’s man- rule, we need to remember that social actors are often agement of its resources). In all these cases, the variable themselves aware of being classified. Even if they are use of analytics is a social process involving reflection, not privy to the details of when, by whom, and how monitoring and adjustment. they have been classified, that this has happened is By ‘social actors’ we mean actors with social ends something of which they are aware, and indeed one of over and above the basic aim of generating and analys- the main ‘facts’ they have to deal with as social actors. ing data (usually for profit): that basic aim in itself is of We need to become sensitive to what Beer (2009: 998) little sociological interest. The broader sociological has called people’s ‘classificatory imagination’ and, interest starts when there is some tension, actual or over the longer term, the wider ‘social imaginaries’ potential, between the aims that social actors are (Mansell, 2012; Taylor, 2005) that may be emerging trying to achieve and the interpretations of their activ- around these new cultures of data collection. ities that analytics generate. This use of the term ‘social Beer goes on helpfully to distinguish three levels of analytics’ encompasses, but goes beyond, the everyday resulting empirical research: first, regarding the ‘technical’ use of the term ‘analytics’ to mean the meas- ‘organizations that establish and activate Web 2.0 urement and reporting of internet data. The mutual applications’; second, regarding the ‘software infra- intertwining of human and material agency is hardly structures and their applications on the web’; and a new insight (Pickering, 1995: 15–20), but it acquires third, regarding how the first two levels ‘play out in a special interest when analytics’ operations are opaque the lives of those that use (or do not use) particular to non-expert social actors who must work hard to web applications’ (2009: 998). We would like in this acquire control over them. short article to build particularly on Beer’s third level, One key variable in such research is what is mea- and on the lessons of our own empirical researches, to sured and analysed, the ‘object’ of analytics. The under- map out some more detailed and concrete ways of lying data’s relationship to an organisation’s online researching the everyday uses of data and analytics presence may be more or less direct: direct if the data from a social perspective. The result is to open up a is literally about that organisation’s online presence much wider and more varied space of agency and (numbers of unique users, their characteristics, types reflexivity than allowed for in philosophical accounts. of interaction with online content); or indirect if the The likely outcome may be no less critical of Big data is not about an organisation’s online presence, Data’s implications, but will develop critique through but is generated or presented online, becoming part of a more nuanced characterisation of ‘Big Data’ as a how that organisation is judged by online visitors variegated space of action, albeit one very different (online reviews, debates). The closeness, or distance, from the spaces in which pre-digital social actors of the relation between the object of data analysis operated. and the general aims and practice of social actors Couldry and Powell 3 clearly will shape the degree of tension and reflexivity mediated messages solidifies control and results in that exists over the implementation of analytics. At one things like propaganda, but we can also see how alter- end of the spectrum will be cases where analytics are native media producers can wrest control of ideas used directly to support other mechanisms of power and their representation to challenge that kind of (e.g. performance management); at the other end will hegemony. be cases where what is at stake in the use of analytics is Broadcast models have however been overtaken, the broad redefinition of an organisation’s aims and for important purposes, by models of mass self- performance, with no direct impact on the evaluation communication. Whereas institutionalised mass media or management of individuals. In the former case, is structured to disseminate messages from one to social analytics may merge into the study of manage- many, mass self-communication is structured to invite ment and power; in the latter case, social analytics may continual input of data by individuals. This reorganisa- be something closer to a phenomenology of how social tion of media production initially seemed to promise a actors and organisations with social aims appear to reconfiguration of the top-down production of ideology themselves, and to the world, under digital conditions. and the bottom-up resistance to it, but as political– Other variables when doing social analytics will economic analyses have developed, we are beginning include the degree of technical expertise of the actors to see how such shifts have also led to the production involved, including the degree to which they can go of data replacing the production of audiences. beyond merely using off-the-shelf analytics to customis- If the exemplary product of institutionalised mass ing them, or perhaps even developing their own ana- media is propaganda, the exemplary product of mass lytic tools and data-collection designs. Financial and self-communication is data. A mass media apparatus other resources will also affect how far the processes requests information to be disseminated from the one which social analytics studies can develop, or get to the many; its economic model uses this information blocked, for example, if the staff to do the analytic to generate an audience whose attention can be sold to work that would enable a richer re-evaluation of an an advertiser. In the mass self-communication model organisation’s digital presence cease to be available. individuals are still part of an aggregate product to be Expertise and resources are, of course, variables in sold, but instead of their attention on a single message any fieldwork setting. produced for broadcast, it is their individual acts of Within these basic parameters, however, social ana- communication that comprise the ‘Big Data’ and lytics promise a rich vein of inquiry into the conditions drive much media value-extraction. of data use and analytics use, from the perspective of Early critics of mass self-communication noted that social actors who are not principally experts in relation the model encouraged individuals to create ‘content’ to data or algorithms, but who look to them to do cer- that was then sold to others in order to capture their tain work towards other ends. It has so far been explored attention (Terranova, 2000; van Dijck, 2013). However, in the context of community and civic activism, but it has ‘content’ is still expressive, even when it is sold to cap- the potential to be expanded to many more areas. ture attention. A more complicated issue concerns the data that is produced, often unwittingly, which now generates much of the value in the newest iteration of Data as media the contribution economy. Many everyday activities For media scholars more generally, the shift to a data- now produce data without requiring human meaning- rich environment poses challenges for a robust under- construction (or even basic consent). The rise of sensor standing of how agency and expression might still work networks has meant that increasingly individuals within that environment. The critical tradition in media are producing not ‘content’ composed of messages con- and communications has largely been concerned with taining intrinsic or constructed meaning, but mere data the operation of power in the construction of systems of – temperature readings, status updates, location coord- symbolic mediation – for example, the function of ideo- inates, tracks, traces and check-ins. Not one of these logical systems (in the Marxist tradition) or the individual data-types is necessarily meaningful in itself Gramscian concept of hegemony. These strategies – but taken together, either through aggregation, cor- have allowed media and communication scholars to relation or calculation, such data provide large ‘work backwards’ through systems of symbolic medi- amounts of information. The difference between this and the ‘content’ that mass self-communication prom- ation in order to understand the process and initial starting points of mediated ‘messages’. This focus on ises to distribute is that the meaning of data is made not the symbolic quality of media messages allows us to semantically (through expression and interpretation) examine power relationships from several different but through processing – especially the matching vantage points. Within traditional broadcast media of metadata (Boellstorf, 2013). Big Data sets are forms we can observe how the symbolic control of composed of numerous pieces of information that can 4 Big Data & Society be cross-compared, aggregated and disaggregated and Voice, transparency and power made very finely grained, not things whose creators necessarily endowed with meaning. In mining the The rise of analytics presents a significant normative data, more insights are made available about more challenge for scholars, activists and others who seek aspects of everyday life but no opportunity is provided to understand how humanity, sociability and experi- for these insights to be folded back into the experience ence are represented. The daily practices of grappling of everyday life. In this context, is there any scope, as with data and with the consequences of data analyses Boellstorf urges, for integrating the epistemic perspec- generate new questions about what and whose power tives of ethnography back into the calculative logic of gets exercised through such practices, and to what meta-data? degree such exercises of power are satisfactorily made All along, the political economy of personal data, as accountable. One approach to these challenges is anticipated by Gandy (1993), has been concerned with through attention to problems of voice (Couldry, value created through the aggregation and calculation 2010). Voice, understood as a value for social organisa- of individual traces. Even if we leave aside the expres- tion (Couldry, 2010: ch. 1), involves taking into account sive quality of individual acts of communication online, agents’ practices of giving an account of themselves and the production of data as a by-product of everyday life their conditions of life. The value of voice is essential to practices enacts a particular political economics of the workings of any models so far developed of demo- media, undertaken within a situation of pervasive sur- cratic institutions, but it is not immediately compatible veillance and generalised authoritarianism (Cohen with a world saturated with the automated aggregation 2012). But the potential disconnect between system of analytic mechanisms that are not, even in principle, and experience, phenomenology and political economy, open to any continuous human interpretation or can be overcome by examining on the ground agents’ review. strategies for building alternative economies of infor- While the notion of voice insists upon organisational mation. Such alternative economies are being devel- processes being accountable to the subjectivities and oped in several areas related to environment and expressiveness of all, the movement towards more sustainability, including projects that use data sources casual, automatic sensing and its calculative rather to make provenance and supply chains visible, and than epistemic logic seems to eliminate this account- those that encourage individuals and communities to ability. Yet clearly something similar to ‘voice’ is collect data as a means to make environmental issues required in this new world, and this is not just a visible by challenging conventional data collection. matter of democracy: ‘we have no idea’, wrote Paul Academic projects like Wikichains (Graham, 2010) Ricoeur, ‘what a culture would be where no one any and start-up companies like Provenance.it (2013) aggre- longer knew what it meant to narrate things’ (Couldry, gate various forms of data about the production, 2010: 1, quoting Ricoeur, 1984: 29). At present, the distribution and supply chains of manufactured objects proxy for voice in the algorithmic domain is the as a means of drawing attention to their long-term notion that data gathering processes ought to be trans- ecological and economic costs. While Provenance.it parent, and the logic of calculation revealed. A focus on remains anchored in a consumer-based economic transparency could begin to foreground notions of model, it does illustrate how alternative modes of accountability in data calculation, ownership and use. data collection and analysis could shift agency and rep- Notions of transparency have been discussed with resentation, especially if it permitted for greater reflex- respect to government production and use of data ivity. Similarly, NGOs like Mapping for Change (2013) (Tkacz, 2012). Yet despite pledging to make public have supported individuals and community groups in data collection transparent, governments like the US gathering environmental data (like air quality and and the UK in fact collect much more information noise) as a means of engaging with gaps and flaws in via surveillance projects and partnerships with informa- official data. These actions intervene in efforts to use tion technology companies. With the reform of the such environmental data within top-down governance USA’s National Security Administration, perhaps processes. As Gabrys (2014) identifies, such citizen sci- more attention will begin to be paid to the data collec- ence efforts must be enfolded and imagined in processes tion practices of the technology sector, making more of of environmental governance or ‘biopolitics 2.0’. These them visible. This kind of transparency goes part of the examples illustrate two ways that an alternative eco- way to establishing accountability, but it still fails to nomics of information might employ calculation of address accountability and reflexivity. A refined multiple data sources or generation of alternative concept of transparency that is sensitive to the sources to illustrate or critique power relations, meaning that data trails might form (even if it cannot although they also illustrate the ambiguity of account- be sensitive to the meaning inherent in their produc- ability within these processes. tion) might go some way to addressing this. This is a Couldry and Powell 5 videoAndAudio/channels/publicLecturesAndEvents/ tricky proposal: unless and until the unconscious pro- player.aspx?id¼2120 (accessed 21 November 2013). duction of data can be conceived of as a form of expres- Couldry N, Fotopoulou A and Dickens L (forthcoming). sion, the philosophical basis for such an expansive Real Social Analytics: A Contribution Towards the transparency will be difficult to establish. One possible Phenomenology of a Digital World. way to proceed might be to highlight not just the risks Gabrys J (2014) Programming environments: Environmental- of creating and sharing data but the opportunities as ity and citizen sensing in the smart city. Environment and well. The practices of social analytics and citizen science Planning D: Society and Space 32(1): 30–48. have the potential to establish these opportunities, Gandy O (1993) Toward a political economy of personal ambiguous as they may be. information. Critical Studies in Mass Communication We hope that, as the debates about Big Data and 10(1): 70–97. society continue and their democratic stakes become Graham M (2010) ‘WikiChains: Encouraging Transparency in clearer, the values implicit in the terms ‘voice’ and Commodity Chains’ Research Project. Available at: http:// ‘transparency’ will themselves begin to converge in www.oii.ox.ac.uk/research/projects/?id¼75 (accessed 30 more satisfying ways than are at present possible. May 2014). Lash S (2002) Critique of Information. London: Sage. Lash S (2006) Dialectic of information? A response to Taylor. Declaration of conflicting interest Information Community & Society 9(5): 572–581. The authors declare that there is no conflict of interest. Lash S (2007) Power after hegemony: Cultural studies in mutation. Theory, Culture & Society 24(3): 55–78. Funding Mansell R (2012) Imagining the Internet. Oxford: Oxford This research received no specific grant from any funding University Press. agency in the public, commercial, or not-for-profit sectors. Mapping for Change (2013) Services: Citizen Science. Available at: http://www.mappingforchange.org.uk/ References services/citizen-science/ (accessed 30 May 2014). Pickering A (1995) The Mangle of Practice. Chicago: Chicago Beer D (2009) Power through the algorithm? Participatory University Press. web cultures and the technological unconscious. New Provenance.it (2013) About Provenance. Available at: https:// Media & Society 11: 985–1002. www.provenance.it/about (accessed 30 May 2014). Boellstorff T (2013) Making Big Data, in Theory. First Ricoeur P (1984) Time and Narrative, Vol. 2. Chicago: Monday, [S.l.], September 2013. ISSN 13960466. Chicago University Press. Available at: http://firstmonday.org/ojs/index.php/fm/ Ruppert E (2011) Population objects: Interpassive subjects. article/view/4869/3750 (accessed 27 January 2014). Sociology 45(2): 218–233. boyd d and Crawford K (2011) Critical questions for Big Taylor C (2005) Modern Social Imaginaries. Durham, NC: Data: Provocations for a cultural, technological and schol- Duke University Press. arly phenomenon. Information, Communication and Terranova T (2000) Free labor: Producing culture for the Society 15(5): 662–679. digital economy. Social Text 18(2): 33–58. Castells M (2009) Communication Power. Oxford: Oxford Tkacz N (2012) From open source to open government: University Press. A critique of open politics. Ephemera 12(4). Available at: Cohen J (2012) Configuring the Networked Self. New Haven: http://www.ephemerajournal.org/contribution/open- Yale University Press. source-open-government-critique-open-politics-0 Couldry N (2010) Why Voice Matters. London: Sage. Couldry N (2013) A Necessary Disenchantment: Myth, (accessed 30 May 2014). Agency and Injustice in a Digital Age. Inaugural lecture at Van Dijck J (2013) The Culture of Connectivity. Oxford: LSE. Available at: http://www.lse.ac.uk/newsAndMedia/ Oxford University Press. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Big Data & Society SAGE

Big Data from the bottom up:

Big Data & Society , Volume 1 (2): 1 – Jul 1, 2014

Loading next page...
 
/lp/sage/big-data-from-the-bottom-up-L0LOyNk71t

References (25)

Publisher
SAGE
Copyright
Copyright © 2022 by SAGE Publications Ltd, unless otherwise noted. Manuscript content on this site is licensed under Creative Commons Licenses.
ISSN
2053-9517
eISSN
2053-9517
DOI
10.1177/2053951714539277
Publisher site
See Article on Publisher Site

Abstract

This short article argues that an adequate response to the implications for governance raised by ‘Big Data’ requires much more attention to agency and reflexivity than theories of ‘algorithmic power’ have so far allowed. It develops this through two contrasting examples: the sociological study of social actors used of analytics to meet their own social ends (for example, by community organisations) and the study of actors’ attempts to build an economy of information more open to civic intervention than the existing one (for example, in the environmental sphere). The article concludes with a consideration of the broader norms that might contextualise these empirical studies, and proposes that they can be understood in terms of the notion of voice, although the practical implementation of voice as a norm means that voice must sometimes be considered via the notion of transparency. Keywords Agency, reflexivity, analytics, political economy, voice, transparency changes. Without a doubt, the information types that Introduction management and governance take as their starting- We are living through a transformation of governance – point have changed: it is digital infrastructures of col- both its mechanisms and reference-points – which is lection, transmission, analysis and presentation that likely to have profound implications for practical pro- have made possible continuous data-mining. cesses of government and everyday understandings of Compared to representative sampling, such new the social world. A shift is under way from discrete approaches to data collection are totalising; they are forms of intervention in social space based on intermit- also characterised by the aggregation of multiple data tent and/or specific information-gathering to continu- sets through the use of calculation algorithms. This ous processes of management based on total and seemingly increased role for algorithms has led some unremitting surveillance (Ruppert, 2011). Both man- commentators to focus on the dominance of ‘algorith- agement and government increasingly are becoming mic power’ (Lash, 2007), an approach that leaves no predicated upon the continuous gathering and analysis room for agency or reflexivity on the part of ‘smaller’ of dynamically collected, individual-level data about actors. We posit that emerging cultures of data collec- what people are, do and say (‘Big Data’). However mis- tion deserve to be examined in a way that foregrounds leading or mythical some narratives around Big Data the agency and reflexivity of individual actors as well as (Boyd and Crawford, 2011; Couldry, 2013), the actual the variable ways in which power and participation are processes of data-gathering, data-processing and constructed and enacted. organisational adjustment associated with such narra- tives are not mythical; they constitute an important, if highly contested, ‘fact’ with which all social actors must London School of Economics, London, UK deal. This article will offer a social approach to the construction and use of such data and related analytics. Corresponding author: The possibility of such a social approach to Big Data Nick Couldry, Department of Media and Communications, London has, until now, been obscured by unnecessarily general- School of Economics, Houghton St., London WC2A 2AE, UK. ised readings of the consequences of these broad Email: N.Couldry@lse.ac.uk Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http:// www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (http://www.uk.sagepub.com/aboutus/open- access.htm). 2 Big Data & Society This more agent-focused inquiry into the conse- Doing social analytics quences of algorithmic calculation’s deep embedding in everyday life has been foreshadowed in some earlier Our first example of a more agent-focused account of debates, notably Beer’s (2009) response to Lash’s (2007) Big Data is what has been called ‘social analytics’ (see argument that ‘algorithmic power’ has changed the Couldry et al., forthcoming, for a much more detailed nature of hegemony. As Beer (2009: 999) noted, soci- account). A social analytics approach is an explicitly ology must also ‘focus ... on those who engage with the sociological treatment of how analytics get used by a software in their everyday lives’. Such a focus does not range of social actors. Such an approach aims to cap- come naturally within Lash’s broadly philosophical for- ture how particular actors reflect upon, and adjust, mulations of issues in social theory which foreground ‘a their online presence and the actions that feed into it, collapse of ontology and epistemology’ (Lash, 2006: through the use of ‘analytics’. ‘Analytics’ here is used 581), and a new power-laden regime of ‘facticity’ broadly to cover both basic analytics (the automated (Lash, 2007: 56) in which ‘there is no time, nor space measurement and counting installed within the oper- ... for reflection’ (Lash, 2002: 18). If that were right, ation of digital platforms and associated websites, why pay close attention to what actors say when they apps and tools) and the adjustments made by actors ‘reflect’ on their position in the social world? But this themselves in response to such measurement and count- analytic closure is unhelpful. ing operations. Platforms that count and sort online Needed instead is a more open enquiry into what data, such as Google and Facebook, work automatic- actual social actors, and groups of actors, are doing ally via algorithms, often allowing users only limited under these conditions in a variety of places and set- degrees of manual adjustment (van Dijck, 2013). tings. Without denying of course the ‘generative’ Other adjustments around those operations may take importance of algorithms (Lash, 2007: 71) when direct digital form (a website redesign) or organisa- embedded in modes of calculation, processing and tional form (an adjustment in an organisation’s man- rule, we need to remember that social actors are often agement of its resources). In all these cases, the variable themselves aware of being classified. Even if they are use of analytics is a social process involving reflection, not privy to the details of when, by whom, and how monitoring and adjustment. they have been classified, that this has happened is By ‘social actors’ we mean actors with social ends something of which they are aware, and indeed one of over and above the basic aim of generating and analys- the main ‘facts’ they have to deal with as social actors. ing data (usually for profit): that basic aim in itself is of We need to become sensitive to what Beer (2009: 998) little sociological interest. The broader sociological has called people’s ‘classificatory imagination’ and, interest starts when there is some tension, actual or over the longer term, the wider ‘social imaginaries’ potential, between the aims that social actors are (Mansell, 2012; Taylor, 2005) that may be emerging trying to achieve and the interpretations of their activ- around these new cultures of data collection. ities that analytics generate. This use of the term ‘social Beer goes on helpfully to distinguish three levels of analytics’ encompasses, but goes beyond, the everyday resulting empirical research: first, regarding the ‘technical’ use of the term ‘analytics’ to mean the meas- ‘organizations that establish and activate Web 2.0 urement and reporting of internet data. The mutual applications’; second, regarding the ‘software infra- intertwining of human and material agency is hardly structures and their applications on the web’; and a new insight (Pickering, 1995: 15–20), but it acquires third, regarding how the first two levels ‘play out in a special interest when analytics’ operations are opaque the lives of those that use (or do not use) particular to non-expert social actors who must work hard to web applications’ (2009: 998). We would like in this acquire control over them. short article to build particularly on Beer’s third level, One key variable in such research is what is mea- and on the lessons of our own empirical researches, to sured and analysed, the ‘object’ of analytics. The under- map out some more detailed and concrete ways of lying data’s relationship to an organisation’s online researching the everyday uses of data and analytics presence may be more or less direct: direct if the data from a social perspective. The result is to open up a is literally about that organisation’s online presence much wider and more varied space of agency and (numbers of unique users, their characteristics, types reflexivity than allowed for in philosophical accounts. of interaction with online content); or indirect if the The likely outcome may be no less critical of Big data is not about an organisation’s online presence, Data’s implications, but will develop critique through but is generated or presented online, becoming part of a more nuanced characterisation of ‘Big Data’ as a how that organisation is judged by online visitors variegated space of action, albeit one very different (online reviews, debates). The closeness, or distance, from the spaces in which pre-digital social actors of the relation between the object of data analysis operated. and the general aims and practice of social actors Couldry and Powell 3 clearly will shape the degree of tension and reflexivity mediated messages solidifies control and results in that exists over the implementation of analytics. At one things like propaganda, but we can also see how alter- end of the spectrum will be cases where analytics are native media producers can wrest control of ideas used directly to support other mechanisms of power and their representation to challenge that kind of (e.g. performance management); at the other end will hegemony. be cases where what is at stake in the use of analytics is Broadcast models have however been overtaken, the broad redefinition of an organisation’s aims and for important purposes, by models of mass self- performance, with no direct impact on the evaluation communication. Whereas institutionalised mass media or management of individuals. In the former case, is structured to disseminate messages from one to social analytics may merge into the study of manage- many, mass self-communication is structured to invite ment and power; in the latter case, social analytics may continual input of data by individuals. This reorganisa- be something closer to a phenomenology of how social tion of media production initially seemed to promise a actors and organisations with social aims appear to reconfiguration of the top-down production of ideology themselves, and to the world, under digital conditions. and the bottom-up resistance to it, but as political– Other variables when doing social analytics will economic analyses have developed, we are beginning include the degree of technical expertise of the actors to see how such shifts have also led to the production involved, including the degree to which they can go of data replacing the production of audiences. beyond merely using off-the-shelf analytics to customis- If the exemplary product of institutionalised mass ing them, or perhaps even developing their own ana- media is propaganda, the exemplary product of mass lytic tools and data-collection designs. Financial and self-communication is data. A mass media apparatus other resources will also affect how far the processes requests information to be disseminated from the one which social analytics studies can develop, or get to the many; its economic model uses this information blocked, for example, if the staff to do the analytic to generate an audience whose attention can be sold to work that would enable a richer re-evaluation of an an advertiser. In the mass self-communication model organisation’s digital presence cease to be available. individuals are still part of an aggregate product to be Expertise and resources are, of course, variables in sold, but instead of their attention on a single message any fieldwork setting. produced for broadcast, it is their individual acts of Within these basic parameters, however, social ana- communication that comprise the ‘Big Data’ and lytics promise a rich vein of inquiry into the conditions drive much media value-extraction. of data use and analytics use, from the perspective of Early critics of mass self-communication noted that social actors who are not principally experts in relation the model encouraged individuals to create ‘content’ to data or algorithms, but who look to them to do cer- that was then sold to others in order to capture their tain work towards other ends. It has so far been explored attention (Terranova, 2000; van Dijck, 2013). However, in the context of community and civic activism, but it has ‘content’ is still expressive, even when it is sold to cap- the potential to be expanded to many more areas. ture attention. A more complicated issue concerns the data that is produced, often unwittingly, which now generates much of the value in the newest iteration of Data as media the contribution economy. Many everyday activities For media scholars more generally, the shift to a data- now produce data without requiring human meaning- rich environment poses challenges for a robust under- construction (or even basic consent). The rise of sensor standing of how agency and expression might still work networks has meant that increasingly individuals within that environment. The critical tradition in media are producing not ‘content’ composed of messages con- and communications has largely been concerned with taining intrinsic or constructed meaning, but mere data the operation of power in the construction of systems of – temperature readings, status updates, location coord- symbolic mediation – for example, the function of ideo- inates, tracks, traces and check-ins. Not one of these logical systems (in the Marxist tradition) or the individual data-types is necessarily meaningful in itself Gramscian concept of hegemony. These strategies – but taken together, either through aggregation, cor- have allowed media and communication scholars to relation or calculation, such data provide large ‘work backwards’ through systems of symbolic medi- amounts of information. The difference between this and the ‘content’ that mass self-communication prom- ation in order to understand the process and initial starting points of mediated ‘messages’. This focus on ises to distribute is that the meaning of data is made not the symbolic quality of media messages allows us to semantically (through expression and interpretation) examine power relationships from several different but through processing – especially the matching vantage points. Within traditional broadcast media of metadata (Boellstorf, 2013). Big Data sets are forms we can observe how the symbolic control of composed of numerous pieces of information that can 4 Big Data & Society be cross-compared, aggregated and disaggregated and Voice, transparency and power made very finely grained, not things whose creators necessarily endowed with meaning. In mining the The rise of analytics presents a significant normative data, more insights are made available about more challenge for scholars, activists and others who seek aspects of everyday life but no opportunity is provided to understand how humanity, sociability and experi- for these insights to be folded back into the experience ence are represented. The daily practices of grappling of everyday life. In this context, is there any scope, as with data and with the consequences of data analyses Boellstorf urges, for integrating the epistemic perspec- generate new questions about what and whose power tives of ethnography back into the calculative logic of gets exercised through such practices, and to what meta-data? degree such exercises of power are satisfactorily made All along, the political economy of personal data, as accountable. One approach to these challenges is anticipated by Gandy (1993), has been concerned with through attention to problems of voice (Couldry, value created through the aggregation and calculation 2010). Voice, understood as a value for social organisa- of individual traces. Even if we leave aside the expres- tion (Couldry, 2010: ch. 1), involves taking into account sive quality of individual acts of communication online, agents’ practices of giving an account of themselves and the production of data as a by-product of everyday life their conditions of life. The value of voice is essential to practices enacts a particular political economics of the workings of any models so far developed of demo- media, undertaken within a situation of pervasive sur- cratic institutions, but it is not immediately compatible veillance and generalised authoritarianism (Cohen with a world saturated with the automated aggregation 2012). But the potential disconnect between system of analytic mechanisms that are not, even in principle, and experience, phenomenology and political economy, open to any continuous human interpretation or can be overcome by examining on the ground agents’ review. strategies for building alternative economies of infor- While the notion of voice insists upon organisational mation. Such alternative economies are being devel- processes being accountable to the subjectivities and oped in several areas related to environment and expressiveness of all, the movement towards more sustainability, including projects that use data sources casual, automatic sensing and its calculative rather to make provenance and supply chains visible, and than epistemic logic seems to eliminate this account- those that encourage individuals and communities to ability. Yet clearly something similar to ‘voice’ is collect data as a means to make environmental issues required in this new world, and this is not just a visible by challenging conventional data collection. matter of democracy: ‘we have no idea’, wrote Paul Academic projects like Wikichains (Graham, 2010) Ricoeur, ‘what a culture would be where no one any and start-up companies like Provenance.it (2013) aggre- longer knew what it meant to narrate things’ (Couldry, gate various forms of data about the production, 2010: 1, quoting Ricoeur, 1984: 29). At present, the distribution and supply chains of manufactured objects proxy for voice in the algorithmic domain is the as a means of drawing attention to their long-term notion that data gathering processes ought to be trans- ecological and economic costs. While Provenance.it parent, and the logic of calculation revealed. A focus on remains anchored in a consumer-based economic transparency could begin to foreground notions of model, it does illustrate how alternative modes of accountability in data calculation, ownership and use. data collection and analysis could shift agency and rep- Notions of transparency have been discussed with resentation, especially if it permitted for greater reflex- respect to government production and use of data ivity. Similarly, NGOs like Mapping for Change (2013) (Tkacz, 2012). Yet despite pledging to make public have supported individuals and community groups in data collection transparent, governments like the US gathering environmental data (like air quality and and the UK in fact collect much more information noise) as a means of engaging with gaps and flaws in via surveillance projects and partnerships with informa- official data. These actions intervene in efforts to use tion technology companies. With the reform of the such environmental data within top-down governance USA’s National Security Administration, perhaps processes. As Gabrys (2014) identifies, such citizen sci- more attention will begin to be paid to the data collec- ence efforts must be enfolded and imagined in processes tion practices of the technology sector, making more of of environmental governance or ‘biopolitics 2.0’. These them visible. This kind of transparency goes part of the examples illustrate two ways that an alternative eco- way to establishing accountability, but it still fails to nomics of information might employ calculation of address accountability and reflexivity. A refined multiple data sources or generation of alternative concept of transparency that is sensitive to the sources to illustrate or critique power relations, meaning that data trails might form (even if it cannot although they also illustrate the ambiguity of account- be sensitive to the meaning inherent in their produc- ability within these processes. tion) might go some way to addressing this. This is a Couldry and Powell 5 videoAndAudio/channels/publicLecturesAndEvents/ tricky proposal: unless and until the unconscious pro- player.aspx?id¼2120 (accessed 21 November 2013). duction of data can be conceived of as a form of expres- Couldry N, Fotopoulou A and Dickens L (forthcoming). sion, the philosophical basis for such an expansive Real Social Analytics: A Contribution Towards the transparency will be difficult to establish. One possible Phenomenology of a Digital World. way to proceed might be to highlight not just the risks Gabrys J (2014) Programming environments: Environmental- of creating and sharing data but the opportunities as ity and citizen sensing in the smart city. Environment and well. The practices of social analytics and citizen science Planning D: Society and Space 32(1): 30–48. have the potential to establish these opportunities, Gandy O (1993) Toward a political economy of personal ambiguous as they may be. information. Critical Studies in Mass Communication We hope that, as the debates about Big Data and 10(1): 70–97. society continue and their democratic stakes become Graham M (2010) ‘WikiChains: Encouraging Transparency in clearer, the values implicit in the terms ‘voice’ and Commodity Chains’ Research Project. Available at: http:// ‘transparency’ will themselves begin to converge in www.oii.ox.ac.uk/research/projects/?id¼75 (accessed 30 more satisfying ways than are at present possible. May 2014). Lash S (2002) Critique of Information. London: Sage. Lash S (2006) Dialectic of information? A response to Taylor. Declaration of conflicting interest Information Community & Society 9(5): 572–581. The authors declare that there is no conflict of interest. Lash S (2007) Power after hegemony: Cultural studies in mutation. Theory, Culture & Society 24(3): 55–78. Funding Mansell R (2012) Imagining the Internet. Oxford: Oxford This research received no specific grant from any funding University Press. agency in the public, commercial, or not-for-profit sectors. Mapping for Change (2013) Services: Citizen Science. Available at: http://www.mappingforchange.org.uk/ References services/citizen-science/ (accessed 30 May 2014). Pickering A (1995) The Mangle of Practice. Chicago: Chicago Beer D (2009) Power through the algorithm? Participatory University Press. web cultures and the technological unconscious. New Provenance.it (2013) About Provenance. Available at: https:// Media & Society 11: 985–1002. www.provenance.it/about (accessed 30 May 2014). Boellstorff T (2013) Making Big Data, in Theory. First Ricoeur P (1984) Time and Narrative, Vol. 2. Chicago: Monday, [S.l.], September 2013. ISSN 13960466. Chicago University Press. Available at: http://firstmonday.org/ojs/index.php/fm/ Ruppert E (2011) Population objects: Interpassive subjects. article/view/4869/3750 (accessed 27 January 2014). Sociology 45(2): 218–233. boyd d and Crawford K (2011) Critical questions for Big Taylor C (2005) Modern Social Imaginaries. Durham, NC: Data: Provocations for a cultural, technological and schol- Duke University Press. arly phenomenon. Information, Communication and Terranova T (2000) Free labor: Producing culture for the Society 15(5): 662–679. digital economy. Social Text 18(2): 33–58. Castells M (2009) Communication Power. Oxford: Oxford Tkacz N (2012) From open source to open government: University Press. A critique of open politics. Ephemera 12(4). Available at: Cohen J (2012) Configuring the Networked Self. New Haven: http://www.ephemerajournal.org/contribution/open- Yale University Press. source-open-government-critique-open-politics-0 Couldry N (2010) Why Voice Matters. London: Sage. Couldry N (2013) A Necessary Disenchantment: Myth, (accessed 30 May 2014). Agency and Injustice in a Digital Age. Inaugural lecture at Van Dijck J (2013) The Culture of Connectivity. Oxford: LSE. Available at: http://www.lse.ac.uk/newsAndMedia/ Oxford University Press.

Journal

Big Data & SocietySAGE

Published: Jul 1, 2014

Keywords: Agency; reflexivity; analytics; political economy; voice; transparency

There are no references for this article.