Paper II: “Ethics and Data Futures”

How to reference this paper:

Pink, S., A. Markham, Y. Akama, E. Gómez Cruz, P. Lacasa, M. Poblet, S. Sumartojo (2016) DATA ETHNOGRAPHIES (2): ethics and data futures. Available online at

DATA ETHNOGRAPHIES (2): ethics and data futures

Sarah Pink, Annette Markham, Yoko Akama, Edgar Gomez Cruz, Pilar Lacasa, Marta Poblet, Shanti Sumartojo

In this Data Ethnographies position paper we outline our agenda for approaching the relationship between ethics and data futures. The paper is based on our second data ethnographies workshop which brought together a group of colleagues whose work interfaces around ethnographic practice, design and ethnographic futures research, internet ethics and digital privacy, and personal data. Two key issues were at the centre of our discussions: first how ethnographic approaches to data can enable and might also call for new thinking about ethics and futures; and second how we confront the questions of the ethics and temporality of ethnographic research in data-worlds.

The session was prefaced by Position Paper 1 – personal data in an uncertain world – which put forward the argument for a data ethnographies approach. The first paper set the ground through a discussion of how it feels to live in a world of data, and the unique positioning and capability of ethnography to create new insights and understandings that contest and situate assumptions about how data can stand for individuals or society.

Dataethnographies II.jpg

Defining Data

Much existing discussion about data concerns questions of surveillance, fear of the loss of privacy, or the idea that data does (or can) permeate, report on, or influence all domains of our lives. As danah boyd and Kate Crawford expressed in early discussions about ‘Big Data’:

Like other socio-technical phenomena, Big Data triggers both utopian and dystopian rhetoric. On one hand, Big Data is seen as a powerful tool to address various societal ills, offering the potential of new insights into areas as diverse as cancer research, terrorism, and climate change. On the other, Big Data is seen as a troubling manifestation of Big Brother, enabling invasions of privacy, decreased civil freedoms, and increased state and corporate control (2012: 663-4).

One way to get at these issues is by exploring how people experience data from an ethnographic approach, which is a way of defining the concept in situ, rather than taking data as a priori or given. Thus, we began workshop 2 by reviewing contemporary questions and debates about what we think data is. As discussed in position paper 1, this strategy underpins a data ethnographies approach already, since an ethnographic appreciation of data enables us to understand it as an experiential phenomenon, something that is in the world with us rather than as separate from us. Because an ethnographic approach to data defies its definition as an object that can manifest a truth, it questions how data can be mobilised for either social change or surveillance agendas. However we also acknowledge that data remains and functions well beyond our experience of it at the close and local level. Data become obdurate and the surveillance and loss of privacy that data practices bring to bear are felt in the world. As the construction and use of data are embedded in power relationships, they remain relevant.

Pursuing these complexities from an ethnographic perspective invites us to consider how our practices of research, analysis and dissemination are also implicated in these processes and how we might participate in more deliberate ways to create ethical and responsible approaches to researching and creating interventions for change in the world (with) data.

How data is defined, including what affordances and qualities are attributed to it, is central to understanding how different actors interpret its potential for change, intervention or audit, and to the ethics of how data is used or treated. With reference to this Annette Markham brought our attention to the history of the term itself, which has been explored in existing work (e.g. Rosenburg 2013). A basic etymology of the term links the 1640s use of the term ‘data’ to the Latin datum (thing given), and dare (to give) (The Online Etymology Dictionary, accessed 10th April 2016). As Markham continued, when data is operationalized as a discrete object with obdurate properties that can be measured and transported, the notion remains categorically different from – and in a sense opposed to – the very idea of process (Markham, 2013). Over time, data has come to represent something incontrovertible (Rosenberg, 2013). These are the kinds of definitions a data ethnography approach can trouble. We can see examples of this in contemporary essays that critique the simplistic notions of data and Big Data (e.g., Wang, 2015) as well as research studies that provide nuanced accounts of how data is a part of materiality in everyday life. For example, Pink, Sumartojo, Lupton and Heyes LaBond, through their ethnographies of self-tracking cyclists, have sought to rethink the ontology of data through ethnographic analysis, to define it not as an ‘object’ but as a ‘thing’ that is part of a digital materiality (Pink, Ardevol and Lanzeni 2016) whereby things are leaky (Ingold 2008), open and always unfinished.

Ethnographic understandings seek to redefine data, or seek to emplace it according to narratives that contest those who perceive data as an object, or representation. Such alternative definitions open up the possibility to explore what happens when we deny (or suspend) the status of data as given and uncontestable, and instead focus on where it is made as imperfect, where it is damaged or when it is incomplete.

Yet, we need to be aware that there is not one single dominant approach to or definition of data that we should see data ethnographies as resisting or coming up against. As Marta Poblet pointed out, data is defined and treated very differently in different domains (e.g. information science, knowledge management, computer engineering, etc.). Even within disciplines there are multiple definitions of data (some using data as a synonym of information, some others considering data as the basic input for information, etc.). In information science, Zins (2007) documented up to 130 definitions of data, information, and knowledge formulated by 45 scholars. This may be a matter of disciplinary or design focus. For instance as Pilar Lacasa pointed out, it is likely that computer scientists don’t frequently think about whether we feel comfortable with our data, because they’re focusing on using it for other purpose. Or, in the case of lifelogging technology developers, as discussed by Fors, Berg and Pink (2016), developers might have a specific imaginary relating to users that is not based on actual everyday contexts of use as they would be presented if researched ethnographically. Data is also used very differently by ordinary people and corporations. We note that these differences matter, in that the affordances and characteristics of data are constructed by different stakeholders, some of whom have more voice than others. If data is perceived as prior to or separate from its interpretation, it gains a neutral status. This makes it more difficult to contest its power, as Edgar Gómez Cruz noted. This is where we need to seek for ethnographically and to contest, the ‘hidden politics’ that are part of this.

At the same time, we do not seek to describe a data ethnographies approach oppositionally, but to instead to see it as existing in the context of an ecology of different approaches to and understandings of data, each of which perform different tasks with data and have different outcomes. Some (or most) of the approaches to data we have discussed above, therefore conceptualise data as having a different ontological status and trajectory through the world to those attributed to it by a data ethnographies approach. This is one of the key differentiators between the kinds of investigations that an ethnographic approach can mobilise and the assumptions that underpin most existing approaches to data and to big data analytics. Such forms of difference might be useful or compatible at times, for some types of investigation. We might even think about when the relationship between data and truth becomes useful or relevant, and what it means in different circumstances to understand data as ‘true’, reliable or representative. These considerations become complex beyond the scope of our discussion in this paper, but for instance they lead us to probe questions like: what is the difference between data collected by self tracking and by drones? However as Shanti Sumartojo pointed out, this can never simply be a completely benign type of difference or relativity, since there will always be a politics to this, which will play out in multiple ways. One of they key things to maintain our awareness of is that the political agendas, power relations, understandings of society and human perception and activity in the world that are wielded by approaches that objectify data/ treat data as objective, or given, will most likely be different to and contested by the ethnographic approach we are calling for. Ethnography, we argue, needs to be mobilised into an equally prominent role.

This raises the question of if there is a persuasive role for ethnographic research in this context. Or should an ethnographic research agenda seek to take on a more activist and contesting role. That is to show that there is another dimension, not only to what data can and cannot tell us about people and society, but also to seek to refigure how data is conceptualised at all on an interdisciplinary stage, and to pull out what the ethical and political implications of this are. Such a role however will need to go beyond the typical way that anthropologists often use ethnographic findings to critically chip away at theoretical generalisations by citing the specific. Instead it needs to be a role that uses ethnography in dialogue with theory to build alternative arguments and visions for interdisciplinary collaborations concerning the roles data is playing in contemporary society and the roles it might play in how we plan and design futures.

Data and Ethics

Annette Markham, our invited guest for workshop 2, developed a conversation with us through a focus on the question of ethics and data. Markham’s work on Internet research ethics is very relevant to this field of discussion, and in part we drew on this. However we also connected these discussions to the ideas about impact, a future oriented ethics and the implications of this for how we think about data ethnographically.

The approach to ethics introduced by Markham, based on her experience of collaborating to develop the AOIR (Association of Internet Researchers) ethics framework, emphasised the use of questions rather than statements in guidelines about ethical practice and conduct. Emerging from an effort to enable scholars from different disciplines to reach consensus, this shift more broadly comments on how ethical guidelines often fail as much as they help since, as Markham pointed out, ethics are judgements, critical choices, and decisions in and for the specific moment. This shift toward ethics as choices we make allows us to think more clearly about impact. If we are to consider a future oriented ethics, then this involves taking responsibility of our choices, and considering what we want the future impact of these choices to be.

As part of the ethics of data ethnographies, we need to avoid falling into what has been referred to as the ‘datafication trap’. Datafication as defined by Kenneth Cukier and Viktor Mayor-Schoenberger (2013) refers to: ‘the ability to render into data many aspects of the world that have not been quantified before’. This is importantly ‘not the same as digitization, which takes analog content – books, films, photographs -and converts it into digital information, a sequence of ones and zeros that computers can read. Datafication is a far broader activity: taking all aspects of life and turn them into data’ (Cukier & Mayor-Schoenberger, 2013). It includes ‘behavioural metadata, such as those automatically derived from smartphones, like time stamps and GPS-inferred locations’ and may be used for a range of purposes ranging from surveillance to citizen empowerment (Kennedy, Poell and van Dijck 2015). However the process is not just a technical transformation which would involve separating various elements of experience into discrete units that can be transported and measured, but a conceptual one – of believing that such a thing is possible.

Our concern with the notion of datafication is that as it attempts to describe a certain state of affairs, as it occurs in one moment, it also flattens human experience, in a way that ethnography always defies, by acknowledging and insisting that whatever we label as ‘data’ is ‘rich’ and ‘lively’, rather than fixed, frozen or representing something ‘true’ about the world. Ethnography can enact data as something to be studied rather than something that is taken for granted. This small shift, Markham suggested, gives us means to co-opt the term datafication to stand for what ethnography can tell us about how data is part of our worlds, as long as ‘data’ is considered a ‘thing’ rather than self-evident or a priori in any situation.

This brings a new and exciting dimension to the way that we view ethics from a data ethnographies perspective. That is, if we take data for its thing-like affordances, to see it as open, malleable and leaky, part of the everyday world rather than as separate from it, the possibility of treating data as incontravertible is excluded. The ethics of objectification have been the topic of long established arguments in sociology and social anthropology. Take for example the history of colonial photography (eg Edwards 1992). When something is characterized by researchers as objectively or scientifically standing for something else, this is part of an objectifying practice, that seeks to make the ‘other’ stand still, to define her or him and to enter into a power relation of dominance and subjugation. A data ethnographies approach specifically avoids this by engaging with data through a processual ontology, to define data as a ‘thing’. Such an approach, in accord with Markham’s notion of ‘ethics as questions – which asks us to consider what the implications of our research choices might be in the future, offers us a way of thinking about data and ethics whereby our approaches to each are compatible with each other because it gives the same ontological status to ethics and to data (see also Markham’s 2015 post on similar issues here

An example that was raised referred to the similarities and differences between the way we are conceptualising data and the ways in which it is conceptualised in social media marketing. This example complicates the discussion because in some ways we are using the same definitions of what data is/could stand for – in that we are interested in, for instance, personal data collected by people using devices in their everyday lives, uploading this, and sharing it with known and unknown third parties. However this is where the agreement ends, since a data ethnographies approach understandings the meanings and uses of the data in different ways. As Markham pointed out it is important for us not to conflate these approaches.

Data Futures

The ethics of data ethnographies have, as we have already noted, implications for the kinds of impact our work might have in yet to be experienced or known futures. In our first workshop we discussed some of these issues in relation to health insurance companies, and how data might be used by corporations of these kinds. Here, in this workshop Yoko Akama drew our attention towards the ethics of the future orientation of design. Data, particularly that which is derived from huge conglomerate sources, is becoming increasingly a material or source for driving questions for and informing design practice. This is worrying in many ways, since as we have outlined here and in position paper 1, metrics derived through Big Data always represent a partial and non-representative sample (Baym, 2013) and thus do not accurately or adequately represent how people engage with and experience the world. While some might fall back on the positivist argument that we simply need to improve our measures, people –as everyday designers– will intentionally or unintentionally ensure that data is incomplete, dispersed and unfinished. As Pink discussed with reference to self-tracking in workshop 1, there are growing movements towards this, and more recently for example the unfitbits online initiative ( has come to our attention.

Yet what is perhaps more worrying is that designers, developers, and policy makers will continue to take Big Data at face value, as an object or representation of a truth that can be extracted from and that reflects society. Similar to Markham’s work in focusing our attention on how the terminology associated with data contains problematic assumptions, the Design+Ethnography+Futures Lab (at established by Yoko Akama and Sarah Pink precisely works towards disrupting these assumptions, by creating more generative forms of uncertainty with which to move on into the as yet unknown futures that Big Data analytics cannot predict. This work, we propose also needs to be undertaken regarding the assumptions that are embedded in contemporary uses of Big Data.

There is a significant role for a data ethnographies approach in prompting researchers to consider more fully the possible imaginaries for relationships between futures, design, ethnography, ethics, and data. There is definitely a need to bring together discussions around ethics and the future orientation of design with what this means for the ways that data is being or will be used in design practice and design futures research. Design might have a role in answering questions like that posed by Marta Poblet when she asked how we can protect ourselves as citizens from uses of our data – not simply as individuals, but as part of a community. Should we as academics be taking more of an activist stand in this field? Bringing together data ethnographies, design futures research and practice, and activism would be an exciting and productive combination.

Other ways to bring data into our notions of future, involve forms of making. For example, Markham discussed her Creating Future Memory project:, which invites us to consider how and where data will play a role in the ways we know and remember, as well as how we think about futures. That, is what role is data playing in how we understand temporalities, as researchers, designers and as ordinary people encountering data in our everyday lives? What role does data ethnography have in providing understandings of this, and how can such understandings enable us to better plan and design for ethical data futures? How, and to what extent can we, and do we want to use ethnography in ways that are accountable for these futures?

Summing up

This position paper builds on Position Paper 1 by extending our definition of what data ethnography means, and the discussion of the value it can bring to the role and meaning of data in our lives. Here, as in Position Paper 1, we have continued to explore how data is a dynamic ‘thing’ with affordances, relationalities and futures, rather than a flattened, objective, standardised series of facts.

Recognising this, we have called for consideration of the future implications of data ethnographic research by treating ethics as a series of questions about the potential impact of our activities, as well as those impacts implied by the activities of others who define data in ways different to us. In doing so, we have to a certain extent glossed over close definitions and debates about what ethics, data, design and futures might be, for the sake of pushing forward an agenda relating to the issues and questions that we believe need to be interrogated in order for us to move forward with a new ethnographically informed and implicated ethics for data futures. Further work is needed in the fine tuning of the questions to be asked and how they will be developed.

Nonetheless these questions this position paper has raised compel us to be accountable for the futures that our research might help bring into being, as well as requiring us to interrogate ethnographically the possible futures that can be envisaged through other framings and definitions of data. We need to ask, amongst other things, what data futures would be preferable, and what role can we play as ethnographers in creating future alterities that acknowledge the relationship of data to the messy everyday life contexts where it is made and used.

A final thought is how, given that we as academics are working in our own ‘datafied’ industry, might consider these issues in relation to shaping our own futures and those of colleagues who will follow us into similar roles.


Baym, N. (2013). Data not seen: The uses and shortcomings of social media metrics. First Monday, 18(10). doi:10.5210/fm.v18i10.4873

boyd, d. and K. Crawford (2012) CRITICAL QUESTIONS FOR BIG DATA, Information, Communication & Society, 15:5, 662-679, DOI: 10.1080/1369118X.2012.678878

Cukier and Mayer-Schoenberger (2013). The Rise of Big Data How It’s Changing the Way We Think About the World. Foreign Affairs.

Fors, V., M. Berg and S. Pink (2016) ‘Capturing the ordinary. Imagining the user in designing and using automatic photographic lifelogging technologies’ in S. Selke (ed) Lifelogging: Theoretical Approaches and Case Studies about Self-Tracking, Springer VS.

Markham, A. (2013). Undermining ‘data’: A critical examination of a core term in scientific inquiry. First Monday, 18(10). doi:10.5210/fm.v18i10.4868

Ingold, T. (2008). Anthropology is not ethnography. Radcliffe-Brown lecture in social anthropology. Proceedings of the British Academy, 154, 69–92.

Kennedy, H., T. Poell, and J. van Dijck (2015) ‘Data and agency’ Big Data & Society, 2 (2) DOI: 10.1177/2053951715621569

Pink, S., E. Ardevol and D. Lanzeni (2016) ‘Digital Materiality: configuring a field of anthropology/design?’ in S. Pink, E. Ardevol and D. Lanzeni (eds) Digital Materialities: anthropology and design. Oxford: Bloomsbury

Rosenburg, D. (2013) ‘Data after the fact’ in L. Gitelman (ed) Data is an Oxymoron. MIT Press.

Zins, C. (2007) ‘Conceptual approaches for defining data, information, and knowledge’. Journal of the American society for information science and technology, 58(4), 479-493.