By Rachel Thomson with Janet Boddy, Jette Kofoed, Niamh Moore & Sharon Webb
The Reanimating Data project has involved a great deal of reflection in relation to ethical issues and even some ‘carefull risk taking’. This is a term coined by Niamh Moore and colleagues (2021) to capture how methodological innovation is often associated with the need and appetite to delve deeply into research ethics and to make contributions to new ethical thinking. Our project has demanded that we extend or extrapolate conventional ethical thinking in order to cover the new terrains into which our methods took us. For the Re-Animating Data (RAD) project this included; working beyond the original consents that shaped the data in our archive; negotiating the ‘ambivalent gift’ of returning data to the community from which it was extracted; taking care-full risks to alert ourselves to the kinds of ethical sensitivities involved, including the right to not be reminded or to remember intimate conversations from the past; and developing modes of working as secondary analysts that held tensions between the need to be playful with data and respecting the integrity and significance of the material. We are busy writing a paper to capture this learning.
We are currently thinking about employing the reanimating model with another data set and decided that it was a good time to take advice from fellow travellers about how we might think creatively about research ethics in the future. As part of this work we were grateful to secure support from the Sussex Digital Humanities Lab to hold a workshop that staged a conversation between members of the Reanimating Data team (Niamh Moore, Rachel Thomson & Sharon Webb) with Jette Kofoed our visitor from Aarhus University, our ethics adviser Janet Boddy and other members of the Sussex research community, including the ethics research committee. In this blog we capture some of the key insights from the conversation and think through how we might work with these in the RAD project and beyond. Quotations from the panel discussion are marked in italics and are slightly edited for the purpose of readability. Each key insight from the conversation is marked by a heading followed by a number of quotations which are then commented on and responded to in plain text. We have not noted the individual speaker – considering that the fruits of this collective discussion are collaborative.
The ethics of remembering and forgetting
You raised the right to be forgotten. So, we’re having these conversations in some domains, but in the social sciences you still barely have the right to be remembered. In fact, you don’t. There’s also this kind of collision… you know these stories don’t belong to me [the researcher], and I certainly shouldn’t destroy them. You know they should be made available and be in the public realm.
Sociological interviews 20 years ago become historical artifacts. It’s like what happens in those, intervening years, it becomes like it kind of shifts disciplines. … for a historian it is like ‘what! you’re destroying data?’, but actually, understandably, from ethical perspective [you have agreements]. But in 50 years’ time that may be the only trace of these things.
People are way more conscious of data and rights around data, and where it goes, and what it does than they were 20 years ago, probably even 10 years ago….. [In research] We are giving people the opportunity to put what they want to say on the record. So [it’s] unlike all of our other kinds of data that’s being harvested, you know our horrible Amazon shopping list, all the things that we don’t want people to know, and we don’t want to be the historical trace of us.
Thinking about queer temporalities within that. So as life courses shift so too identity shifts. So, the right to be remembered in the way that you want to be remembered is important, and the right to be forgotten in the way that you want to be forgotten.
What I’ve learned from archaeologists in teaching with them at the ethics courses, is that a very old rule is that you never dig everything out. You leave something behind for the next generation, because you know that those coming after you might be able to do it more diligently, or to do it differently.
Working across disciplines can be generative for ethics. Norms and assumed ‘good practices’ in one field are challenged by norms and practices in another. Where the social sciences (which generally work with those who are alive) tend to lean on anonymity as a methodological norm and (uninspected) ethical good, historical approaches provide different perspectives. On the one hand, oral history traditions encourage us to think about ‘testimony’, the invitation to members of the public to become part of history and in doing so to use their own names. On the other hand, historians are alert to the very partial traces of the past that are left to us in the archive. Certain voices disappear. So not only can we make a strong case for the data of social science to become part of the public archive, but we also need to think carefully about whether anonymity is always an ethical good. This may be a question better discussed with participants rather than a decision made on their behalf. We may not always know or understand what counts as sensitive data – and what is sensitive may change over time as people enter different stages of life, and/or when they are no longer there to make decisions. What does it mean to delegate care of data? Can we imagine a prolonged or unending ethics – that involves returning to data or unresolved ethical challenges from the past.
Care, curation and infrastructures
.. my imagination of what an archive was […]: a box into which you, you know, post your data. Instead of: an archive is an infrastructure that you build that puts data in connection with other data and other things, and those standards, or what makes that possible.
What are the limits of our capacity to care for participant? So, for instance, we might retire or get made redundant from the university, and no longer have access to the archive that we saw ourselves as looking after. If we’re talking about long temporalities, the duty of care that we feel very strongly, can that be delegated? Can that be passed on? Can that be shared?
It has to be a constant conversation, and it relies upon infrastructures that are responsive to users, and which allows research participants to know exactly what’s out there about them, and enabling that kind of ability to kind of request takedowns or request changes or request.
We went for the most restrictive creative commons license, because we didn’t want commercial use. We didn’t want derivatives. We didn’t want, you know, sharing with attribution. And so, thinking about how those types of standards might benefit conversations about the kind of reuse imagined.
What does that traveling of data entail [for example data made available for secondary analysis by new researchers])? Does that mean that you can give it to me, and I can just violate it in in some other form or direction?
We cannot have that conversation [about AI and research ethics] outside of understanding technology and what challenges are being created for us to think about the human capacity to think and feel and to connect and to notice each other, and to be present with each other when we’re doing research.
Universities and Libraries play a vital role in assuring the preservation of data and archives enabling researchers to imagine the longevity of research data and to trust that our promises made in the present will be honoured long into the future. But what does it mean to care beyond the immediate ‘now’? In the Reanimating Data project we have worked with a dual approach to archiving the WRAP data set that assumes a long-term preservation approach using the institutional repository and a much more user-friendly and participatory approach that draws on the platform of Omeka. We work with CC BY NC SA creative common licenses to make our intention for the data clear to potential users – enabling sharing and adaptation but requiring attribution and restricting commercial uses. We also create metadata using the Dublin Core standard that enables us to link individual items in the archive into a cat’s cradle of digital relationalities that enable interested others to find the data. These infrastructures are vital and alongside the knowledge and theory of archivists, they enable us to leverage access to marginalised collections and to bring these into a truly ‘public sphere’. Yet this is not an automatic or automated process. It involves careful labour (the work of data preparation, creating metadata, maintaining platforms) and careful risk taking as we work out and work through the consequences of our decisions. It is productive to think about care beyond the immediate situations. New technologies necessarily pose new affordances which take time to understand while also learning of their pitfalls. We are beginning to grasp how digital archives can enable us to enliven research. The ethical labours of today can be written into the way that materials survive and the ways that future users experience them and allow us to scrutinize what care might mean in the future life of the data.
Democratising ethics
Participants are making decisions anyway, about what they tell us, and what they don’t.
We very rarely allow that our participants may have ethical frameworks themselves, and may have ways of working through ethical issues in their kind of everyday lives.
The thing that people were most concerned about in archiving was about representation, not just of them personally, but of their community and the potential for research to do ethical harm in relation to perpetuating stigmatization, political harm as well.
The problem of missing data turns into a really political project which is about: if you want to represent people who don’t trust research, then you’ve got to do it in a completely different way. And I think that’s really productive, because it’s not just by mistake that it is missing data.
What do we want to save? It is not just the funded research.
Ethics is utterly in the writing.
We tend to spend a great deal of time thinking about professional ethics and institutional procedures in order to identify and minimise risk. Yet, in an era that provides us with the technological tools for co-production, to distinguish between risk and harm and in doing so we need to recognise and engage with the ethical frameworks and insights of participants. We know that people deliberately absent themselves from what they perceive as potentially stigmatising knowledge projects. So, when starting new projects and building the archives of the future we need to think prospectively about inclusion and dignity. This is something captured by Rachel and Liam in their paper ‘starting with the archive’ (Thomson & Berriman 2023) where they mapped out stages in co-production of an archive (permission; negotiating the idea of ‘the public’; becoming data; publishing and compliance) as well as principles that could underpin all stages of this work (a sense of possibility; a commitment to co-production; understanding and promoting shareability; and securing posterity (a safe home) for the archive). When we do this kind of work carefully and collaboratively, we find out interesting things. For example, that people have their own language and insights about how they want their documentation to be handled by others. These kinds of consultations can also sensitise us around data sets from the past, when it is harder to find and involve the original participants. For example, in the Reanimating Data project the experience of the original research team encouraged us to not seek out original participants, understanding the potential burdens of being reminded of intense and sensitive memories which would outweigh the potential ethical goods associated with renewing consent. Such a decision places moral obligations on the researchers and archivists in shaping how materials are used, and how participants are represented or presented.
Not all research has the benefit of ethical review, or at least not all stages of the research are interrogated in this way. The ethics of writing, although perhaps the most sensitive of all ethical issues, generally falls outside of ethical procedure. Realising how the ethical practice is visible not only in consent forms or the paragraphs describing the ‘ethics’, but also how ethics is in all the details of how we present the settings, the holding of bodies, affects and ways of engaging. We want to discuss in much more detail the ethics of writing.
Ethical muscle: fallibility and tensions between procedural and relational ethics
Credible researchers go back and, you know, sometimes they unmake their previous decisions.
Are these researchers ethically competent? Are they demonstrating their capacity to think [..]. are they saying I’m a trustworthy person who knows what to do when I don’t know what to do.
‘Having’ ethics is such a weird, such a weird thing to say, isn’t it? I have ethics meaning: I have the approval. The institution has said that I’m good.
We can see a future where you get brought into a project to write a really good ethics application.
It would be an interesting exercise for us to ask ChatGPT to do an ethics application for a fantasized project.
[Take an ethical scenario], and know how to unpick it, work your way through it. [The aim] is not to answer it. It’s not about being right, is it? But having be having the capacity to think.
I like this idea of sticking with the messiness. You know, allowing that kind of complexity and messiness to kind of be part of the research journey.
Part of our conversation focused on the importance of preparing researchers to become ethically competent. This of course means understanding procedural ethics and having the confidence to work with an ethics committee. Ideally this takes place in stages, seeking ethical approval as a study unfolds and having the confidence to work iteratively in collaboration with critical friends. Ideally, we would seek ethical review for each set of methods as they arise in a longer project. But ethical competence also involves having time, space and supervision so that ethically complex situations and examples can be explored not merely for the purpose of achieving approval of an ethics application. This involves working through the emotional responses of researchers and participants to different parts of the study. Jette’s (2015) account of ethical hesitancy was embraced as a way of maintaining a research position when working with sensitive topics that give rise to pressures towards ’helping’ and ‘solving’ the problems encountered. Hesitancy can support researchers to be able to operate in the present tense without collapsing into safeguarding reactions – although a good understanding of safeguarding is a crucial underpinning of working in an ethical way. Jette shared examples from a Danish context of extended courses for doctoral researchers where they are invited to talk through ethical challenges (and joys) that they experience and to explore how to implicate themselves in their data. The aim of this training is not simply for them to become compliant in relation to ethical procedures but to build the capacity to think and act ethically and to build ‘ethical muscles’ collaboratively with peers in safe spaces. Allowing fallibility to be part of ethics is not an invitation to carelessness, or to not care about making mistakes. Rather it is a manner of acknowledging the careful risk-taking as part and parcel of all research processes. ‘Having’ ethics, we find, seems to be a new lingua developed through ethics applications and ethics assessments where once ‘approved’, you ‘have ethics’. Such possessive ethics must be challenged by a vivid language of ethics, involving terminologies such as careful risk-taking, hesitancy, prolonged ethics, democratizing ethics, giving back, ethics of writing, ethics of leaving traces behind. Questions of supervision and peer support are a vital part of such terminology and of ethically robust research – as well as the ability to slow a project down so that the granular and iterative character of the research process becomes evident. A willingness to learn, implicate ourselves, communicate and to change our minds are characteristics of healthy relational ethics.
These thoughts, ideas and values are incredibly useful as we begin to think about developing a pilot stage for our new Working Mothers project. We also generated ideas and insights that are valuable for Ethics Committees and ensuring that ethic committees continue to learn and work in conjunction with a research culture that promotes learning in relational ethics.
References
Kofoed, J., & Staunæs, D. (2015). Hesitancy as ethics. Reconceptualizing Educational Research Methodology, 6(2). https://doi.org/10.7577/rerm.1559
Moore, N., Dunne, N., Karels, M., & Hanlon, M. (2021). Towards an Inventive Ethics of Carefull Risk: Unsettling Research Through DIY Academic Archiving. Australian Feminist Studies, 36(108), 180–199. https://doi.org/10.1080/08164649.2021.2018991
Thomson, R., & Berriman, L. (2023). Starting with the archive: principles for prospective collaborative research. Qualitative Research, 23(2), 234-251. https://doi.org/10.1177/14687941211023037