Beyond the Field: Ethics after Fieldwork in Politically Dynamic Contexts

Abstract

As researchers, when do our ethical obligations end? How should our ethical obligations respond to dynamic and unstable political contexts? Political scientists frequently work in dynamic political situations that can pose new ethical questions beyond those existing at the point of fieldwork. Yet, research ethics are often conceived in terms of a static, if not hermetically sealed, field site that remains frozen in time at the point we conduct fieldwork and collect data. I argue, first, that we need to consider more systematically how a dynamic field intersects with ethical obligations. Second, I argue that new and unexpected ethical questions can emerge after exiting the field, including responsibilities to research participants, dissemination, and publication, and returning to the field, which should be a part of how we conceive of ethical obligations.

The best laid schemes o’ Mice an’ Men,

Gang aft agley,

An’ lea’e us nought but grief an’ pain,

For promis’d joy!

—Robert Burns, “To a Mouse” (1785)

Political scientists frequently work in dynamic contexts which can change dramatically after exiting the field. This dynamism can pose new and unexpected ethical questions outside of those we conceived before and during field research. When it comes to ethics, the dynamic nature of research contexts and the ethical questions this dynamism can pose beyond the field are often overlooked or sidelined in favour of ethical questions that arise within the field, as a site beyond our institutions and desks where we conduct research and gather data.¹ In this article, I argue that questions of ethics frequently arise beyond the field and form a crucial dimension of our overall ethical obligations.² In other words, we have both a duty to think about and to address the questions of ethics that arise beyond the field.

In considering ethics beyond the field, I do not intend that researchers should be encouraged, required, or incentivized to be clairvoyants. Our role is not to be able to imagine, or predict, what changes may or may not occur after we have left the field. Rather, our task is to transform how we think of field research sites, as contingent contexts that might change. In turn, we should conceive of questions of ethics as an ongoing consideration—and obligation—that does not end when we leave the field.

My motivation here comes from my research in Crimea, where I conducted field research on Russian identification and engagement with Russia between 2012 and 2013. In a matter of three weeks in early 2014, Crimea shifted from an uncontested Autonomous Republic, within Ukraine, to an internationally disputed de facto republic within the Russian Federation. With these political-territorial changes, the lives of those who had participated in my research in Crimea were changed dramatically and—most likely—forever.

While Crimea’s annexation is a rare event, political scientists frequently work in dynamic and newly contested sites, such as the Middle East and North Africa (MENA) in the wake of the Arab Spring. While we might expect dynamism and contestation in authoritarian and quasi-authoritarian regimes such as Crimea, I cover three issues with implications that include democratic contexts where states like the United States and the UK can use research data for pernicious ends.³ First are the ethical questions of engaging with research participants beyond the field who reside in dynamic and contested contexts: What ethical questions should be considered when we maintain contact with participants beyond the field? Second are the ethical questions of disseminating research about dynamic and contested contexts: What questions do we need to ask about what we publish and when? Third, I discuss the ethical questions arising from choosing to return (or not) to the field. Finally, I reflect on how questions of ethics beyond the field intersect with policies that advocate for norms of data access and research transparency (e.g., DA-RT).

We Need to Talk About Ethics

The 1978 Belmont Report set out four guiding ethical guidelines that form the backbone of institutional ethics review (IRBs in the United States) and underpin the everyday practice of ethics in the field.⁴ Most prominent has been the consequentialist principle of reducing the harmful consequences of research for participants and maximising the benefits of research for those who participate in research (table 1).⁵ The Belmont report also requires researchers to recognize the deontological principles of research: the rights of participants to be autonomous consenting participants and to be treated with respect and dignity in the process of research.⁶

Table 1 Ethical principles and guidelines stemming from the Belmont Report

Political science, as a discipline and profession, had remained relatively silent on research ethics.⁷ More recently, in response to methodological innovations like experiments and the recognition that political scientists often conduct research in dangerous contexts, discussions of ethics have become more prevalent. Debates in political science, as well as across the social sciences, have also concerned the appropriateness of applying the Belmont criteria through institutional ethics review because of the biomedical origins of these criteria which some argue are inappropriate for social science.⁸ I am less concerned less with the relevance of IRBs and more concerned with how the Belmont criteria are operationalised in the process of research. In particular, I draw attention to operationalising the Belmont criteria in particular for questions of ethics beyond the field, as much as within the field. For example, for harm, we need ways of conceptualising and operationalising harm, in terms of its form (e.g., physical, material, psychological, political, reputational), degree and locus (e.g., research participants, researchers, gatekeepers and research assistants, and the wider context in which the research is conducted). These dimensions of harm are challenging to consider before and during field research. They can be even more challenging when we consider ethics beyond the field in unpredictable, dynamic, and contested contexts.

In the following section, I review debates within the two prominent political science literatures—ethics of experiments and “dangerous” field sites—to consider how each has addressed issues of ethical questions such as conceptualising consent and harm. In particular, I note how each set of debates has discussed, to some extent, ethical questions beyond the field in passing but have not to date focused on these questions directly.

The Ethics of Field Experiments

With the explosion of field experiments in political science in diverse and complex contexts, ethical concerns about the conduct of experiments have garnered attention and concern, particularly since the Montana get-out-the-vote scandal (GOTV) in 2014.⁹ In terms of assessing the locus of harm, research participants in field experiments are usually unaware they are participating and subject to interventions in their lives. This can be crucial to the methodology of field experiments that are designed to avoid observation bias caused by participants knowing their behaviour is being observed (the Hawthorne effect).

As Teele argues, this is often (wrongly) presented as an ethical dilemma: between a question of methodological clarity and ethical responsibility. However, Teele argues, this is not an ethical dilemma so much as an ethical sacrifice, where we signal that methodology has priority over ethics.¹⁰ This, at least implicitly, has been the rebuttal of some scholars who argue that knowledge gained from experiments has a normative value.¹¹ For example, some have argued that the kinds of knowledge gained are on critical topics (e.g., election behaviour, conflict) and might be hard to acquire from less interventionist methods, even if this knowledge is gained by causing potential harm, such as by manipulating democratic outcomes.¹²

These ethical sacrifices, in the name of data, have consequences on the context in which the research is conducted. These effects could include long-lasting effects downstream—because experiments can seek to alter the field after the experiment has ended. This is problematic because it is typically research participants, rather than social contexts, that are the focus of institutional ethics review. As Michelson explains, while research participants were the focus of academic debate over the Montana GOTV scandal, there is also the ethical question of whether the experiment influenced election results, as was raised by officials and politicians in Montana.¹³

More broadly, experiments (in particular those which might be seen as intervening in elections) can decrease the legitimacy and credibility of research and the academic profession. This might, in certain circumstances, increase wider social distrust in political systems.¹⁴ In turn, scholars may find it difficult to research in that context or similar contexts because of unwilling participants. Experimental methods can also provoke or extenuate a wider public distrust, from medicine to politics. For example, it is widely reported that with the coming of the AIDS epidemic in the United States, the Tuskegee experiment had a lasting impact in terms of black communities’ mistrust of (predominantly white) government medical programs.¹⁵ Assessing harm, in terms of its forms, degree, and loci in experimental methods, is, therefore, a complicated endeavor that extends beyond the duration of the experiment.

The Ethics of “Dangerous Fieldwork”

The second domain in which discussions of ethics in political science predominate concerns the ethical questions arising from doing “dangerous fieldwork” in sites of recent, ongoing or pertained conflict.¹⁶ Conducting “dangerous fieldwork” involves a complex locus of harm to assess: researchers and research participants can be exposed to dangerous degrees of psychological and physical harm. Participants add to this challenge by spanning a complex spectrum of violence, from victims through combatants to perpetrators, whose competing perspectives and interests the researcher has to balance.¹⁷ For example, those perpetrating violence can also be especially willing to engage with researchers about life histories.¹⁸ The researcher has to navigate ways of minimizing the kinds of influence on the research that these different interests might seek to impose.¹⁹

Debates within dangerous fieldwork also require us to consider how researchers, as well as gatekeepers and research assistants, are a significant though often overlooked potential locus of harm. As Cramer et al. note, researchers can often overlook the harm they may themselves be exposed to. This can be for several reasons, including empathy for research participants and an institutional review process that emphasises concern for participants (over researchers).²⁰ There can also be a cachet attached to dangerous research, and a fetishisation of danger, in particular among graduate students who are incentivized to do more to stand out from their peers.

Assessing and limiting harm in dangerous fieldwork relies on the researcher’s “informed moral judgment” about the context in which they are working.²¹ Such judgments can be affected and impeded by the research process, including the researcher’s exposure to trauma that may make them more likely to expose participants to higher degrees of harm.²² Calculating harm includes considering whether the research should be conducted in the first place. It also includes deciding what materials are published and when, in particular when conflicts are ongoing and participants in the field—and in the wider social context—remain at risk from the dissemination of research materials.²³ A more sceptical perspective, proposed by the anthropologist Fluehr-Lobban, suggests that it can be hard “or impossible” to assess, avoid or prevent harm that research participants (and researchers) might be exposed to because we cannot know or predict what will happen within the context of the field.²⁴

This points to the importance of considering ethics beyond the field since we publish material after exiting the field. Our research has the potential to affect the field after we have left, while the field has the potential to change in significant ways, in terms of the level of violence and who holds power.²⁵ Our evaluation of the harmful impact of publishing can be based on our knowledge of the field and our relationships with research participants. As I discuss later, these evaluations become more challenging once we have left the field, and in particular when the field is dynamic and a newly contested context. We have to reconfigure our knowledge of the field against these changes while being beyond the field, where it can be difficult to gain the necessary knowledge to make such judgments.

Ethics and the Banality of a Post-Soviet Research Context

Before turning to questions of ethics beyond the field, I want to first reflect on the institutional and research context of my study in Crimea. Compared to the United States, where institutional ethics oversight via IRBs is federally mandated, there is greater variation in the UK between institutions in terms of the procedures and extent of ethics review.²⁶ For example, the institution where I conducted my graduate training was, at the time, relatively laissez-faire when it came to institutional oversight and ethical review that, beyond the supervisor, involved little institutional involvement or oversight. The instances which would have required institutional ethical review would have been those involving “vulnerable” participants, such as minors. This is not the case across all UK institutions. For example, my experience of another UK institution with a large medical school is that all research conducted with human subjects must be reviewed by an ethics review board.

A further important institutional context concerns graduate training, where we had little in the way of discussion of research ethics beyond supervisory support. This is partly explained by how interviews in political science are conceived: as a method that extracts knowledge from elite informants as knowledge-holding actors, typically policymakers.²⁷ Here, ethical questions primarily concern issues of consent, anonymity, and what is “on” or “off” the record.²⁸ By contrast, political scientists with an ethnographic sensibility conceive of interviews differently: as an evolving working relationship between researchers and research participants.²⁹ Researchers use interviews to gain insight into multiple different perspectives, meanings, and lived experiences from a diversity of participants.³⁰ While this is not the case for all programs, within methods training classes in political science that emphasise elite interviews there can be little scope to discuss the ethical questions of entering the everyday lives and realities, including living in people’s homes.

At the time of field research in 2012 and 2013, Crimea was an uncontested region of Ukraine and a fairly banal post-Soviet context.³¹ Ukraine was neither fully democratic nor autocratic. It was not a site of conflict or potential violence. Yet it was systemically corrupt, in the throes of a corrupt and unpopular Yanukovych government. My greatest concern in Crimea was avoiding the gaze of, and connections with the state, especially the security services. This concern is typical of post-Soviet researchers, and of those working in quasi-authoritarian and authoritarian regimes more broadly, who have to contend with high degrees of distrust of the state.³² Aside from jokes that I was a “spy”, I was often faced with justifiable questions concerning what had brought me to Crimea and who was funding me. As best I could, I ensured participants of my independence as a researcher and of my personal motivation to study the topic. I was also clear how the research was funded by a UK research council, even if this proved challenging to explain.

My second concern was avoiding signed consent forms, which can provide an overly official and bureaucratic atmosphere to research and undermine the idea that research is voluntary.³³ Using signed consent forms (as often required by IRBs) can sometimes heighten the potential harm to which participants are exposed, by making confidentiality harder. These concerns are typical of researchers working in quasi-authoritarian and authoritarian regimes, including post-Soviet space and the MENA region in the wake of the Arab Spring, and of those working with vulnerable populations, such as sex workers.³⁴ Instead, I sought oral consent, gained by explaining the purpose of the research to the participants (as best as I realistically could), by providing an information sheet and discussing the research with participants, in terms of who I was and the purpose of the research.

Beyond this, Crimea was an accessible site to conduct field research in 2012 and 2013. Individuals were easy to find (e.g., via social media and existing contacts) and were willing to meet and discuss their experiences and, often, to act as gatekeepers for accessing further participants. Crimea was also accessible because little bureaucracy was involved. For example, I needed neither a visa nor research permit to conduct research in Ukraine and could travel visa-free to Crimea, by land or air, for up to 90 days. By contrast in more autocratic, post-Soviet field sites, such as Russia, researchers are required to acquire a researcher visa backed by an invitation from a local research institution.³⁵ This created a bureaucratic trail that might make some research more feasible than others. Today I could not access Crimea so easily and with the same relative invisibility. Now a de facto part of Russia, Crimea is accessible only via Russian research visas. Meanwhile, visiting Crimea from Russia is considered illegal under Ukrainian law making it hard to travel to Ukraine afterwards. It is possible that the only researchers, going forward, who will be able to conduct field research in Crimea will be those with Russian citizenship.

The greatest ethical issue facing my research has been how Crimea has changed since 2013. At the time, perhaps naïvely, Crimea’s status seemed so uncontested as to render these questions of Russian identification and engagement as banal, rather than an issue of contemporary contestation. Thus, what I (and the majority of the participants) had neither conceived of, nor predicted, were the events that would take place in the months after, that would take Crimea from a banal to a contested context in a matter of weeks, as the first case of territorial annexation in Europe since 1945.³⁶

While Crimea might be an extreme case, it reflects a broader set of concerns regarding the research that political scientists frequently conduct in dynamic and contested contexts. In the context of the Arab Spring, researchers who have or are conducting field research in the MENA region face both new objects of analysis, from protests to repression, and new (and heightened) ethical challenges to participants and themselves.³⁷ In my graduate cohort also, colleagues faced similar challenges where the field changed during and after fieldwork, from research on minority groups in eastern Turkey to the practices of the state in western DRC. Changing field research sites, in other words, are likely a common occurrence. The point is that we have not been framing our ethical obligations with this understanding of the likeliness of, but at the same time unpredictable, change to the cases and contexts in which we work. At the same time, my contention that we have ethical obligations extends beyond contexts that become contested once researchers have left the field.

Ethics Beyond the Field

In the following sections, I discuss three ethical issues that arise beyond the field: engaging with those in the field, publishing about the field, and returning to the field. These issues have all arisen in the process of working with Crimea as a dynamic and newly contested site, after leaving the field. However, these issues face many field researchers, whether they work in newly contested sites, like post-Arab Spring MENA countries such as Egypt, dangerous contexts of ongoing or recent conflict, and across the spectrum from contexts that are authoritarian, quasi-authoritarian, or liberal democracies. In other words, dynamic and newly contested contexts provide a most likely and salient context for having to address these ethical questions beyond the field. Yet our ethical obligations extend beyond the field regardless of the kinds of research political scientists do, in the way they engage with and represent the field after the period of data collection has ended.

In focusing on these three issues—of engaging, publishing, and returning—I return to the Belmont criteria to address how these questions intersect with harm minimisation, beneficence and the rights of participants to justice and autonomy (table 2). Using this framework moves beyond a focus just on the principle of “do no harm”. For example, I consider how the assessment of harm intersects with the need to “give back” to those research participants and assistants (by returning and sharing data) and with the rights we provide to participants and assistants to choose to participate in the research and their rights to be treated with dignity, respect and privacy in how we represent and engage with the field while being beyond the field.

Table 2 Applying Belmont criteria to ethical questions arising beyond the field

Engaging Beyond the Field

Field research, particularly immersive qualitative methods such as ethnography, interviews, and focus groups, relies on establishing a network of contacts, gatekeepers, research assistants and research participants in the field research site.³⁸ Interview-based field research requires developing working relationships with participants in the field through whom we observe the dynamics and inner logics of what we are studying.³⁹ As Fujii highlights, these working relationships are critical for ethics, too, by providing the necessary space for the dignity and respect between researcher and participant, and which is more essential than establishing trust and rapport.⁴⁰

These relations may be asymmetrical: we may enter the personal lives of our participants, but our participants are less likely to enter our personal lives. Such relationships may not be “real” relationships, like between friends or colleagues, but are researcher-participant relationships built within the field.⁴¹ These relationships, thus, have a certain power relationship that may be “reciprocal, asymmetrical or exploitative.”⁴² We also hang around with research participants, to be a part of their everyday life. Often, we can become friends with research participants, from the simple act of becoming a participant’s friend on Facebook, to establishing a relationship that can be crucial to immersive data collection, and to feel more at home in the field.

We use our participants, as well as local research assistants, to learn about the field. We often use research participants and assistants as gatekeepers, to seek out other research participants, and to provide access to events, such as protests, meetings, and commemorations that can be crucial for data collection and contextualization. In Crimea, research participants informed me of what was going on where, on where to go and how to get access. Through research participants, and the organisations they were part of, I was able to see the various expressions of Russian identification, such as the invite-only Russian cultural festival held every year. I was able to observe who was attending such events (primarily pensioners), what they were wearing (from military uniforms to the Russian tricolour), and how they were behaving.

Our participants may also use us, as political resources, within the social context of the field.⁴³ In this sense, we have little control over the participants. We cannot control how participants use us within their webs of relations. Just as we cannot control the dynamism and unpredictability of the field, we cannot control who participants in our research become or what they do, for example, regarding their careers after the field research has ended.⁴⁴

In turn, as a kind of research that is predicated on relationships, our ethical obligations to the participants, gatekeepers, and research assistants require us to be able to assess and minimize the potential form and degree of harm to which they might be exposed. In the field, harm assessment and minimization concerns ensure that participants are informed about potential harm and risk, and consent to participate on this basis. Researchers are also required to inform and discuss with participants about the nature of anonymity and confidentiality, and about data exposure so that participants can make informed autonomous decisions about the degree of harm and risk they might be exposed to.

Our relations, however, extend beyond the field, as we maintain contact with those in the field or might seek to conduct long-distance strategies of research. In turn, our ethical obligations to these participants and assistants do not cease when we leave the geographical site of the field. For example, we have a responsibility to ensure that participants’ data remains secure for as long as we retain it. In many instances, it can also be useful to maintain research relationships, to maintain contact, and to be engaged in the field in ways that are useful for data collection, analysis, and reflection. Social media makes this increasingly feasible, with the ability to observe in real time how developments are affecting research participants and how they are articulating themselves vis-à-vis such developments. If we maintain these relationships beyond the field, then we maintain ethical obligations. Even if we do not maintain these relationships, we maintain ethical obligations to participants (and gatekeepers and assistants) to treat them with dignity, respect, and privacy, for example in how we talk and publish about the field (which I discuss next).

We typically conceive of the period of informed consent as concerning the period in the field. This logic assumes the field stays relatively constant once we exit. This is not an assumption we can make as researchers who work in dynamic and unpredictable contexts, which can shift rapidly in the degree and form of contestation. There might be instances when we should re-seek consent, or address participants’ willingness to participate in research and to be published about, if we plan to use the data for new purposes or unforeseen outlets. At the same time, we may not be able to contact participants to gain such consent, in particular if we are able to return to the field as I will discuss later. This means we might need to have a more realistic engagement with the kinds of assurances we can offer participants about how their data will be used and thus seek initial consent on this basis.

Further, it might be ideal to call on participants when the field undergoes a period of rapid and significant change so that they can be a window into the field from beyond the field. At the same time, this poses questions of ethics that go beyond our focus on ethical obligations within the time and space of field research, and that require us to reconsider the harm to which we are exposing those involved in our research and the nature of consent we acquired from participants.⁴⁵ It might be also be ideal from a methodology perspective, also, to call on research assistants to continue to gather data when or if we cannot access the field. Of course, this poses forms and degrees of harm that are hard to conceive because of the uncertainty of the context, and our distance from it. In both cases, calling on participants and assistants might expose them to forms of harm, which may be hard to gauge and evaluate not least because distance may render these forms of harm less visible or even invisible to the researcher. This is not to suggest that it is necessarily possible to foresee all the forms and degrees of harm to which participants, gatekeepers, assistants, and researchers might be exposed when in the field, but more to emphasise how this becomes much harder from beyond the field.

With Russia’s annexation of Crimea, initially, I attempted to engage with research participants, given that it occurred just six months after I had left the field. Before long, I no longer felt comfortable engaging with participants about annexation. I was disconnected from the field, and this field had shifted dramatically, politically and geopolitically. I was no longer able to sense potential harm to participants and to contextualize different perspectives on events. So removed, I felt uncomfortable about prying when the day-to-day lives of those I knew had become so insecure.⁴⁶ I could observe some participants remaining in Crimea, some leaving; I could observe some participants supporting annexation while others opposed annexation. In other words, it would have been opportune to collect data—to observe beyond the field how my participants experienced annexation and how they positioned themselves vis-à-vis annexation. Without the ability to understand the field, however, I no longer felt able to make judgments that were crucial to minimizing the potential degree and form of harm to participants in ways that became difficult to assess beyond the field, particularly those that have changed so dramatically.

Publishing Beyond the Field

The ethics of what we publish is central a question of ethics beyond the field. We make choices about how we use the material, what we write, and when we seek to publish material in journals, and increasingly online (e.g. draft conference papers, blog articles).⁴⁷ At the same time, we have professional incentives that promote an ethos of publish and perish; such an ethos might be in tension with the ethics of what should be published to minimize the potential for harm to and to ensure the dignity and respect of research participants. What is published, and when, is thus integral to the question of ethics.

It is imperative to think about the potential for harm to be caused, to research participants, assistants and gatekeepers and the wider context, by what material is published.⁴⁸ In dangerous research contexts, we are attuned to the notion that we need to think ethically about dissemination and publication, where the forms of potential harm are serious, such as material and physical harm to research participants and assistants, and political and physical harm to those in the wider social context.

Dynamic and contested, but not necessarily dangerous, contexts can also pose challenging and serious ethical question concerning what can be published beyond the field. Dissemination strategies must consider and take account of the unpredictability of the field in how we approach the use of the data. In simple terms, researchers should abide by standards of privacy, by upholding anonymity and confidentiality, to minimize the potential forms of harm and protect the rights of participants to dignity and privacy. However, anonymity is not an end in itself. In small research contexts, it might be clear to others in the social contexts who participants are even if they are anonymized. More broadly, then, researchers can adopt other strategies, such as using a layer of “protective abstraction” from the original material.⁴⁹ For example, researchers can discuss data in a more aggregate way, in terms of common themes in the research, or taking research away from specific identification characteristics of participants or the social context, toward the more abstract.⁵⁰ The challenge is to match this against the desire to make a credible argument and to navigate regulatory frameworks of transparency, as I discuss later.

It is also challenging to assess the forms, degree and locus of harm, because of the context’s unpredictability. For example, as discussed earlier, the dynamics of the field can alter the nature of data and the relationship between research participants and the field significantly and, potentially, in ways that increase potential harm to these participants. In turn, this can affect the meaning, salience, and sensitivity of the data that we have or may seek to publish in ways that might be harmful to research participants and the wider social context. For example, Brewer advocates a strategy in dangerous research contexts of avoiding acquiring “guilty knowledge”, and thus avoiding having to deal with this knowledge.⁵¹ However, this assumes an unchanging research context, as if the kind of knowledge that is “guilty” is also unchanging. In dynamic and contested research contexts, such an assumption is problematic. As I found in Crimea, it is hard to predict what might be salient or guilty at a particular point. Rather, this can change in response to the changing research context.

Further, as Fujii explains, authors lose control of their material and its potential impact once it is published.⁵² This I have experienced first-hand, not through academic work or engagement because this is less accessible to participants in the field and individuals outside of the field, but through the faster-paced world of writing blog articles. On the one hand, writing blog articles allows researchers to transform academic knowledge into knowledge that is more widely accessible and meaningful and to communicate with a larger audience, and back to research participants and assistants, as part of the “giving back” of research to those who help to produce it. On the other hand, writing blog articles exposes us to this larger audience. We cannot control how our material is read, interpreted, and politicized by this larger audience.

For example, in December 2015, I published a piece in the Monkey Cage blog, where I argue that in Crimea there was a plurality of meanings of being Russian and a lack of support for a change away from the then status quo. In other words, I argue there was little support among those I interviewed for the kinds of reality that emerged in 2014 with Russia’s annexation of the peninsula. Within a day, the article was translated into Russian and posted on a Russian news website (InoSmi). Within a few weeks, the Russian Ministry of Defence’s TV station, Zvezda, created a 60-minute documentary with Russian commentators and politicians discussing and critiquing the research. The documentary used video clips from YouTube, sharing my face and details about my professional life, such as my educational background. This made me feel uncomfortable and visible in a way that I had neither anticipated nor consented to. I had lost control of the research material. It was a personal choice to publish work that was critical of the Russian government (but not critical of my participants). Some researchers adopt deliberate strategies not to criticise the government in states in which they conduct research, to protect their participants, their work with research collaborators, and their access to the field.⁵³ My goal was to critique the way that Crimea’s annexation was framed as inevitable by journalists in the west; this made criticism of the Russian government hard to avoid. Of greater concern of this new visibility was the potential impact on the social context of the field. The material was anonymized and told an aggregate-level analysis. It did not identify participants, but this was not the point. Even if participants could not be identified from the research, they had not consented to the material being used in a way that was so visible and politicized. The material was being used as a tool by a federal government, now an occupying power, to discredit the research and its methodology.⁵⁴ This example demonstrates the difficulty of knowing the harm that one’s research can do: I do not think anyone was directly harmed by the data being used in this way because individuals cannot be individually identified. Yet it is in part contributing to the securitisation of an issue that became politicised in unforeseen ways beyond the field.

For some, this demonstrates the “impact” of research when our data is used by organizations and authorities beyond academia. My intention is not to advocate for a position that is for or against research that has broader non-academic impacts. However, at least in the UK, researchers are increasingly incentivised to (and often required by funding bodies) to promote the impact of their research.⁵⁵ Rather, my intention is to highlight that the impact of our research, which clearly takes place beyond the field, also has ethical implications. This includes harm to our participants, their social context and researcher, and questions about how we use data in ways that respect the rights of participants to privacy, to respect, and to consent to their data being used for these ends. Meanwhile, the operationalisation of impact does not include beneficence to research participants or their social context, but broader policy and societal impacts for the countries where we produce research.

I wish that I had thought more about how I would disseminate research, and what the potential implications of dissemination might be. There is a brave naïvety that comes with being a graduate student and a novice seeking to publish about the field.⁵⁶ It is the responsibility of graduate programs to have open and realistic discussions with students in methods and field research training, in a way that is (hopefully) methodologically pluralistic to the variety of strategies of field research, and open to thinking about ethics in practice, beyond institutional review, by training and requiring researchers to be reflexive. I do not suggest that researchers become clairvoyants who can anticipate, expect, or imagine they could foresee all potential future events. Rather, we should to some extent be more forward thinking about the potential forms of instability in the research context, and how this might affect what we can publish about the field.⁵⁷ We also need to be more reflexive, as a research community, about the contingency of the field to recognize the impacts this can have on the participants of research, the data collected, analyzed, and published, and on us as researchers, with personal and professional lives.

Returning from Beyond the Field

When I left the field, in June 2013, I had every intention of returning to Crimea. I had invested in Crimea, professionally and personally, building contacts, establishing working relationships, making friends, and acquiring knowledge. I had begun to think of new projects that I wanted to conduct in Crimea. Finally, I wanted to share the data that I had collected, analyzed, and published on with those I had met and spoken with, who were crucial to the research process and development of knowledge that challenged key pre-existing assumptions about Crimea.

Following Crimea’s annexation, it became extremely unlikely that I will return to Crimea, so long as it remains under the de facto governance of Russia, because of how it has shifted politically and geopolitically.⁵⁸ That I would not travel to Crimea to conduct future research is often surprising: would it not be fascinating to conduct research now Crimea had been annexed? There might even be professional incentives to gather such data since I have a body of contacts and relationships on the ground. I leave this “moral judgement” for another researcher.⁵⁹ For now, most academics I know that are conducting field research in Crimea are Russian citizens because of the relative protection this provides Russian citizens, as opposed to foreign citizens, from the gaze of the state.⁶⁰

Returning to the field is a crucial part of field research providing a return on investment, concerning the time, energy, and resources the researcher commits. More substantively, returning to the field enables researchers to give back to the field, because without research participants, there would be no research. This sense of reciprocity has been a key tenet of much anthropological and ethnographic work who see giving back as part of the principle of beneficence, enabling the research to have some sense of benefit to those who participate and make the research.⁶¹ Returning reinforces the sense of respect, commitment and reciprocity to the community/communities that are integral to the research.⁶² As the geographer and ethnographer Karen E. Till argues, returning to the field helps researchers to improve their depth of understanding of the “of the peoples (including the researcher), places and institutions studied”,⁶³ and to reconcile and work through the “conflicts, tensions, differences and gaps” that arise and provoke further questions during the research process.⁶⁴ Fulfilling this kind of ethical contract of reciprocity, between the researcher and participants, by returning to the field is therefore optimal.⁶⁵ However, this is not without challenges. For example, Ellis describes how her participants became unhappy and shut her out of the field after a relationship of seventeen years because of how they interpreted her writing on the field. Writing about sexual practices within the social context, her writing about the social context became personalized by participants as if she was writing directly about specific research participants in the way she used their stories.⁶⁶

This means we need to think about the ethical questions of returning to dynamic and newly contested sites like Crimea. In terms of harm, it is difficult to assess and minimize the degree and forms of harm that past and potential participants might be subjected to, as well potential harm to me as a researcher. For example, the scope of what is politically salient has expanded where the everyday banality of being Russian in Crimea is no longer banal. Talking about Crimea leaving Ukraine is no longer an unlikely event, but an event that has occurred. If Crimea had been a dangerous context when I had conducted research, then perhaps it would be easier to return, and easier to calculate the form, degree, and locus of potential harm. Yet power relations have shifted in Crimea, security has weakened, and the salience of identification has changed and new forms of identification established. This makes it difficult to gauge the degree and forms of potential harm that might be caused by my presence in the field, to myself, research participants, and the social context. The bureaucratic problem of returning to Crimea is overshadowed by the judgment that, at least for now, I could no longer offer significant protection or minimize potential harm, neither to participants, friends, and my host family—nor myself—including the possibility of identifying publically the diverse set of individuals who originally participated in the research. This is demonstrated by the visibility and politicisation of the project following the Monkey Cage article. Neither I nor my research intentions are banal or inert: they are highly political. This could result in harm to research participants and me, should I return to Crimea, or at least could indicate potential forms of harm that would be undue.

In these situations, it can become difficult to think of ways of giving back to participants, assistants, and the broader social context because access to the field and engaging with participants is no longer possible. For example, even the idea of telling participants’ stories that contest the idea that Russian annexation was inevitable, or desirable, before 2014—which is about the best I can hope to achieve—is far removed from participants when this dissemination is to academic and policy audiences and not directly to participants or those within the field.

Finally, in considering the ethics and conduct of fieldwork, it is important to consider critically what is meant by the field, in particular in a world that is increasingly interconnected. We need to be careful about where we talk about the field, as much as how we talk about the field. For example, I have refused participating in conferences in Russia in the wake of annexation because taking the field with me to Russia, where opposing annexation is illegal, would be a source of personal harm and to the organisers of the conference. Thus, returning from the field does not mean leaving the field in an absolute sense. Perhaps thirty years ago, before social media and digital forms of communication, it would be easier to lose touch with participants in the field, as opposed to glimpse into the unfolding of their daily lives from afar. Beyond forms of communication, feminist geographers have argued, the field, in a political sense, consists of “the academy, where research is initiated, where the people we speak with live, and the social contexts and settings in which research is funded and made available to various audiences.”⁶⁷ We take the field with us in everything we do even after leaving the field, from the stories we tell to the articles we write and publish. Thus, we face questions of the ethics in the way we go about this interaction with the field after we have left, in a locational sense.

Ethics Beyond the Field in the Era of Research Access and Transparency

Finally, I want to reflect on how the need to consider ethics beyond the field intersects with moves towards data access and research transparency (DA-RT).⁶⁸ Some, in support of greater access to and transparency of research material among peers, have argued that how data is accessed, produced, and analysed is itself an ethical obligation to the discipline.⁶⁹ However, many researchers have raised concerns that such moves towards research transparency raise more ethical questions than they solve. In particular, this concerns issues of harm and consent, rendering research in dangerous contexts potentially too harmful to conduct.⁷⁰

The contribution of thinking of ethics beyond the field to debates surrounding DA-RT is to critique the notion that data remains inert after it has been collected. Given that fieldwork sites are dynamic and can change dramatically beyond the field, our data is anything other inert. Beyond redacting transcripts for sensitive material at the time, which may render transcripts meaningless, what is sensitive is contingent and context specific. While researchers can use strategies of abstraction to protect participants in how they write about the field, they cannot do this if they are required to provide public versions of transcripts. Researchers might have to constantly be on the redacting offensive, removing material already uploaded online as a field site becomes more contested, or be unable to discern what is sensitive to a sufficient degree once they have left the field. For example, what if I had posted redacted field-notes and interview transcripts from Crimea in 2013: would I have removed or further redacted the transcripts in 2014? Would I have been able to evaluate what should and did not need to be redacted from the transcripts? What purpose would this endeavor have served?

Uploading data in its “raw” but redacted form is thus unlikely to offer much meaningful insight into the process of data collection and analysis. Instead, what researchers can do in a way that maintains both ethical and transparency obligations, is to be more reflexive. As Pachirat argues, the job of good ethnography is to preserve “enough of the scaffolding” of the research process in writing up to enable the reader to discern for themselves the process of research and location of the researcher.⁷¹ In other words, such a “reflexive openness” is a form of transparency that is methodologically and ethically superior to providing access to data in its raw form, at least for qualitative data.⁷² Our job is to be able to write up data in ways that protect participants, ensure their rights to privacy and respect, convince our peers that we did what we claimed, and locate ourselves in the process of research, including how we use data to answer the questions we pose.

Conclusion

To expect the unexpected is a frequent piece of advice researchers give their peers.⁷³ However, it is frequently an afterthought, rather than a central tenet of research. One of the benefits of field research is, precisely, to get “out of the armchair” and to meet the unexpected in the field, comparing our assumptions about the world with the voices of those who are critical to our research. Studying politics is precisely about the contingency, dynamism, and contestation of the real world. Yet when we conduct research and ponder the ethical implications and obligations of this research we smooth this contingency and dynamism away. The challenges of identifying and addressing ethics obligations arising because of contingency and beyond the field might seem insurmountable. However, we cannot ignore these challenges away, as if they will not be really existing problems so long as we ignore them.

In concluding, I want to reflect on the lessons and strategies that researchers might take in considering questions of ethics beyond the field. The process of institutional ethics review is a necessary step in opening up the research to scrutiny, but it is not sufficient. Ethics review offers neither the necessary training nor tools to deal with the daily practice of ethics in and beyond the field. Among political scientists, and the profession of political science, we need to be stimulating honest conversations about the experiences, practices, and challenges of field research in and beyond the field. We need to be encouraging each other, and especially our students, to think about how we engage with participants, how we publish, and whether we return. We do not have to be clairvoyants but we are also not neutral observers. We carry with us sensitive material and have choices about what, when, and how we talk about, disseminate, and publish this material.

Within political science there is a body of expertise about how our material is politicised beyond that which we control, about when and why we might choose to become more politically engaged, and the implications (including ethical) that stem from this. Our job is to provide ways of tapping this expertise, at conferences, in doctoral programs and pre-fieldwork workshops, to further the reach and impact of these conversations in the ongoing training and mentoring political scientists receive—especially preceding PhD fieldwork.

The practice of ethics needs to be a more fundamental part of teaching research methods. This is not a revolutionary point. However, in my own experience as a graduate and early career researcher, what was considered ethics training was overly narrow, focusing on issues of gaining consent rather than on what consent meant or, in fact, what the field meant. Most graduate students begin as relative nobodies in their respective field sites. Many end PhD training seeking tenure-track jobs on the backs of trying to convince search committees they are relative experts. In other words, our field sites are often sources of transformative power for researchers. Rarely did I think about, or was asked or encouraged to think about, the process of transformation in the early stages of field research, in relation to my writing and being out of the field, until it was too late. There may be programs that do encourage such kinds of thinking and discussion but this is far from a standard of training across the discipline. While I have learnt from these experiences, there is more we can be doing for future generations of scholars to make them aware of the responsibilities that field research entails, within and beyond the field, as they transform from a nobody to trying to be a somebody.

Finally, if reflexivity is the answer to questions of identifying and addressing ethical questions within and beyond the field and making research more transparent, we have to train scholars in the practice of being reflexive and doing reflexivity.⁷⁴ This reflexive endeavour needs to stretch beyond those who are typically engaged in more reflexive practice such as ethnographers. Rather, we need a wholesale commitment within political science to a training paradigm—from experimentalists, survey methodologists and those using interviews—that encourages (if not requires) researchers to locate themselves, their impact vis-à-vis participants, and the context of the research, and to address the questions and ethical questions arising from this location. This means—for example—reimagining oneself as a research participant rather than researcher: how would I feel if receive falsified material informing me of candidates in an upcoming election? How would I feel if transcripts of an interview with me concerning personal or political questions were uploaded online? Would this affect my trust in political and academic institutions? If the answer is yes to these questions, as I suspect it might be for many of us, then we need to be training and facilitating researchers to address questions of reflexivity in the design, conduct, and writing up of research.