Research among Indigenous and other historically marginalized peoples is often exploitative. In recent decades, research codes and contracts have been developed with the aim of enhancing the ethical behaviour of researchers and the inclusion of Indigenous viewpoints. Such instruments try to prevent issues such as ‘ethics dumping’ (in which unethical research practices are exported to lower-income countries1) and ‘helicopter research’2 (in which privileged researchers (for example, from high-income countries) work in lower-income contexts and/or with less privileged groups, with minimal involvement from or benefit to local communities and researchers). In this context, codes such as the ‘Global Code of Conduct’ — which is based on the ‘San Code of Research Ethics’ — aim to urge researchers to be transparent about ethics and inclusion so that voices from Indigenous peoples are better heard and taken up in research, including in citations, authorship and peer reviewing2. They can be important empowering instruments, although they are not legally binding.

As research codes and contracts aim to reach a broad spectrum of researchers, they are very general in nature and not discipline-specific (a notable exception is the code on DNA sampling)3. This means that there is much room for interpretation on what exactly is meant when the codes and contracts promote or require behaviour that is related to rather abstract concepts such as ‘respect’, ‘honesty’, ‘(justice and) fairness’, ‘care’ or ‘process’1. It is also important to acknowledge that specific ethical concerns and issues can vary among different research disciplines3. We base our ideas mainly on ethnography, which is the study of human cultures and societies in their own context — typically through long-term fieldwork and participant observation. As a very broadly applicable methodology, it has often also informed ecological and evolutionary studies (for instance, ethnographic studies containing traditional ecological knowledge).

Although we strongly support the values promoted in the codes, it can be difficult for researchers and participants to determine what exactly these entail in practice. In addition to their empowering qualities, codes and contracts contain substantial barriers that are often overlooked. To better manage expectations regarding these instruments, our aim is to increase awareness about three related challenges among Indigenous peoples, researchers, editors and reviewers. We (1) identify how power relations and gatekeepers can influence codes and contracts, sometimes resulting in their unethical application; (2) note ambiguities regarding the impact and benefits of research; and (3) identify practical barriers to implementing such codes.

Representation and unethical application of codes and contracts

We start with the most important challenge: the unethical application of codes and contracts by managing authorities. Because they are subject to interpretation, authorities sometimes use codes to their own benefit or to prioritize their values, while ignoring the principle of empowerment that underpins the codes. These instruments can thus become political in themselves: organizations that are responsible for permissions and the implementation of a code can direct research according to their own agendas. If so inclined, they can delay research or deny permission altogether — sometimes without reasonable explanation or with questionable justifications. For instance, in 2019 a nonprofit organization (NPO) representing a group of San, which based its practices on the San Code of Research Ethics, forbade a researcher to speak to marginalized (or, as they termed them, ‘uneducated’) community members, arguing that they would not be able to understand the researcher’s questions. The NPO representatives preferred to direct the researcher to ‘preapproved’ respondents in the community4. Many San ostensibly represented by this NPO ultimately felt uncomfortable with this representation because the representatives lived far away, exhibited a sense of superiority, dominated interactions and even threatened community members. Some San were unaware that they were being represented in this way or were simply disinterested, whereas others strongly disapproved of this NPO’s behaviour and felt that they failed to undertake democratic elections and decision-making processes.

This situation lays bare the structural problems with representations of communities more generally, in which heterogeneous viewpoints and different interests are too often ignored5. Southern African communities tend to be regarded as homogeneous, but this assumption fails to recognize structures of power both between and within communities6,7 (Box 1). Thus, although ‘communities’ have become a cornerstone in research and development work, communities rarely have unitary opinions8. If codes and contracts are based on essentialized ethnic identifications, it remains unclear who exactly belongs to a group as ethnically heterogeneous as ‘the’ San9,10 (Fig. 1). It can thus be difficult to get all individuals in a community to support a research code that is focused on a specific, but not always clearly identifiable, group — particularly when codes and contracts can be used to curb people’s individual agency.

Fig. 1: San distribution in southern Africa.
figure 1

Adapted with permission from ref. 15, Springer Science+Business Media, LLC. Publ. note: Springer Nature is neutral about jurisdictional claims in maps.

Full size image

Further complications surround the fact that disparate San groups, spread over numerous countries, are often assumed to function as one collective group. For example, in 2003–2004 the Hoodia benefit-sharing agreement was signed, whereby royalties were to be paid and shared among the San people of South Africa, Namibia, Botswana and Angola11. Although a trust was set up in South Africa to receive and distribute the revenue, South African San communities and other non-South African organizations have reported difficulties in accessing these.

Who shares in the benefits and what are these?

Most research codes and contracts explicitly state that research should have ‘impact’ or be ‘beneficial’ to the community. However, given that communities contain people with diverse interests, benefits for one group of people can disadvantage others. For instance, the creation of community-based nature conservation areas and tourism may mean that some people will receive jobs and revenue, whereas others might be constrained in their livelihoods as a result (for example, through limitations on hunting, agriculture or keeping livestock)12. Who decides what a benefit is and is not, whom it is for and how it will be distributed thus become important questions.

Those who hold power over codes and contracts may make unfeasible requests for benefits, such as requests for cash, hotel stays with dinner and travel costs, or for researchers to invest in large development projects. On one occasion, a local and national government official made a general statement to community representatives that researchers — irrespective of the specific details of the research plan — should buy cars and build schools and hospitals for the community. Such a focus on immediate material community benefits reflects a partial understanding of what research impact might entail, emphasizing research as an applied practice and broadly focused on the implementation of ‘development’ (through the creation of jobs, education, health-care improvement and so on). Although valuable, given the long history of ethnographic, ethnolinguistic and ethnohistorical research conducted with the San (which has also informed ecological and evolutionary research), it is important to acknowledge that fundamental research — wherein knowledge creation is a goal in itself — can prove beneficial to Indigenous groups globally. In southern Africa, research with San that confirms their long regional tenure has been instrumental in land restitution following colonization and apartheid9,10,13. Applied research, focused on direct impacts, cannot do without the much larger body of fundamental research, as the impacts of any research are rarely foreseeable.

Furthermore, many researchers of all descriptions who undertake fundamental research have initiated or become involved in a variety of initiatives, including community tourism, water projects, COVID-19 workshops, human rights workshops and educational support, among others. In some cases, researchers have forged institutional connections between San and potential donors, relevant nongovernmental organizations or government officials. Still others have exposed political struggles and patterns of suppression and subjugation (including physical and financial exploitation) that has sometimes led to collective advocacy and activism14. Researchers may also contribute informal modest benefits (financial or otherwise, such as hospital lifts, food sharing and supporting school necessities) that are highly valued locally.

Practical barriers

In southern Africa it can be practically difficult to prepare for research owing to the unresponsiveness of organizations who have committed to implement a code; in our experience, many do not respond to emails (often owing to limited online access). Procedures can be unclear; uncertainty regarding how and where to get contracts signed and which organizations are responsible can lead to lengthy delays. Researchers experience additional difficulties in securing permissions to work with San living in national parks, such as in Bwabwata National Park, Namibia, or the Central Kalahari Game Reserve, Botswana. These extra hurdles mean that research is further constrained in these areas, as is potential development or activism based on this research.

Moreover, most San seem unaware of the codes and contracts or are simply not interested. This applies especially to the many remote and marginalized San, resulting in serious problems with the distribution of information regarding the codes and a lack of awareness around the ethical values and ideals articulated, fuelling the disinterest. Community or individual disengagement can also be a symptom of codes and contracts emanating from organizations that are politically, geographically and culturally distant to the San; signing a contract can also be seen as imposing bureaucratic cultures onto them. Codes are often applied at the national level and thus at times disenfranchise local interests or those whose communities transcend national borders. Local people are excluded not only from decisions about research by representative authorities in charge of codes and contracts, but also by governments, and many San and researchers believe decision-making about research should take place at local levels.

Conclusion

To be clear, we do not advocate against ethics instruments nor do we intend to show that we as a team have all the answers. Rather, we wish to open a dialogue around best practice, which recognizes that codes and contracts are not a panacea but instead pose inherent challenges and can have adverse effects alongside their potential for empowerment. And despite such instruments, researchers might still behave unethically — while many will behave ethically without them: there is no substitute for meaningful, interpersonal relationships as the foundation for ethical engagement. The challenges that we describe are complex and there is no single solution that will work for all Indigenous groups. There are, however, actions that could be taken towards improvement. Furthermore, as it is not always clear who exactly can be identified as ‘Indigenous’ and who not, it is important to note that the points we raise are — to varying degrees — also applicable to other groups.

Indeed, positive initiatives are coming from San groups themselves that aim to increase their participation and agency in the research process. The //Ana-Djeh San Trust in Namibia is exploring how San people can interview researchers with the goals of disseminating results and increasing understanding of the research process for the people involved. If codes and contracts are to be used, broader community training about their contents is essential — something the Tsintsabis Trust in Namibia (representing a group of Hai//om and ǃXung San) is investigating. More broadly, where possible, implementing organizations should make information accessible while striving to be more reachable and responsive, both to researchers and the Indigenous peoples they serve. This will empower people to judge the merits of their involvement with a particular project and to decide when research is problematic. It may also be useful for such organizations to collaborate with a small group of trusted researchers, preferably from different disciplines, who can then work together with the Indigenous peoples when evaluating research requests. A pitfall of this final strategy is that researchers might then be prone to take the lead and perhaps even make the decisions, so it would be critical for such researcher roles to be monitored and appropriately limited. Such steps will ensure research with the San fulfils the ethical ideals that have driven the development of codes and contracts from the outset.