The vision espoused in the Draft National Open Science Policy is constituted by the following elements: (a) equality of opportunity; (b) environmental sustainability; (c) democratization of knowledge; (d) inclusive socio-economic development; and (e) scientific research. Given that the South African Constitution is the supreme law of South Africa, and the South African state has a duty to respect, protect, promote, and fulfil the rights in the Bill of Rights, it would significantly strengthen the Draft National Open Science Policy if its vision were explicitly grounded in the South African Constitution—in particular its Bill of Rights. This can be achieved by linking the constituent elements of its vision to constitutionally enumerated rights. Such links may be apparent in some instances—such as equality of opportunity and the right to equality—given that the right to equality is well known and often referred to in policy discourse. However, such a link may be less apparent in the case of the element scientific research. Does scientific research have any link with the South African Constitution?

The answer is yes. Scientific research enjoys an explicit link with the South African Constitution in the form of the right to freedom of scientific research (contained in section 16(1)(d)). By invoking the right to freedom of scientific research and unpacking its meaning and purposes, the Draft National Open Science Policy can, going forward, provide a more solid basis for the relevance and importance of scientific research—and by extension open science—in South Africa’s constitutional dispensation. The right to freedom of scientific research serves purposes that are at the core of our constitutional value system: promoting individual autonomy, facilitating the search for truth, and supporting democracy (Thaldar and Steytler, 2021). We briefly elaborate on each of these purposes.

While the Draft National Open Science Policy deals at length with intellectual property rights, it completely omits other kinds of property rights. It is important, for example, to consider not only the copyright in datasets of human genetic data, but also common law ownership of the human genetic data that make up such datasets. Note that the question of ownership of human genetic data is distinct from—and yet interacts with—data subjects’ privacy rights in their genetic data and researchers’ possible claims to intellectual property rights related to such genetic data. Stated differently, the legal nature of human genetic data is not one-dimensional, but multidimensional (Thaldar et al., 2022). These various legal dimensions interact with each other—one right can limit another in specific, defined ways (Thaldar et al., 2022). It would therefore be a serious mistake to conceptualize human genetic data in only one or two dimensions and ignore the other dimension(s), as this would render an incomplete, and likely incorrect, understanding of the rights applicable to human genetic data. Yet, this is unfortunately what the Draft National Open Science Policy does.

We are not alone in calling for policy engagement with the issue of human genetic data ownership. A 2018 report by the Academy of Science of South Africa (ASSAf) entitled Human Genetics and Genomics in South Africa: Ethical, Legal and Social Implications (ASSAf, 2018) (the ASSAf report) called for this topic to be “carefully and vigorously debated and clarified for the South African context” (ASSAf, 2018). However, the ASSAf report does proffer a substantive position of its own, namely, that the “custodianship” of human genetic data ought to be preferred, and “ownership” avoided (ASSAf, 2018). We suggest that any normative inquiry about the desirability of human genetic data ownership should be informed by inter alia existing common law property rights. This is important, not only because respect for existing rights is a well-established norm, but also because the existence of existing rights may pose significant practical legal challenges to policy options that threaten to encroach on existing rights. For example, if, hypothetically, private research company X is the owner of the genetic data of thousands of South Africans, a policy that proposes that all genetic data of South Africans ought to be made public property effectively proposes that the state ought to expropriate private research company X’s property. This may require an excessive amount of state resources to accomplish, which raises the question of whether the policy objectives (such as greater accessibility of the genetic data) cannot be attained through different means (than making all genetic data of South Africans public property) (Kabata and Thaldar, 2023). However, apart from briefly referring to a “traditionally” held legal view regarding human biological samples, the ASSAf report does not present legal analysis on whether human genetic data satisfy the criteria for ownership in South African law.

Such an analysis has since been embarked on by Thaldar et al. (2022), showing that a human genetic data instance—i.e., the computer file containing the sequence data—is indeed susceptible of private ownership in South African law. This conclusion is important in the context of developing an open science policy, as private ownership rights in human genetic data can be a powerful tool to either facilitate or hinder greater access to such data. The Draft National Open Science Policy’s principle of “as open as possible, as closed as necessary” can only be sensibly applied if there is clarity regarding the parameters of legal rights in human genetic data.

The Draft National Open Science Policy refers to the Nagoya Protocol to the Convention on Biological Diversity (Secretariat of the Convention on Biological Diversity, 2011) (the Nagoya Protocol) and states that: (a) the Nagoya Protocol deals with access to, and benefit sharing of, genetic resources; and (b) the Nagoya Protocol has “gained interest with the idea of extension to other genomic data” (Secretariat of the Convention on Biological Diversity, 2011). First, we discuss the exact legal scope of the Nagoya Protocol, and second, we suggest that the idea of extending the scope of the Nagoya Protocol is controversial and can only serve to detract from the positive vision of the Draft National Open Science Policy.

The Convention on Biological Diversity (Secretariat of the Convention on Biological Diversity, 1992) (the Convention) provides that each state has sovereign rights over its genetic resources, meaning that each state can decide how its genetic resources will be governed, including how rights will be vested in its genetic resources. This principle is often referred to as “genetic sovereignty”. But does this principle include human genetic data? The Convention defines “genetic resources” as “genetic material of actual or potential value” (Secretariat of the Convention on Biological Diversity, 1992); “genetic material”, in turn, is defined as “any material of plant, animal, microbial or other origin containing functional units of heredity” (Secretariat of the Convention on Biological Diversity, 1992). Since the word “animal” can be interpreted as including humans, the Conference of the Parties to the Convention (1995) clarified that the Convention does not apply to human genetic material. When adopting the Nagoya Protocol, the Conference of the Parties to the Convention (2010) again recorded that the Nagoya Protocol does not apply to human genetic material. Accordingly, the principle of genetic sovereignty is limited to non-human genetic material.

But, would it not be a good idea to lobby for an expansion of the Convention and the Nagoya Protocol to also apply to human genetic data? After all, human individuals that belong to the same ethnic group share certain genetic similarities. As such, if individual members of a certain ethnic group participate in a genetic research project, such individuals may be providing valuable genetic information, not only about themselves as individuals but about their entire ethnic group. Should such ethnic groups not be entitled to control access to—and benefit from—such genetic information? Moreover, from a national perspective, should the human genetic data of South Africans not be viewed as a natural resource, similar to water or gold, that should be managed by government as public property? This was indeed the position taken by the ASSAf report (ASSAf, 2018). However, we suggest that this position would be difficult to sustain in the South African context for the following legal and policy reasons.

We now turn to the positive aspects of the Draft National Open Science Policy that we identified in the introduction above. Our analysis takes the form of a comparison of the Draft National Open Science Policy with another draft policy published for public comment by the South African government, namely, the Draft National Policy on Data and Cloud (Department of Communications and Digital Technologies, 2021) (see Table 1 below). It is relevant to note that these two draft policies were produced by two different government departments: the former by the Department of Science and Innovation; the latter by the Department of Communications and Digital Technologies. Many of the policy objectives of the Draft National Policy on Data and Cloud deserve support. These include: (a) encouraging universal access to broadband connectivity; (b) eliminating regulatory barriers and enabling competition in the data and cloud sector; (c) supporting the development of small, medium, and micro enterprises; and (d) promoting research, innovation, and technological developments in relation to the cloud. However, the Draft National Policy on Data and Cloud is premised on the ideological position that (drastically) greater state control of data is the best solution. For the reasons stated below, we are critical of this position.

While there is an important overlap between the objectives of the Draft National Open Science Policy and the Draft National Policy on Data and Cloud, as both draft policies aim to facilitate South Africans’ free access to data, the two policies propose to accomplish this objective through radically different means: the Draft National Open Science Policy through incremental, incentivized moves towards open science, within the existing intellectual property legal framework and the right to private property; the Draft National Policy on Data and Cloud, on the other hand, through the nationalization of all data generated in South Africa, the disruption of the intellectual property legal framework, and government control of access to data.

At a level of principle, these two approaches are clearly ideologically incompatible. Which of these two approaches would be better aligned with the values of the open and democratic society that South Africa aspires to be? Without doubt, the Draft National Open Science Policy. When examining the concept of an “open society” in the South African Constitution, Justice Ackermann (in Ferreira v Levin, 1996) relied on Karl Popper’s magnum opus, The Open Society and Its Enemies (Popper, 1945). This judgment points to the political philosophical source of the concept of an “open society”, and opens the door to learn from this source (Jordaan, 2017). In The Open Society and Its Enemies, Popper (1945) famously proposed that policy-making (or social engineering) in an open society should be “piecemeal”, rather than “utopian”. Piecemeal social engineering denotes small-scale social “experiments” that can be modified or reversed based on results in the social “laboratory”. Utopian social engineering, on the other hand, denotes large-scale policy interventions that seek to modify human behavior to conform to policy ideas at any cost—i.e., without error-correcting mechanisms along the way (Popper, 1945). The Draft National Policy on Data and Cloud’s proposal to nationalize all data generated in South Africa is not only far-reaching, but contains no intermediate steps to provide for learning and adapting based on real-world effects. As such, it leans dangerously towards the kind of utopian social engineering that Popper warned against. Fortunately, the Draft National Open Science Policy does not fall into this trap.

Also at a practical level, the Draft National Open Science Policy offers a more realistic pathway to reaching the objective of facilitating South Africans’ free access to data. It respects the existing legal frameworks and individual rights, while envisioning a cultural change towards open science that will be championed by an official body, the Open Science Advisory Council, and incentivized along the way. By contrast, the Draft National Policy on Data and Cloud proposes to break down established legal frameworks and rights, and impose government control of data. It is human nature to respond positively to incentives, and to respond negatively to infractions on one’s reasonable expectations, such as the expectation that one’s individual rights will be respected. Furthermore, obvious questions can be raised regarding the efficiency of government control of data. Accordingly, if South Africans’ free access to data is the public policy objective, the Draft National Open Science Policy offers a much more attractive policy pathway to accomplish this objective.