1 Introduction

Over the last decade heritable human genome editing (HHGE) has moved from being an idea to becoming a reality. HHGE involves the re-writing of parts of the human genome (single nucleotides or longer stretches of DNA) in a manner that ensures the transmission of these changes to future generations. With the emergence of CRISPR-Cas9 in the early 2010s (Ishino et al. 2018) and the birth of the first CRISPR-modified children in 2018 (Cyranoski 2018; Greely 2019), it has become clear that scientists are now able to introduce such targeted heritable modifications, and to do so with increasing precision.

HHGE is controversial because it not only allows researchers to delete or replace disease-related variants from the human gene pool, but also to genetically enhance humans.Footnote 1 Because of its potential to fundamentally transform medicine and human society more generally, the research surrounding HHGE has become a key battleground for debates about academic freedom and self-governance. The question is: given the various ethical, social and scientific challenges—as well as opportunities—that this technology brings with it, how ought we to govern its development and use? Should scientists and other experts decide what the technology should look like; what it can be used for; who gets access to it; and who might benefit from it or risk negative impacts? Or is there a role for non-scientists, including regular citizens, in these decision-making processes? And if so, what should this role be?

There is now a strong consensus among various stakeholders, including scientists, that some sort of public engagement or participation is necessary in this process (Conley et al. 2023).Footnote 2 Despite this consensus, public engagement exercises have been slow to emerge and only over the last few years have we seen first efforts materialise. Whilst this is encouraging, it is important to ask what these engagement exercises mean. Are we witnessing a shift in the way science and technology are governed? Are these participatory exercises meaningful, in the sense that the public is given relevant decision-making power? Or are they just window dressing?

To answer these questions, it is important to first develop a better understanding of what “meaningful participation” means. Participation can come in many forms, and it can be used for various purposes. Not everyone calling for more public participation will thus be calling for the same thing. So even though we seem to have a broad consensus on the need for public participation in HHGE, there might be a lot of dissent hidden within that consensus.

In the first part of this paper, I will introduce a framework that can help us better structure the debate about meaningful engagement (Sect. 2). The framework consists of three distinctions: negative/positive; weak/strong; and shallow/deep forms of engagement. The first distinction is about the motivation behind the participatory practice, the aim it is supposed to achieve. The second is about the power that is ascribed to the publicFootnote 3 in the decision-making process. The third distinction is about the point in the process at which the public is asked for input. I will argue that most authors who call for “meaningful” engagement ask for strong and deep engagement that mainly pursues positive goals.

In Sect. 3, I will use this framework to look at existing engagement initiatives in the context of HHGE. This analysis will show that these initiatives don’t deliver the strong and deep engagement that many advocates of public engagement are calling for. This raises the question of why we still don’t have more meaningful engagement (assuming that this is a valuable goal). What are the hurdles, and how could they be overcome? I will discuss some of the hurdles that have been identified in the literature on public participation and discuss how they could be used to improve the situation in the context of HHGE. I will also propose a further hindrance that has so far not received much or any attention in the literature on public participation and the governance of HHGE, namely, the power structures within science itself.

In Sect. 4, I will expand on this point by highlighting the fundamentally non-democratic organisation of science, which contains elements of oligarchy, meritocracy, epistocracy, and probably also some local tyranny. The people who call most of the shots in a research field are Principal Investigators (PIs) and institutional leaders (who are often current or former PIs). They are not elected but have reached their position of power in part because of their experience, their past work, and because of the connections to other powerful people they have established over the years (hence the mix of oligarchy, epistocracy, and meritocracy). Importantly, such a system will have little or no place for deep and strong public participation, which is a tool of direct democracy.

This analysis points to a different obstacle and thus a different path to meaningful engagement: to prepare the ground for deep and strong public participation we might have to first re-shape the way in which power is distributed within science itself. More specifically, I will suggest the creation of science-community assemblies at institutional level that are tasked with shaping key research priorities and approaches surrounding HHGE-related research activity. These assemblies would not just include PIs but also postdoctoral research fellows, PhD students, and technicians, chosen by lot for a specific period of time. Such a body would introduce a flatter decision-making process within science and could thus form a stepping stone on our path towards a deeper and stronger form of public participation, as it would help to break up existing hierarchies and power structures. This move towards more participation could build on existing examples of a softening of power structures, such as the inclusion of students in hiring panels in Germany or the inclusion of laypeople in funding decisions and research supervision in the Alzheimer’s Society in the UK (see Sect. 3.3). The hope would be that these and other bridging practices could pave the way to a more inclusive practice of decision-making in research on genome editing and beyond.

This proposal is of course tentative and speculative. The claim here is not so much that assemblies are “the” way forward or that they could serve as a replacement for meaningful engagement. Instead, the idea is that this is could be a part of the way forward that has so far been neglected in the debate about meaningful public participation in the context of HHGE (and science and technology more generally).

Ultimately, the exact manner in which a flatter hierarchy could be established in science is a question that should be discussed and decided by the people who will be affected by it, i.e., the practising scientists, policy-makers, and the broader public that funds science through its tax payments. The main point this paper is trying to establish is that we have to fundamentally re-think the way in which power is distributed within science before direct democratic tools such as deep and strong public participation can gain a genuine foothold.

2 The rise of public participation

There is now a broad consensus, including among scientists, that the governance of HHGE and other controversial areas of research and technology requires some form of public participation. In this section, I will survey key aspects of the debate about public participation that has developed over the last four decades. This analysis will show that a push for “meaningful” public participation sits at the core of this debate.

2.1 The broader debate about public participation in decision-making

In the second half of the 20th century calls for a more direct involvement of the public in modern democracies grew louder. This was in part due to a “deliberative” or “participatory” turn in democratic theory (Dryzek 2002; Bherer et al. 2016). Driving this turn was a worry regarding the authenticity of (representative) democracies. “Authenticity” here means that the citizens of a democratic state should have the opportunity to have a substantial rather than just a symbolic influence on the decisions that are being made by the government on their behalf.

In the late 20th century, it became clear that this authenticity, and with it the legitimacy of democratic systems, was in decline (Fung 2015). This initiated a search for new tools that could help reverse or at least slow down this decline. Participatory practice was one of these tools. From the 1990s onwards we can thus observe a steady increase in the use of participatory governance tools across the globe (Fung 2015). This was an increase both in terms of the number of participatory processes being realised, and in terms of scope, i.e., the number of problems that were being addressed with participatory tools (ibid). Examples include Participatory Budgeting, which originated in Porto Alegre in Brazil in 1989 and which has since spread globally (Ganuza and Baiocchi 2012), or the British Columbia Citizen’s Assembly on Electoral Reform in 2004 (Carty et al. 2008).

At the same time, calls for more public engagement in science and technology started to emerge. These calls were—at least in part—driven by a worry about a growing public mistrust of experts and their ability to manage the complex technologies that are an integral part of modern societies (Jasanoff 2003; Wynne 2006). Examples that triggered these worries are controversies over the development and the use of genetically modified organisms (GMOs) or the management of nuclear waste, and also high-profile failures such as the Challenger Shuttle explosion or the Chernobyl disaster.

These developments raised questions about experts’ ability to predict and control the power of our technological innovations.Footnote 4 This gave rise to a debate about the need for more public participation. Research and technology, so the idea, should not simply be left to scientists or experts, at least when they carry a risk to society. The broader public, however defined, should be able to play at least some role in the decision-making process.

But what should this role be? And how much power or influence should regular citizens have over the governance of science and technology? Different authors have come up with different answers to these questions, leading to a complex landscape of approaches to public participation in science and technology. In the following, I will try to give more structure to this debate by distinguishing between two types of motivations (negative and positive) and two types of approaches (weak/strong and shallow/deep) to public engagement.

2.2 The negative/positive distinction

As already hinted at above, for some scientists and policy-makers public engagement is seen as a tool for tackling growing public mistrust. This public mistrust is seen as problematic because it a) threatens to undermine the authority of science and the evidence-based policy-making that builds on it and b) carries the risk of imposing stricter regulation on science, which could reduce scientific freedom. The call for more public engagement is thus in part an attempt to keep the public out of science and policy-making.

There are two ways, roughly, in which public engagement should help protect science and policy-making: 1) by educating the public, and 2) by enabling scientists and policy-makers to gather more information about the views of the public on a specific technology or issue.Footnote 5

The first approach builds on the idea that people who reject a particular part of research or technology do so out of ignorance; if they only knew more about the issue, so the thought, they would be more positive about it. This “deficit model” has largely been rejected by the scholars studying it, but it stubbornly remains a part of the discussion about the governance of science and technology (Cortassa 2016; Simis et al. 2016). Even in the recent debate about HHGE it features as a prominent reason for keeping laypeople out of science or science-related policy-making (see Waltz et al. 2024b).

The second approach (i.e., gathering more information through engagement with the public) seems more benign, in the sense that it does not necessarily assume that citizens are somehow deficient. Instead, it assumes that through the engagement process researchers and policy-makers can get a better picture of what the public thinks about a specific issue. This information can then be used to try and defuse potential flashpoints or outcries (Burall 2018) by either embarking on targeted information campaigns (potentially dipping back into the deficit view) or by installing self-imposed controls or restrictions.Footnote 6 Self-imposed restrictions are usually preferred to having “outsiders” impose new regulations on science and technology, as scientists or experts can decide themselves what restrictions to impose and when they can be lifted.

Both of these approaches see public engagement as a form of self-defence: the aim is to protect the autonomy of science and policy-making, leaving decision-making in the hands of experts. I describe this as a “negative” motivation for engagement because the public is seen as a threat to science, a negative factor that must be removed or at least reduced through engagement. Importantly, the public is not asked to be part of the decision-making process itself.

These negative motivations contrast with the positive reasons that can also be behind calls for public participation. In these cases, participation is not seen as something that helps protect science or policy-making from negative influences but that helps improve it. Archon Fung, for instance, identifies three positive roles for public participation: providing legitimacy to policy decisions, increasing the effectiveness of decision-making, and increasing justice (Fung 2006).

The first point has to do with the influence that citizens have on the people who make decisions, be that about laws or specific policies. In a representative democracy this influence mainly consists in the ability to elect the decision-makers. In science there are no elected officials, but legitimacy can still be increased if citizens have an opportunity to play a more direct role in the decision-making process.

Increased effectiveness is another alleged benefit of public participation. Effectiveness here means that the decision-making process actually solves the complex problems it is supposed to solve. In this case public participation is mainly treated as an epistemic asset: the inclusion of the broader public can re-shape how a problem is framed, and it can diversify the kind of information that is considered during the decision-making process (Fung 2015). This can include an expansion of the types of risks and benefits that are seen as relevant (Wynne 2006), or the range of regulatory actions and policies that are being considered (National Research Council 2008). This expansion of viewpoints does not mean that public participation automatically improves every decision. But by broadening the range of options or aspects that are being considered, the decision-making process is put in a better position for finding solutions that work (or so the argument goes).

The third advantage that Fung highlights is justice: we know that the preferences of a powerful minority can capture a decision-making process, which can then lead to unjust outcomes for groups that have less power (even if they represent the majority). Bringing the broader public into the decision-making process can help balance the influence that different interests have on the outcome of the process (Fung 2015). This is also relevant for worries about the undue influence of commercial interests on decision-making, a problem that is of increasing importance in a science that is closely intertwined with industrial actors and interests.

Additional advantages of broader participation can be an increase in accountability (Jasanoff 2003) and in the quality control of scientific outputs (Funtowicz and Ravetz 1993). In the case of technologies such as genome editing, which are often labelled as “emerging” technologies, engagement is also seen as an important tool for developing a shared vision of what our technological futures should look like (Jasanoff et al. 2015; Hurlbut 2015). This matters because the technological advances that are emerging are not neutral; they reflect prior moral commitments. But these moral commitments should not be set or imposed by a narrow group of experts. Instead, they should be discussed and developed in a more inclusive and co-produced manner (see, e.g., Wynne 1993, 2006; Hurlbut 2015).

These negative and positive motivations for public engagement are tightly coupled to differences in the amount of power that is allocated to the public and to the types of participation practices that are deployed in the process. In the next section I will use Archon Fung’s “Democracy cube” (Fung 2006) to think about this diversity of practices and roles.

2.3 The weak/strong distinction

Fung highlights that engagement entails more than just organising a public hearing or sending out a survey—the public should be engaged in a more meaningful manner (Fung 2006). To capture this power and the diversity of practices that surround it, he proposes a cube as a framework within which different approaches can be placed. The three axes of the cube are: “participants”; “communication and decision mode”, and “authority and power”.

In terms of participants Fung identifies three broad classes of actors that might be involved in the decision-making process: state actors; minipublics; and the general public. The term “minipublic” is taken from Robert Dahl’s work and refers to groupings of citizens—ranging from tens or hundreds to thousands (but not millions) of people—that are assembled to deal with a specific issue (Fung 2006). A minipublic can be composed of professional stakeholders only, or it can contain lay stakeholders, or it can be formed via random selection from the general public.

The axis “communication and decision mode” sets out how a group of participants would go about dealing with an issue at hand. These modes can include the simple listening to others (for instance if a citizen takes part in a public hearing as a spectator), or the voicing of everyone’s preferences (for instance if citizens take to the microphone at a citizen assembly), or it can be a more sophisticated exercise that includes moderated deliberation and negotiation. The first two modes correspond to what Rowe and Frewer (2005) call “public communication” (the sponsor of an engagement exercise conveys information to the public) and “public consultation” (information is conveyed from the public to the sponsors of the engagement exercise). The deliberative mode corresponds to their notion of “public participation”.Footnote 7

The third axis of Fung’s cube is “authority and power”. This specifies the power which each group, with its specific composition and mode of discussion, should ultimately have in the decision-making process. Here Fung distinguishes between three types of power: the process can be entirely for the personal benefit of the participants (for instance increasing their understanding of an issue); it can be merely informative for decision-makers; or it can be part of the actual decision-making process (co-governance or direct authority). The first two types comprise what I will refer to here, following Nancy Fraser (1990), as “weak” publics. The third type, which includes co-governance and direct authority, is what I refer to as “strong” participation. In this last case the public is a decision-maker.

2.4 Shallow and deep participation

From the above we can already see how these distinctions start to come together: calls for public participation driven by negative motivations are often proposing a weak role for the public: citizens should either be educated, or they are seen as a mere source of information that helps experts build their strategy. In the positive approach there is often (but not always) a focus on strong participation, as its proponents argue that decisions are better (i.e., more legitimate, just, accountable, etc.) if they are taken by a diverse group of actors. What matters in such a deliberative and inclusive process is not so much statistical representation, but rather the fact that there is an opportunity for everyone to be included in a dynamic and open process of decision-making (Biegelbauer and Hansen 2011). This is also why minipublics, which rarely are statistically representative of a population, are often seen as an adequate basis for deliberative exercises.Footnote 8

The positive reasons for participation also highlight another focus of this drive for participation, namely, the point in the decision-making process at which the public should be involved. As we have seen above, there is a worry that important assumptions or values are injected into a research program or technology at the beginning of the process; the very framing of a problem, or of a possible future, is seen as something that needs to be developed communally.

Hurlbut (2015) and Jasanoff et al. (2015) develop a version of this view in the context of HHGE. But the notion that we need to collectively think about possible futures and use such foresight or anticipation in the management and development of emerging technologies (or in policy-making more generally) has also been central for other strands of the debate about governance in the 21st century. Examples include the idea of “anticipatory governance” (Fuerth 2009; Guston 2010, 2014); “participatory foresight” (Nikolova 2014; Bourgeois et al. 2017), and Responsible Research and Innovation (Sutcliffe 2011; Stilgoe et al. 2013).

What matters for all these approaches is not only public engagement but also the timing of such engagement. Those who think about the development of science and technology as a linear process tend to use the term “upstream engagement” to highlight that non-scientists should have some influence from the very beginning of the process (Royal Society and Royal Academy of Engineering 2004; Wilsdon and Willis 2004). Others who are sceptical of the “stream” or “pipeline” views of science and technology might avoid this terminology (see, e.g., Joly and Kaufmann 2008; Stirling 2008) but will raise similar issues regarding a “deeper” involvement of the public. The general idea is that non-scientists should not just have a say on technology once it has been developed. Even though this would be a form of strong participation—as the public is given decision-making power—it is not seen as meaningful by these authors because the public is only allowed to sign off on established facts and visions; key choices have already been made and are thus imposed on them (Wynne 2006). In such cases we can therefore speak of strong yet “shallow” participation, as opposed to strong and “deep” participation.

2.5 The importance of diversity and deliberation

Before I apply these three distinctions to an analysis of emerging participatory practices in the context of HHGE, I want to highlight two conclusions we can draw from the above. First, the push for positive, strong, and deep participation is closely connected to a worry about diversity: the call for broader participation is, in part at least, about the benefits of having a wider set of views and backgrounds present (and influential) at all stages of the decision-making process. Advocates of positive, strong and deep public participation are worried about the narrowness of the group of people who currently call the shots. Experts, for instance, are taken to represent a limited range of the theoretical knowledge, the (lived) experience, and also the skills that are needed to come up with solutions for the complex problems that modern society faces. This is not only about the problem of following a “scientific perspective” that is too narrow to grasp the complexity of the issues at stake (see, e.g., Sarewitz 2015). It is also about a lack of diversity in terms of gender, race, age, disability, or socioeconomic background.

Second, deliberation is often seen as crucial to arriving at meaningful decisions that represent an informed public opinion on the issue at hand. Non-deliberative opinion polls could of course be used to gather the views of the public on a specific issue. If an adequate sampling protocol is applied, such a poll could even capture the diversity of views that the public holds. But as Adashi and co-workers point out, deliberative engagement allows us to gather public input that is not just “informed” in the sense of building on adequate knowledge, but also in the sense that it is based on group reflection and judgment (Adashi et al. 2020). This echoes the idea of “democratic reason” and collective intelligence that Hélène Landemore uses to make her case for the epistemic benefits of deliberative democracy (Landemore 54,55,a, b). To move forward, then, what is needed is not just any deliberation (for instance the ideal hypothetical deliberation that (Kitcher 2011) proposes), but actual deliberation among a diverse group of actors, including members of the general public.

3 The debate about public participation in the governance of HHGE

Urgent calls for public engagement in the context of HHGE started to emerge in 2015, after rumours began to circulate that researchers in China were using novel genome editing tools (CRISPR-Cas9) to modify human embryos in vitro. These rumours turned out to be true: in 2015 Liang and co-workers reported that they modified the genomes of non-viable human embryos (Liang et al. 2015). Even though none of these experiments resulted in an actual pregnancy, the fact that this research was being conducted created a stir in the research community and beyond.

In response to these rumours, several papers were published in high-profile journals which stated that at this early stage the technology should not be used (Lanphier et al. 2015) or that its use should be “strongly discouraged” (Baltimore et al. 2015a).Footnote 9 Both of these papers stressed the need for more public engagement on the issue.

Over the following years this need for public engagement was highlighted again and again at academic conferences, in papers, reports, books, and position statements (see, e.g., Baltimore et al. 2015b; Chan et al. 2015; Jasanoff et al. 2015; Sarewitz 2015; Baylis 2017; NASEM 2017; Nuffield Council on Bioethics 2018; German Ethics Council 2019; Blasimme 2019; Baylis 2019; Jasanoff et al. 2019; Vijlbrief et al. 2020; Adashi et al. 2020; Andorno et al. 2020; Dryzek et al. 2020; National Academy of Medicine, National Academy of Sciences, and The Royal Society 2020; Greely 2021; WHO Expert Advisory Committee on Developing Global Standards for Governance and Oversight of Human Genome Editing 2021).

Interestingly, already a quarter of a century ago there were calls for strong and deep public participation in the context of what was then called “inheritable genetic modifications” (IGMs). These calls were made in a report by an inter-disciplinary working group that was assembled by the American Association for the Advancement of Science (AAAS). In this report, published in 2000, the authors formulated many of the worries and recommendations that we find again 15 years later. For instance, the report calls for the establishment of a public oversight mechanism which has authority over IGM activities (Frankel and Chapman 2000, p. 51) It also calls for public involvement in setting the priorities of IGM research so that the future direction of the technology is not just defined by commercial interests, which already then threatened to dominate the field (ibid, p. 50). Furthermore, the report highlighted that it is up to society as a whole to “consider how much evidence of safety and efficacy will be required before permitting either human clinical trials or non-medical applications” (ibid, p. 25). A key claim of the report was thus that broad and inclusive public engagement is required not only to decide if such technology should be used (once available) but also to shape its future trajectory (ibid, p. 5).

A quarter century after this initial call for deep and strong public participation, and a decade after the calls that were published in 2015, we now see the emergence of several public engagement exercises that directly address HHGE. In the next section I will introduce these cases and analyse the type of engagement they represent, using the framework developed in Sect. 2.

3.1 Four examples of public engagement

The first example of emerging public engagement in the context of HHGE is the Dutch DNA-dialogue, which was set up by a multidisciplinary consortium of 11 organizations and that took the form of a national survey and 27 moderated dialogues conducted between October 2019 and October 2020 (van Baalen et al. 2021; Houtman et al. 2023). The goal of this initiative was to learn about the opinions, and potential change of opinions, of Dutch society about the clinical application of HHGE. The dialogue aimed at gathering a diversity of views from the public that could inform political decision-making (van Baalen et al. 2021, p. 4). To this end the engagement process was carefully designed to include a wide range of perspectives, covering not just different age groups (including school children) or professions, but also, for instance, people with a migrant background. In addition to gathering opinions a further goal of the process was to stimulate a broader and long-lasting process of collective reflection and opinion formation on HHGE (ibid).

The main insight gained from this project was that roughly 70% of the Dutch population support the use of HHGE if used to tackle severe genetic disease, whereas only 8.5% would support HHGE-mediated genetic enhancement (Houtman et al. 2023). Interestingly, the analysis of the outcomes indicated that engaging in dialogue did not change the acceptance rate of participants (ibid).

Importantly, the DNA-dialogue pursued a positive approach: whilst informing the public about the technology was part of its aims, this was not done to avoid a backlash or to reduce public rejection of HHGE (Houtman et al. 2023). Instead, the main aim was to use the insights gained from the public to inform a collective discussion about the desirability of the technology (van Baalen et al. 2021).

This, however, also means that the project is a case of weak engagement: it was an exercise in gathering information on the hopes, values, and concerns of the broader public in relation to the use of HHGE. The goal was not to have the public make decisions. Furthermore, the dialogue mainly focused on whether and how the technology should be used in clinical practice (i.e., after it has been developed). This means that the DNA-dialogues did not aim at deep engagement. According to the framework proposed in Sect. 2, the Dutch DNA-dialogue would therefore have to be classified as a positive, weak and shallow engagement exercise.

The second example is a deliberative public engagement exercise conducted in South Africa (Thaldar et al. 2022). The engagement consisted of a minipublic of 30 people from a variety of backgrounds that met over three consecutive evenings on Zoom to a) be informed about the technology and b) engage in a moderated deliberation on the potential uses of the technology and its ethical and social implications.

The outcomes of this exercise were similar to those of the Dutch DNA-dialogue. A vast majority of participants supported the use of HHGE to tackle serious genetic conditions (assuming the technology is safe and effective), whereas the use of HHGE for genetic enhancement was clearly rejected (ibid, p. 11). An important difference between the two studies is that the South African project found strong support for the use of HHGE to introduce immunity to infectious diseases, in particular HIV, whereas a majority of participants in the Dutch study rejected such uses. Those who endorsed editing for immunity in South Africa often mentioned issues such as the high economic cost of HIV/AIDS in the country, demonstrating that the acceptance of specific uses of the technology will be highly context-dependent (ibid, p. 14).

The aims of this engagement exercise were mainly positive, as the project leaders emphasise the importance of broader public debate when tackling fundamental questions such as our humanity (ibid, p. 2). These questions, so the implication, cannot or should not be answered by experts alone. The authors, however, also mobilise some of the more negative motivations, as they highlight that both the rejection and the uncritical endorsement of emerging technologies are often driven by misconceptions about the underlying science. One goal of the project was therefore to educate the public and thus remove misconceptions that can lead to baseless fear (or optimism). This highlights that the positive and negative elements that were described in Sect. 2 are not mutually exclusive—any given exercise in public engagement might be driven by both motivations.

The engagement process that should achieve these goals was by design weak and shallow. The public was not involved in decision-making processes, as the aim was simply to gather the views of the public to help policymakers “identify where lines should be drawn” (ibid, p. 2). The deliberation did not have any binding power, meaning it is a weak form of engagement. It also primarily focused on questions of how the technology should be used (once developed), indicating that it is again a shallow form of participation.

A third example of a HHGE-related engagement exercise comes from the UK, where the public participation charity “Involve”, together with Wellcome Connecting Science, organised a citizen’s jury on HHGE in 2022 (Wellcome Connecting Science 2022). The jury of 21 people was designed to reflect patients who are eligible for genomic services in the NHS (genetic counselling or genomic medicine services) and was asked to discuss the question of whether HHGE should be allowed in the UK for serious genetic conditions. The jury members deliberated over four days and then developed 15 recommendations. As in the other two projects discussed above, the vast majority of participants supported the use of HHGE for serious genetic conditions (81%).

The aim of this exercise was summarised by the organisers as follows: “Whilst the editing process is not currently 100% accurate, scientists predict that it soon will be. This means that public audiences should be brought into the conversation now about the application of the technology so that policy makers can take account of societal perspectives when discussing the legislation” (Wellcome Connecting Science 2022). As with the two previous examples, the engagement was thus an information-gathering exercise—the recommendations were merely informative and had no binding force. Furthermore, the jurors were only asked about how the technology, once developed, should be applied. This exercise, thus, is again an example of a shallow and weak engagement.

What is interesting about the outcome of this particular engagement process is that the jurors repeatedly stressed that it is was necessary to give individuals with lived experience of a condition genuine influence on the decision-making process in science. The jurors emphasised that they did not want to stifle scientific advancement, but that scientific research should factor in lived experience, implying a desire for strong and deep public engagement. I will return to the idea of giving “experts by experience” decision-making power in Sect. 3.3, where I discuss the approach to funding and research oversight implemented by the Alzheimer’s Society in the UK.

The fourth example I want to highlight here is a citizen’s jury funded by the Genomics Health Futures Mission (GHFM) of the Medical Research Future Fund (MRFF) in Australia in 2021 (Nicol et al. 2022). In this case a group of 23 people met for deliberations over the course of 4 days and addressed the question: “Under what conditions (or circumstances) might the application of human genome editing technology be acceptable”. This question is almost identical to the questions asked by the UK and South Africa studies.

Participants were asked to come up with recommendations which were then provided to relevant government and policy bodies, as well as the broader public, with the hope that the report will “assist” the relevant people in their decision-making. The main thrust of the exercise was therefore again positive: members of the public were not seen as some sort of obstacle that needs to be overcome or moulded in a particular way. Instead, the jury was seen as a key source of input, providing diverse perspectives to decision-makers which they would otherwise lack.

However, as there is no binding power to the recommendations and as the discussion focused only on the application of the technology, it is again an example of a weak and shallow engagement exercise. The public was involved in a positive spirit, but it was not meaningfully engaged (if we apply the above definition of “meaningful engagement”).

3.2 A step forward?

The four engagement exercises were powerful and insightful initiatives that represent an important step forward. They were carefully designed and executed (see, for instance, Fig. 1 in Nicol et al. 2022 for an overview of the various pre- and post-process activities, as well as the additional surveys, interviews with participants, and briefings that were conducted as part of the Australian citizen jury), and they provided policy-makers and scientists with a more nuanced and critically informed public opinion about the question at hand. All studies also build, at least in part, on a positive rather than a negative motivation, thus realising at least one aspect of the push for more meaningful public participation.

But their careful approach is also their main weakness: if we look at these exercises from the perspective of the framework developed in Sect. 2 then they don’t deliver. First, all four exercises were weak: the public was not allowed to sit at the table of scientific decision-making. Second, they were all shallow interventions, as the public was only asked to comment on how the technology should be applied once it has been developed. Two of the three criteria for meaningful participation have therefore not (yet) been realised.Footnote 10 This raises the question of why we don’t see more strong and deep forms of participation in HHGE.

3.3 Why is it not working?

There are several possible explanations for the lack of deep and strong engagement in the context of HHGE. First and foremost, meaningful public participation is not a globally accepted good. In many countries there is no tradition of asking the public for their views on particular issues. It is little surprise, then, that public engagement, a tool of direct democracy, will find little uptake in states with weak or no democratic institutions.

However, there are at the same time many democratic states across the globe that practice direct forms of democracy, for instance through binding referenda. In most countries such referenda are relatively rare, but there are cases (e.g., Switzerland or California) which have a long tradition of holding regular popular referenda. The problem is that even in these contexts we don’t observe strong or deep public participation in science.

It could be argued that this is due to the highly technical and complex nature of the questions and problems raised by science. But the practice of popular referenda in countries like Switzerland undermines such an explanation. The Swiss public is regularly asked to vote on highly complex issues such as the type of fighter jet the Swiss army should be allowed to buyFootnote 11 or whether curbs on genetic engineering should be implemented.Footnote 12 Giving Swiss citizens direct power over complex questions of high consequence for the future trajectory of the country has not led to the implosion or demise of Switzerland, showing that, in principle at least, a more inclusive approach to complex decision-making is feasible.

But despite such proof-of-principle cases, it has to be added that even in the case of Switzerland strong and deep public engagement in science is not the norm. Whilst there might be popular referenda on core scientific issues, there is no permanent meaningful public participation in the governance of everyday research. This is not due to a lack of interest from the public. As the case of the UK citizens jury discussed in Sect. 3.1 illustrates, citizens often express a keen interest in having a say on how scientific research is planned and governed.

This willingness to be involved is also demonstrated by the activities of the Alzheimer’s Society, a registered charity in the UK which supports research on Alzheimer’s with a fundraising income of over £ 100 million per year.Footnote 13 The society has a unique approach to how it governs its funded research, as it involves laypeople with lived experience by giving them direct decision-making power regarding what is being funded and how funded research is steered and monitored. These “research network volunteers” are, for instance, part of the team that reviews grant applications and they also monitor and support research projects funded by the charity.Footnote 14 Whilst these are small-scale examples, both the UK citizens jury and the case of the Alzheimer’s Society show that even in countries with no or almost no direct democratic tradition there is a clear appetite and capability within the public to take on a more direct role in scientific decision-making.

But if meaningful participation is possible and the public is interested in being involved, then what is the main hindrance to stronger and deeper public participation in science? Maybe the problem lies, in part at least, on the other side of the engagement process, i.e., the scientists. Maybe these key stakeholders are ultimately the ones blocking stronger participatory practices?

This explanation, however, is challenged by recent empirical evidence. Surveys of researchers have shown that a majority of scientists have a positive view of public engagement (Rose et al. 2020; Cadigan et al. 2022; Waltz et al. 2024b). This mainly relates to weak engagement (educating or informing the public) but there also seems to be a more positive attitude towards a deeper involvement of non-scientists (see, e.g., the study by Cadigan et al. 2022). These positive views are in line with the finding that only a minority of researchers (18%) see “the public” as a negative “other” (Simis et al. 2016). The majority of researchers (in the cohort that was surveyed) either did not think of the public as an “other”, or if they did then they had a neutral or even positive view of the relationship between science and the public. For instance, 15% of the researchers thought that the public’s perspective, understood as a non-scientific other, can be beneficial for science (ibid, p. 407).

But if scientists have a significant appetite for, or at least an openness towards, various forms of engagement, and if the public is willing to be involved, then why is there still not more of it? In part the lack of engagement might be due to systemic barriers to engagement, in particular a lack of infrastructure and incentives (Rose et al. 2020; Burns et al. 2021). As Scheufele and co-workers highlight, this is not just an issue of time, money, and “credit” within the research community but also a problem with the broader set of practices that is on offer. When it comes to public engagement there is no one-size-fits-all approach, as different publics or research fields will require different engagement practices (Scheufele et al. 2021). This means that we need clearer mechanisms for engagement that scientists can make use of, if we expect engagement to become a more common occurrence in areas such as HHGE (ibid). At the moment, the lack of incentives and clear action mechanisms might hinder public engagement, despite the positive attitude many scientists have towards participation.

Another factor that is often highlighted as a hindrance to stronger participation is the framing of the debate, and in particular the narrative of “objective” risks that dominates the debate about HHGE. As Almeida and Ranisch point out, a key feature of the recent ethical debate about HHGE has been a shift from debates about categorical questions (such as “Does the impossibility of obtaining consent from future generations categorically rule out HHGE?”) to pragmatic issues, in particular questions of safety and risk-benefit ratios (Almeida and Ranisch 2022).

This focus on risk can be a hindrance to public participation because it operates with a narrow and technical understanding of risks (in this case “biological” risk). This type of risk is about issues such as off-target effects of the genome editing process, which can lead to an increased risk of developing cancer or negatively affect the general functioning of the cell.

Once risk is framed in this way, the debate has become a technical issue that has to be addressed by experts in an objective and empirical manner. As Jasanoff and co-workers put it: “Limiting early deliberation to narrowly technical constructions of risk permits science to define the harms and benefits of interest, leaving little opportunity for publics to deliberate on which imaginations need widening, and which patterns of winning and losing must be brought into view” (Jasanoff et al. 2015). In this framing the knowledge-deficit also becomes important again. It is argued that people simply don’t know enough about these biological risks, so they should not be involved in these debates (ibid).Footnote 15

The above analysis suggests at least some measures that could be implemented to improve the status quo. For instance, more effort should be made to tap into and amplify the positive attitudes towards meaningful participation that clearly exist within the public and also science. Putting a stronger spotlight on success stories such as the Alzheimer’s Society could play a role in such an amplification drive.Footnote 16 Also, more effort has to be put into creating an infrastructure that supports meaningful engagement, thus reducing the activation energy needed for individuals or institutions to initiate such practices. Re-thinking the way in which we frame the debate about HHGE and other emerging technologies will also have to be part of this process.

However, it is clear that the above is not an exhaustive list of possible explanations for the lack of meaningful participation in the governance of HHGE. Developing a better understanding of the various hindrances this process encounters will require a detailed empirical investigation that looks at the attitudes towards, and dynamics of, public participating in the governance of genome editing. But in the final section of this paper, I will suggest one additional factor that might hinder meaningful public participation in HHGE and which future analyses of the problem might have to consider in order to foster a more inclusive governance of the technology. This factor is the power structure within science.

4 Power structures and public participation

The problem of power in science, I suggest, has at least two aspects that are relevant to the case of HHGE and public participation: first, there is the particular way in which power is obtained in science. This way is fundamentally non-democratic. Second, the group which obtains power in this non-democratic manner lacks diversity.

4.1 The path to power in science

The key decision-makers in science are usually the heads of individual research groups, often referred to as “Principal Investigators” (PIs). The PI will usually decide who they employ in their laboratory, what projects the employees will work on, when a project is discontinued, when data is ready for publication, etc. PIs will also move through the ranks within an institution, taking up leadership positions such as Head of Department, whilst either keeping their research group or re-starting it when their period of service is completed. In many universities PIs will also be part of the wider leadership team of the institution. PIs will also form the informal leadership circle in a specific research field. This is a more diffuse power structure, but PIs are usually the ones who sit on grant review panels and editorial boards of discipline-specific journals, and who lead the key professional associations. Through these positions, they can control or at least influence much of what is happening in a field.

As a result of this power structure in science, the main group of people who actually produce and analyse data—the technicians, PhD students and postdoctoral research fellows—don’t have any significant decision-making power.Footnote 17 In addition, in countries such as the UK, PhD students are not classified as members of staff, which excludes them from several benefits and also makes it more difficult to include them in decision-making processes.Footnote 18

Within laboratories or institutions there will of course be open discussions about the direction of research or experimental practice (for instance in weekly progress meetings at the laboratory level). The PI will, for instance, gather input from their researchers, as they are closest to the experimental systems used in the laboratory. But this process is not formal and usually does not adhere to democratic principles. It might take the form of consultation or input gathering. But it is always clear who ultimately calls the shots. The direction of a laboratory, and also a research field, is thus set by the powerful top “1%”.

Importantly for the purpose of this paper, the people who are in power are not there because they have been democratically elected. Rather they are where they are because of their “track record”, which includes the papers they have published and the knowledge or expertise they have acquired over time. Science is therefore, in part at least, a meritocracy and epistocracy, at least at the level of individual laboratories or also departments. Professional associations will often allow the membership to elect the president for a set period, injecting a democratic element into the process. But there is usually no opportunity for the membership to select the range of candidates that are put forward. The choice is made by the leadership of the association, and usually only includes well-established researchers who have been PIs for many years.

These undemocratic dynamics can also be observed in the case of HHGE: the majority of the debate about the future direction of this technology is driven by a relatively small group of leading researchers, including Nobel laureates such as Jennifer Doudna or David Baltimore. In a recent article that analyses the evolution of the debate about the ethics and governance of HHGE, Françoise Baylis observes a) that the key players are all powerful and well-established PIs and b) that there is still an unwillingness within this group of key players to engage with deeper ethical questions and to “broaden the tent” (Baylis 2025).

This lack of a broad tent has systemic reasons. Whilst science has epistocratic and meritocratic elements, knowledge and merit are not always the only deciding factors. Race, gender, disability, and socioeconomic status are also important features that affect who can make it to the top (Zivony 2019; Eaton et al. 2020; Swenor et al. 2020; Meeks et al. 2024). Furthermore, connections with the people who already are in power also matter, whether for getting papers published, getting a leadership position, or funding. As mentioned above, candidates for elections are not selected by the membership, but by other PIs. There are therefore also clear elements of an oligarchy present in science, where a relatively small group of people have power over an area, in this case a research group, institution, or a research field.

At the local level of the laboratory or a research institute we can also observe forms of tyranny, as an overwhelming majority of PhD students and postdoctoral research fellows are reporting abusive supervision behaviour by their PIs (Else 2018; Moss and Mahmoudi 2021; Täuber et al. 2022). The abuse of power, i.e., acting like a local tyrant, can also serve as a tool to advance one’s own career (Täuber and Mahmoudi 2022).

These points about racism, ableism, abuse of power, and the advancement of one’s own career also matter for the second point I want to highlight here, namely, the lack of diversity. Science, and especially STEM disciplines, have a general problem with diversity (Botella et al. 2019; Miriti 2020). But at the same time, it has been noted that there is more diversity at the bottom of the hierarchy than at the top. For instance, there is a clear lack of diversity at the level of professor, but much more diversity at the level of PhD students, especially in fields such as biomedicine (Gibbs et al. 2016; Meyers et al. 2018; Olzmann 2020). This is not just diversity in terms of race, gender, disability and socioeconomic background. It can also be a diversity of mindset. Not everyone who embarks on a PhD is determined to stay in academic researchFootnote 19, and not everyone doing a PhD (or science more generally) will simply assume that the “scientific perspective” (Sarewitz 2015) is always the right or best approach. Even among established researchers there is not just one approach or view, as there are often “mavericks” who are set on challenging the status quo of science (Cook et al. 2022).

These points matter for several reasons. First, they matter because strong and deep participation is a tool of direct democracy. It is unclear how such a tool can ever find a home in a fundamentally non-democratic setting. Second, they emphasise that the lived experience of researchers is not one of democratic and inclusive decision-making. Third, there is more diversity within science than is usually acknowledged. When advocates of strong and deep public participation lament the lack of diversity in science, they are thinking of the scientists who lead the field or debate. But we should not take the part to represent the whole. Underneath the narrow layer of the powerful 1% lies a broader and more diverse cohort of the “working class” of science.

4.2 A different way forward?

This suggests an additional path towards deeper and stronger public engagement: changing the internal power structures of science. If there is a flatter and more democratic decision-making structure within science, we will not only have a more diverse group of decision-makers involved, but we will also establish a different culture that helps develop a different lived experience for researchers. Exposing researchers to a different way of decision-making could make them more receptive to having other stakeholders, including laypeople, involved in the process. It has been shown that researchers who are exposed to participatory practices, either by being part of them or because they see colleagues being involved, are more likely to get involved or continue to be involved (Royal Society 2006; Poliakoff and Webb 2007).

To repeat the point, lived experience matters here. In many parts of the world, researchers are not used to a more inclusive decision-making process. In fact, it is often normal that PhD students and technicians, who are all highly trained and knowledgeable individuals, are excluded from key decisions. If it is uncommon for a PhD student to have a role in core decision-making processes, then it is understandable that it seems far-fetched to give a non-scientist any sort of say in how science should be pursued or developed.

There are exceptions to this rule. For instance, students might be involved in the process of hiring new faculty members. Whilst this is rare in places such as the US and the UK, in Germany most universities have a student representative with voting rights on hiring panels.Footnote 20 As with the example of the Alzheimer’s Society and their involvement of “experts by experience”, such islands of inclusive practice can play a crucial role in a broader push for a more inclusive practice in academic decision-making. If we build on these examples and start challenging the exclusive nature of decision-making in academia, we will not only make the jump to deep and strong public participation for issues such as HHGE less extreme, but we will also have a chance of increasing the diversity in the group of decision-makers.

So, what structures could be installed, and what kind of decision-making processes should they cover? First, it has to be stressed that there will be no one-size-fits-all model. As is the case of participatory practices more generally, what works best will depend on the context and the specific aims of the exercise (Scheufele et al. 2021). The question of what this more participatory future could look like is not something that should simply be imposed on those who then have to live in this new reality, i.e., the scientific community, the policy-makers, and the general public, which is funding the research through tax payments.

But there are a few general ideas that could form a starting point for discussions. For instance, we could envision a science-community assembly that includes researchers from all levels in a particular institution, drawn by lot for a set period of time and which either meets regularly or which is set up to deliberate on a specific issue, for instance research related to genome editing. Members could be chosen to serve for one or at most two years to account for the relatively high turnover of members in an academic research setting. The assembly could deliberate on questions of budgeting and the strategic direction of the institution or even individual research initiatives.Footnote 21 For instance, if a PI plans to set up a project related to HHGE this could trigger an automatic deliberation by the community assembly, in addition to the existing ethical review process. The community assembly would not replace Institutional Review Boards (IRBs), as it would not just focus on protecting human research participants (which is often the main role of IRBs) and as it would get involved at an earlier stage—rather than sending a completed research proposal to the internal review process, the community assembly would convene and make decisions when a research idea is first proposed by a fellow researcher. It would thus be a form of a deep inclusive practice.

The power dynamics in such a community assembly would have to be carefully managed. If, for instance, competing PIs had a majority vote they could use the assembly to stifle a colleague’s research agenda. Drawing members by lot and ensuring broad membership could, to some degree, counteract these risks. Also, because this assembly would include technicians, PhD students, and postdoctoral research fellows, the diversity of the decision-makers would likely be increased. By installing such a culture of non-hierarchical decision-making, the ground could be prepared for a more inclusive participatory culture that involves non-scientist citizens.

As the above indicates, this is a highly speculative proposal that has obvious downsides. Apart from the potential abuse of power, it is also not clear how such a re-shaping of internal power structures could come about and whether it would be effective. Regarding the first point: part of the issue is that it is not clear who would have the authority or power to initiate such a shift in how scientific decision-making is structured (my preferred answer to this question would be that this again would have to be an organically grown, bottom-up and co-produced change). But it is also well-known that researchers, and especially early career researchers, already have too many things they need to do in the little time they have available. This goes back to the points mentioned earlier regarding infrastructure and incentives. Researchers are rewarded for publishing papers and bringing in grants, not for sitting on hiring panels or science-community assemblies. If they want to have a career in science, or at least make a living, then they need to focus on their core activity, which is producing research outputs.

However, the incentive structures and the infrastructure of science are not a simple given—they are something that human actors, including scientists themselves, have created and maintained over time. This means they can also be changed. It is now a common claim that academia needs to re-think its incentive structures and move away from a narrow focus on high-impact and high-frequency publications: quality science and scientific qualifications are about more than just data. Calls for such a re-think have especially come from the debates about the so-called “replication crisis” in the experimental sciences, i.e. the idea that most scientific outputs, especially in fields such as biomedicine or psychology, cannot be replicated by other researchers or laboratories (see, e.g., Munafò et al. 2017; Romero 2017; Heesen 2018; Korbmacher et al. 2023).Footnote 22

If a new decision-making infrastructure were installed this could be coupled to such a change in incentive structures. A PhD student, for instance, could gain credit for serving on a panel or an assembly based on the fact that they are developing and/or demonstrating leadership qualities through their activities.

Regarding the second point, i.e., whether such an assembly could be effective: this is an empirical question that we cannot answer at this point. As mentioned in Sect. 3.3, there are examples where a flatter and more inclusive decision-making process has been installed, for instance in the research network of the Alzheimer’s Society. But we will need more research to understand what this success means and what forms of inclusive decision-making will work best within science.

It is also important to highlight that such assemblies would not be an end point or a replacement for public engagement. These assemblies will not provide the level of diversity and dynamics that advocates of deep and strong engagement are asking for. Rather, they would be a tool, a sort of stepping stone on our way towards a more inclusive and deliberative culture within science.Footnote 23

In summary, there is a different path towards stronger and deeper public engagement that includes a re-shaping of the power structures within science. The ground for such a move already seems prepared, courtesy of ongoing discussions about various crises in contemporary science. Turning these crises into opportunities (Fanelli 2018; Munafò et al. 2022; Korbmacher et al. 2023) could help us to change the decision-making culture within science itself, thus paving the way for a more inclusive decision-making process.

5 Conclusions

HHGE is a controversial technology that raises various ethical, social and scientific issues. There is now a broad consensus among scientists and other stakeholders that scientific self-governance is not sufficient in this case; some form of public participation should be part of the decision-making process.

However, as I have shown using the three distinctions between positive/negative; weak/strong; and shallow/deep participation, the public participation that we are witnessing to date is not meaningful: we don’t see any cases of strong and deep participation, i.e., participation that gives the public a true share of the decision-making power and that reaches deep into the decision-making process.

I have then suggested that to properly explain and tackle this absence of meaningful participation we need to pay more attention to the power structures within science itself. Science is a fundamentally non-democratic system that contains elements of an oligarchy, epistocracy, meritocracy, and also a bit of tyranny. As positive, strong, and deep public participation is a tool of direct democracy, the power structures in science are a potential roadblock to realising a more inclusive and meaningful decision-making practice.

One way forward, I propose, is to introduce a flatter and more democratic culture of decision-making within science itself. Realising such a shift will be challenging, and the exact shape and form it takes will have to developed in a process of co-production that involves relevant stakeholders, including taxpayers. But despite the various hurdles such a re-shaping faces, there are indications that such a practice can work, for instance in the context of funding and hiring decisions. Building on these examples, whilst also developing more concrete mechanisms by which such practices can be realised, might allow us to collectively re-think how emerging technologies such as HHGE could and should be governed.