We argue that sustainable approaches to ocean biodiversity governance require researchers to be aware of the histories and legacies of knowledge extraction, imposition, and assumption within their research fields (Belhabib, 2021; Trisos et al., 2021). In addition, ocean biodiversity researchers should actively and continuously strive to acknowledge and address these colonial and inequitable legacies in their work. We acknowledge that this is not an easy task for researchers, even within research that recognises the need to implement paradigm shifts and transformations. Existing knowledge hierarchies have proven to be multi-layered and are often reproduced even when attempting to break down these asymmetrical patterns through approaches that aim to integrate or ‘bring together’ diverse knowledge systems. However, effective change, such as that required to address the intractable problems of marine biodiversity loss and sustainability, will require sensitive and multi-faceted approaches supported politically and financially, to reconstruct the relationship between knowledge and equitable governance. In rising to this challenge, it will be important for researchers to work with, as well as through, such tensions, with a commitment to explicitly face the power asymmetries often cast as invisible within the academy (Haraway, 2016b).

In practice, addressing knowledge hierarchies starts with acknowledging the influence of researcher positionalities. The provision of a positionality statement alongside research outputs is suggested as best practice to make explicit the context within which knowledge is being produced (Trisos et al., 2021; Muhl et al., 2023). As individuals, we are shaped by our geography, race, gender, sex, ability, spirituality, socioeconomic status, discipline, age, and so much more, which again influences our work, our ways of being and knowing (onto-epistemologies), the intended reader (the gaze), and the standpoint we write from (the pose) (Baker et al., 2019; Trisos et al., 2021). Actively considering these influences is a necessary start to the process of breaking down knowledge hierarchies in biodiversity research, and challenging the way we know the world from our own critical thought and reflexivity (Idahosa and Bradbury, 2020).

The authors of this article reject the notion that researchers are neutral actors in knowledge production (Abimbola, 2019; Green, 2020; Trisos et al., 2021) and describe our positionalities at the end of this article. As outlined in Section 7 our positionalities, and thus our understandings of and approaches to biodiversity governance, are influenced by varied geographical contexts, institutions, disciplines and lived experiences. Although many of us work with Indigenous and local communities as co-researchers and collaborators, we do in no way represent Indigenous and local knowledge systems of those we are not directly part of ourselves. The authors speak as representatives of the One Ocean Hub, which is based in Scotland, and various authors also work from, and are located in the Caribbean Island of Barbados, Ghana, Namibia, South Africa, Tasmania and the United Kingdom. However, a majority of authors were educated within Western University institutions, or institutions whose curriculum was founded on Western pedagogy, which will influence the knowledge hierarchies we are exposed to, and consequently impact how we approach ocean biodiversity governance research.

The embedded coloniality of knowledge within ocean biodiversity governance research can be traced to the very beginnings of modern marine environmental sciences. Oceanography, limnology, fisheries science, and related marine sciences emerged as professionalised sciences in the nineteenth and twentieth centuries, becoming institutionalised through the establishment of research laboratories, predominantly in Europe and North America. While first focused on surveying, classifying, and cataloguing the maritime world, often in service of imperial powers which required intimate knowledge of the oceans, these sciences were soon fundamental to proactive attempts to conserve and regulate the exploitation of marine populations and environments (Rozwadowski, 2004, 2013; Hubbard, 2006; Reidy, 2008; Finley, 2011; Reidy and Rozwadowski, 2014; Adler, 2019). This occurred within the context of, and was intrinsically entangled with, Western imperialism and went together with efforts to ‘optimise’ the use of marine environments throughout the world. While histories of marine science have tended to focus on the North Atlantic, where major international organisations including the International Council for the Exploration of the Seas (ICES) were predominantly focused, these developing sciences were also fundamentally shaping—and were shaped by—parallel attempts to order and regulate water bodies throughout both domestic and colonised territories (Anker, 2001; Rozwadowski, 2002; Jennings, 2011; Breitinger, 2022). This encouraged the creation of research institutions across empires staffed with technical experts leading research programmes that were intended to inform the efficient exploitation of marine resources and, especially, fisheries (Jennings, 2011; Breitinger, 2022).

Through this process, scientists became, and continue to be, perceived as an intellectual elite capable of ‘knowing’ nature better than ‘others’. Through methods of data collection and analysis they maintain their position of power as those best placed to understand the natural world. The emphasis of detached objectivity within scientific methods, in which a rational observer investigates the material world without implicating themselves in what is being studied, also works to create a detachment of humans from nature (Morton, 2009). Nature becomes an entity separate from humans, that is a subject for objective study and manipulation (Barad, 2007). This continually reinforces a worldview where humans are separate from and superior to their environment ( (Drayton, 2000; Clarke, 2007; Adas, 2014; Haraway, 2016a, 2016b). Biodiversity, for example, is often assessed in terms of its ecosystem services and the benefits provided to humans like clean air, water, and soil fertility. While this perspective recognises the inter-dependence of humans and nature, it still places human utility at its centre (Muradian and Gómez-Baggethun, 2021). Accompanying this is the tendency in some scientific disciplines to reduce complex systems into more manageable components for easier study. Such an approach has been criticised by scholars in science and technology studies for valuing the efficiency of studying individual components over the interconnectedness of natural systems and humans – which is arguably much more difficult (Keller, 1995; Jasanoff, 2004). This form of scientific objectivity is also used to sustain knowledge hierarchies, where science dominates discourses of nature due to its claim to objectivity in comparison to different ways of knowing that are seen to provide only subjective knowledge or perspectives (Haraway, 2016a).

Fundamental to the contemporary dominance of scientific knowledge is that states have looked to science to inform and reinforce their own claims to authority over the natural world. In doing so, they have helped to enshrine scientific knowledge and, therefore, scientists as the elite voice concerning the natural world (Mukerji, 1989; Gascoigne, 2019; Oreskes, 2021). Governmental patronage has been fundamental to the rising reach and power of natural science in centralised biodiversity governance, just as science has been fundamental to the rising reach and power of governments (both domestically and internationally through imperial and neo-colonial expansion). This is not to suggest that biodiversity research has been beholden to governments or that governments have dictated scientific research agendas. Yet, governmental patronage and funding, especially from the mid-nineteenth century onwards, has played a significant role in shaping both the trajectories of scientific research and in elevating the scientific voice as the central authority on matters of nature (Mukerji, 1989; Gascoigne, 2019; Oreskes, 2021). As Mukerji (1989) and Oreskes (2021) have emphasised in their separate studies of the history of oceanography, the question of who funds scientific research is important as this not only creates opportunities that enable scientists to conduct the work that they want to do but also shapes what research is not prioritised due to the limits of funding. While funding did not confine scientists to specific practical concerns of the state, it did shape the availability of finance for research surrounding subjects that concerned governments.

It is well recognised that the trajectories of science are highly influenced by economic agenda, and the priorities of both funders (Oreskes, 2021) and those undertaking the research (Thorp, 2023). This has played an important role in shaping what we know and do not know about the oceans. This influence sits uncomfortably with the perception that scientific research is autonomous and ‘pure’ particularly since the autonomy of science and scientists is also fundamental to the cultural authority of scientific knowledge. The authority that states seek to exploit to reinforce its own claims of authority over nature.

This is not to question the value or importance of normal science such as the natural sciences, which is critical to informing ocean biodiversity governance, but it is instead to highlight the embeddedness of science within broader governance processes that have constructed knowledge hierarchies to the detriment of other perspectives and knowledge holders. As Mukerji (1989) states, “the power of science lies less in what scientists tell the government than in the cultural authority of science as an institution … it is this cultural authority that makes science so useful to the government.” This is achieved through the process by which governments make use of scientific rationales to inform myriad decisions and policies while, crucially, also justifying these decisions (Mukerji, 1989). In doing so, governments expropriate the voice of science, calling on the cultural authority of the scientific voice to legitimise its own activities. Crucially, there is not a universal scientific voice but, instead, governments can make use of the voices of scientific advisors that align with their agenda while ignoring the voices of those at odds with this. The importance is not in the truth or veracity of these voices but, instead, in their perceived cultural authority and autonomy.

The scientific voice is simplified and ventriloquised – i.e. expropriated and/or reproduced to meet a particular goal – by governments to support and legitimise their decisions over nature. It is this seeming reliance on the scientific voice that provides the justification for governments to act as custodians over nature, as it is governments who have privileged access to this scientific knowledge through their patronage and authority. In the process, the dominant cultural authority of the scientific voice becomes further embedded through its centrality to governmental policies (Mukerji, 1989; Gascoigne, 2019). As it is (some) scientific knowledge that is constructed and exploited as the best possible source of information surrounding the natural world, different knowledges are then neglected or undermined as these are not so readily absorbed within or aligned towards governmental frameworks. This is especially true within the parallel contexts of centralisation and colonisation, in which the scientific voice was regularly used to provide justification for imperial rule through the language of ‘improvement’ and ‘efficiency’ (Drayton, 2000; Tilley, 2011; Ross, 2017; Oba, 2020). Where Oreskes (2021) states that “every history of science is a history both of knowledge produced and of ignorance sustained” it is important to also recognise that there is often a history of knowledge ignored and suppressed too.

Historically, domestic and colonial marine research programmes were led by amateur and professional scientists and relied heavily on the knowledge and observances of marine and coastal communities, and who were best informed of the surrounding marine biology and oceanography (Hubbard, 2006; Jennings, 2011; Jones, 2018; Silver et al., 2022). This was a process in which people who lived with the ocean engaged in intellectual conversation with researchers, leading to the exchange and co-production of knowledge. Yet, these marine and coastal communities were constructed as ‘informants’ in service of scientists and technical experts rather than intellectual agents in these exchanges each with their own realities and perspectives that were deeply rooted in place and experience. The embodied knowledge of these ‘informants’ was then enfolded into methodologies of normal science, scientific reports and recommendations, not only separating this knowledge from its source but also employing this knowledge to inform decisions in forums where the voices of knowledge holders were fundamentally excluded (Tilley, 2011; Jones, 2016; Mavhunga, 2018). Even as they were informing scientific research in both colonial and domestic settings, marine and coastal communities were portrayed by decision-makers as ill-informed and driven by self-interest, so that only ‘rational’ environmental management informed by technical experts and led by governmental agencies could provide for the ‘optimum’ exploitation of marine environments. In some global south contexts, these management systems were conflated with more sinister political racist systems of spatial management excluding communities from their traditional access to the ocean (Francis and McGarry, 2023).

The parameters of what entailed ‘optimum’ usage was also determined by the priorities of domestic and colonial governments, often relating to uncertain or arbitrary targets that were set without the necessary data or even the capacity to collect the data that supposedly underpinned such targets. In the process, governments perceived and constructed themselves as the ultimate custodians and stewards of marine environments, capable of making the most informed and rational decisions surrounding marine environments through a reliance on scientific research (Chirwa, 1996; Drayton, 2000; Clarke, 2007; Jennings, 2011; Breitinger, 2022). This underpinned the construction of systems—legal, political, economic, institutional, and educational—that empowered governments to assume and monopolise the rights to control and make decisions over water bodies (Anker, 2001; Tilley, 2011; Wilson, 2021; Breitinger, 2022). It is important not to overstate the efficacy of these processes, as governments regularly proved unable to manage marine environments in practice so that a pluralism of management regimes and knowledge systems remained—and remains—in practice (Beinart, 2000; Hodge, 2011; Jennings, 2011; Mbatha, 2018).

The development of ‘evidence-based’ ocean biodiversity governance and marine sciences worked to construct and embed knowledge hierarchies within both ocean governance and ocean governance research by the mid-twentieth century whereby marine and coastal communities were excluded from decision-making processes (Tilley, 2011). This enabled governments to tighten their control of natural resources at the same time that “ecologists enlarged their system of global knowledge at the expense of local knowledge” (Anker, 2001). Local knowledge was central to the development of global scientific knowledge of marine biodiversity, but it was extracted, undermined, and relegated in the process. In some cases, their knowledge was weaponised against them, to inform oppressive fortress conservation systems.

Conceptually the definition of biodiversity (previously termed biological diversity) which is most commonly applied in marine science and ocean governance is that provided in the Convention on Biological Diversity (CBD) “the variability among living organisms from all sources including … diversity within species, between species and of ecosystem” (CBD, 1992). The Convention is widely endorsed globally, but the political popularity during the 1990s of the field of conservation biology in the Global North perpetuated the widespread uptake of the term with its explicit goal of conserving biodiversity against increasing human pressure and to promote sustainable use and benefit sharing (Sarkar and Margules, 2002). Diversity in this context is understood to be essential for the resilience of natural systems to change or pressures (Mace et al., 2014). In describing “the variety of life” the term encompasses the full complexity of ecological systems and the relationships between and among organisms from molecules to systems (Sarkar and Margules, 2002). While the CBD intended to promote understanding of the interdependence of humans and nature within governance, in practice, the operational definition of biodiversity remains capricious and is often criticised as poorly encapsulating the ways in which human and non-human actors co-create diverse ecosystems and adapt to changing environmental conditions (Tsing, 2009).

The reverence of science-informed, or ‘evidence-based’ decision-making necessitates that the concept is simplified so that biodiversity can be measured, particularly in marine systems where ecological knowledge is inherently uncertain (Pascual et al., 2010). This simplification is guided by government priorities and applies the use of biodiversity surrogates that represent a simplified measure of biodiversity, to demonstrate changes (Mellin et al., 2011; McQuatters-Gollop et al., 2019). The definition of biodiversity surrogates creates “new natures” (Sullivan, 2013) that allow for biodiversity to be ‘seen’ within existing governance (and economic) frameworks and meet the needs of users in terms of ‘evidence’ (Robertson, 2006). In operationalising biodiversity, these practices determine commensurability to support exchange and substitutability that legitimise its use and degradation through compensation and offsetting or to further accumulation so actors can capitalise on “opportunities of environmental crisis and conservation” (Sullivan, 2013; Sullivan and Hannis, 2017). In doing so, this use of science to justify biodiversity’s use distances the commodity being traded from its physical form and all its inherent complexity and does not readily recognise the “inextricable link” between biological and cultural diversity (Cocks, 2006; Robertson et al., 2014).

The term “biocultural diversity” has been used more frequently to denote this link, primarily in reference to Indigenous traditional communities, but there is potential for its broader application within biocultural discourse (Cocks, 2006). Literature suggests that social groups should be able to choose and define their own bio-cultural diversity (Belay, 2012) and that this flexibility is important to appropriately incorporate cultural and religious values into biodiversity conservation and research and close the growing divide between biodiversity and large sections of society (Escobar, 1998; Cocks, 2006).

Despite a broad understanding of the ocean as a common resource i.e. a “shared resource [with] a defined community that devises protocols, norms and values to manage it” (Rudolph et al., 2020), ocean governance is inherently fragmented by sector. The framework of the 1982 United Nations Convention of the Law of the Sea (UNCLOS) has been the primary forum to address integration of ocean governance, supported by the CBD (1992) and the other Rio Earth Summit instruments (UNGA, 1992), which promote an ecosystem approach for sustainability that embedded the varying ways that people valued nature (Morgera and Razzaque, 2017). Despite this, fragmented and exclusive forms of governance of the ocean have continued to marginalise those most vulnerable to biodiversity loss, which has continued on a trajectory of decline (IPBES, 2019; Tessnow-von Wysocki and Vadrot, 2020; Tolochko and Vadrot, 2021; Erinosho et al., 2022).

In recognition of these trends of decline, calls for the transformation of ocean governance to meet the interlinked demands of sustainability and justice have increased (Strand et al., 2022a; Lombard et al., 2023). The United Nations Conference on Sustainable Development in Rio de Janeiro in 2012 set out a pathway to do this, which finds its most recent expression in the Sustainable Development Goals (SDGs) and yet have still been limited in their influence (Biermann et al., 2022). Another notable action is the United Nations Decade of Ocean Science for Sustainable Development that prioritises the co-design and co-delivery of solution-oriented research and recommends co-design of “transformative science carried out by diverse actors will contribute to sustainable development” and promotes a spirit of inclusivity and openness for transformation through knowledge production and use (IOC-UNESCO, 2021). In support of this transformation, the Ocean Decade has a cultural framework programme that recognises Indigenous knowledge as important to transform ocean sciences and conservation, and encourages support for action (Febrica, 2023).

Furthermore, in the recent Agreement under the United Nations Convention on the Law of the Sea on the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction (the BBNJ Agreement or High Seas Treaty) (UNGA, 2023), a mandate to respect, promote and consider the rights of Indigenous Peoples and local communities “when taking action to address the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction” was outlined to cut across all aspects of the Agreement (Article 7 of the BBNJ Agreement). Whilst current ocean biodiversity governance still suffers from a legacy of sectoral division, its evolution to one that reflects the multiple values held by society and effectively supports biodiversity health is increasingly on global political agenda.

It is widely acknowledged in the literature that bridging or integrating knowledge systems is not easy, however, the challenge of ensuring such knowledge is appropriately considered and incorporated are often less appreciated or acknowledged. Recognising these challenges is important as optimistic visions of knowledge integration can obscure differences between actors and reproduce hierarchies between scientists and local communities in the negotiation of practice and policy (Nadsady, 1999). It is important that technical questions of compatibility and integration do not diminish focus on the ethical and political dimensions of integration practice (Nadsady, 1999; Green, 2009).

The value of Indigenous and local knowledges, including the context, values, and cosmological context within which that knowledge sits (Brosius, 2006), is increasingly recognised as essential for the sustainable management of social-ecological systems. However, there are concerns that in many cases knowledge integration can form a box-ticking exercise and perpetuate historic trends of marginalisation (Wohling, 2009; Green, 2012; Rivers et al., 2023). Research methods often do not involve knowledge holders in research design or analysis and are rarely considered as researchers pursuing their own research questions informed by their own world views (Chilisa, 2017). There is also a tendency to transpose Indigenous and local knowledges into the norms of Western-oriented natural science to fit with demands of Governments (Rivers et al., 2023). Such processes commonly aggregate, abstract and alienate knowledge from its holders (Berkes et al., 2000; Latulippe and Klenk, 2020) and assume a lack of interest of knowledge holders in outputs (Mosimege, 2017; Muhl et al., 2023).

Hierarchies are further imposed via demands for the ‘proof’ of credibility and validity, or even existence, of Indigenous and other diverse knowledge types to science and decision-making before inclusion and integration (Bohensky and Maru, 2011). This disparity is even evident within recent legal developments such as the BBNJ Treaty where consideration of Indigenous knowledge is caveated to apply only “where relevant” (UNGA, 2023). This is despite the often-foundational importance of these knowledge forms for the wellbeing of the communities within which they are held (Latulippe and Klenk, 2020). Demands of proof are rarely placed on natural or normal science from holders of various knowledge types (Moller et al., 2004) and this disparity highlights the often-supplementary status afforded to such knowledge in practice.

Ocean governance has evolved from a long history of colonisation and the interconnected development of normal marine science, and Western- and natural science-dominated biodiversity conservation. Western-oriented knowledge systems are seen as rational and objective approaches to environmental (and ocean) governance, with biodiversity framed within this as a global “asset”. The term “biodiversity” has evolved out of spatial-temporal and socio-political relations and is not universally applicable, and can be considered elitist, due to conflated relationships with settler colonialism and conservation (McGarry, 2023). Local, contextually rooted, knowledge remains central to the development of marine biodiversity knowledge. Yet such knowledge continues to be extracted from local places (and cultures) and relegated via processes of transposition to meet the demands of governance forums dominated by Western-orientated science that seek to base decisions on quantified and statistically certain information. This has led to abstract understandings of biodiversity, and fragmented governance approaches that, in a rejection of a post-normal framing (Funtowicz and Ravetz, 1994), do not adequately manage human impact on the ocean’s biodiversity.

Despite an officially recognised definition of biodiversity within governance at a global level, the case studies presented illustrate the varied ways through which biodiversity is conceptually engaged both within research and governance. These modes of engagement are influenced by the location and context of the research and researchers and:

Our case studies indicate that although the term “biodiversity” is not always explicitly referred to by communities, the importance of social-ecological systems in different cultural settings is common across all contexts. This finding highlights how the term “biodiversity” itself can perpetuate hierarchies if its use overshadows perspectives and concerns that are implicitly connected and/or depend on biodiversity yet do not directly engage with the term. Culturally sensitive, innovative and transdisciplinary approaches to research, like those described in the case studies, allow for the inclusion of different priorities and values in research and research-related outputs. In creating space for a more inclusive expression of priorities and world views, such approaches render these values visible to global governance forums and in doing so support their appropriate integration. The authentic co-design, co-development and co-delivery of innovative research can help dismantle hierarchies that currently exist in knowledge creation.

The case studies presented underline how the structures provided for the production, assimilation and visibility of knowledge are highly influential in addressing knowledge hierarchies. Common across all case studies was the creation of space for the inclusion of all voices to bring the various ways that humans ‘value’ biodiversity into relief, to demonstrate not only that these values exist but their legitimacy and significance to decision-making under existing and emerging international obligations. The various methodological approaches applied to this end are described within case studies and range from arts-based (Erwin et al., 2022; McGarry, 2023) to deeply sociological (Mbatha, 2018; Francis and McGarry, 2023; McGarry, 2023), legal (Shields et al., 2023; Strand et al., 2023) and to natural capital approaches (Niner et al., 2024). These methods present a spectrum through which often obscured or ignored relationships with biodiversity are brought to light. Also demonstrated is how simply providing the opportunity to contribute a view, or have your voice heard on a matter (particularly for marginalised groups) is only a very small part of a meaningful response to breaking down entrenched knowledge hierarchies. Invisible power dynamics are highly influential even when the right people are brought into the room. Accordingly a visible platform is the very basis of what is required, this should be accompanied by a set of research relations that acknowledge historical and ongoing injustices, that elevate marginalised knowledges, ensure participatory parity (Bozalek et al., 2020), that value multiple ways of knowing (Strand, 2023), and that are not primarily focused on science and research as a professional endeavour. Furthermore, the differences between these case studies, including those where methodologies and philosophical groundings sit uncomfortably against each other, are a stark reminder that there will never be a single methodological approach through which to effectively address knowledge hierarchies. Despite often universally common issues arising from and perpetuating knowledge hierarchies in marine biodiversity governance, different contexts will require different approaches sensitive to the context within which research is taking place (Temper et al., 2019).

As the above case studies demonstrate, there are multiple knowledge systems that need to feed into ocean biodiversity governance. This, of course, is not a new finding. Over the past three decades, there have been increasing calls for greater engagement of different knowledge systems and worldviews across various aspects of ocean governance that are implicitly or explicitly linked to biodiversity (Díaz et al., 2018). Yet, marine scientists and policymakers regularly struggle to engage with knowledges that exist outside of the structures of conventional Western-orientated science. This includes difficulty engaging with varying disciplinary perspectives drawn from the arts and humanities as well as with the knowledges of ocean users and communities. Part of the reason for this is the way in which these sciences evolved following what Reid et al. (2021) refer to as a “utilitarian worldview where humans are in control of nature,” which is intrinsically connected to processes of colonisation and has come to be characterised by claims of objectivity that are able to reduce systems-to-be-governed to quantitative models created, controlled, and interpreted by specific groups of experts and governors (Jentoft and Chuenpagdee, 2009; Aguado et al., 2021; Silver et al., 2022).

While scientists acknowledge the uncertainty in the statistical models applied for ocean biodiversity governance, these are commonly seen to be based on the “best available evidence” of the time, which can be refined and verified as new data arises. Such evidence, however, largely refers to cultivated ‘traditional’ scientific data (Reid et al., 2021). Historical perspectives show that scientists have regularly drawn from and relied on local ecological knowledge holders to inform their understanding of marine spaces. Accordingly, the perceived value of today’s “best available knowledge” is founded on the unrecognised contributions of knowledge holders outside of the field of science. It is important that such history is explicitly recognised as it contests binary thinking around scientific and other knowledge-making traditions, which were (and are) not always as separate as is imagined through concepts of ‘modern’ and ‘traditional’ marine governance. The problem in the past has not been that scientists are unwilling to engage with various forms of knowledge or to co-produce knowledge with ocean communities. Instead, it is that such co-produced knowledge has been folded into scientific reports and models that have been used to inform the creation and imposition of laws and policies over ocean and ocean users by external centralised orders without providing the space for input from situated knowledge holders in how their knowledge is used or what decisions are made on the basis of their knowledge (Tilley, 2011; Osseo-Asare, 2014; Mavhunga, 2018; Wilson, 2021).

More recent moves to ‘integrate’ other forms of knowledge beyond traditional scientific and technical data have in large part continued with this trend, in which knowledges that do not fit the ‘normal’ science mould have been taken and reshaped in attempts to fit into such moulds. This has reduced such knowledges to ancillary data to better inform scientific models and approaches for problems and challenges already set and defined by governmental or scientific research agendas (Walley, 2004; Etiegni et al., 2020). Such approaches can fail to acknowledge non-traditional scientific knowledge as being more than ecological knowledge, centred on place-based issues with specific contexts, histories, challenges, experiences, and interests. Extracting such data to integrate and ‘scale up’ into regional or national models or policies not only simplifies and homogenises such knowledge, smoothing the edges to fit more neatly within scientific models or to meet the specific criteria set by scientists or policymakers, but also fails to recognise the ‘living’ and ‘social’ nature of such knowledge and the ‘hybrid’ realities that these represent on the ground. Case study 3 which describes the development of a framework to include a holistic and varied range of biodiversity values of ABNJ shows how rich detailed description of these values and how they vary at a community, regional or national level is missed when representation at an International scale is required. While the framework developed can include and consider all available information, when decisions are required of the International ocean and the relationships of all society are needed to be captured that included will invariably be limited and therefore skewed to that which is easily accessible (as supported by the arts-based methods applied in case study 5) or more readily described. Case studies 1, 4 and 5, in applying research to a more specifically defined scale can offer this detail and clearly demonstrate how knowledge is held and lived by its holders. Such knowledge is not static or homogenous but instead is influenced by a range of effects and transformations that cannot be easily reduced to stereotypes of ‘modern’ or ‘traditional’ knowledge; instead, it is a mix of diverse knowledges that have been co-produced on the ground as communities, environments, technologies, economies, and governance structures have transformed over time. This then provides not only ecological knowledge but also context-specific economic, social, cultural, and political knowledge (Walley, 2004; Williams, 2017; Fischer et al., 2022; Msomphora et al., 2022). Such knowledge cannot be easily reduced and integrated into scientific models (often via quantification) that form the basis of decision making but, instead, constitutes an entire web of different knowledges that should not only inform decisions but that should also inform how problems are framed, what knowledge needs to be produced in order to address such problems, and to what end (i.e. who benefits) (Brugnach, 2017; Fischer et al., 2022).

Co-production provides one approach towards this more active engagement of different knowledge systems across research. As Cooke et al. (2021) describe knowledge co-production:

When approached meaningfully, co-production offers clear scope to challenge knowledge hierarchies by providing the space to collaboratively set objectives and research activities while also interpreting the data from different perspectives and using this to inform the recommendations arising from the research (Maxwell et al., 2020; Cooke et al., 2021; Reid et al., 2021). While co-production is most often seen as a means to bring together and negotiate across distinctive worldviews and values through engagement with diverse communities and stakeholders (Maxwell et al., 2020; Reid et al., 2021), the above case studies also demonstrate that knowledge co-production is also essential in bridging the hierarchies within academic discourses too. Co-production and transdisciplinary research methods are not just complementary but should be seen as intrinsically entwined. Transdisciplinary research cannot bridge the knowledge-action gap without engaging in forms of co-production between knowledge holders, stakeholders, researchers, and policymakers. While meaningful co-production requires a range of researchers whose knowledge includes different and distinctive approaches towards collaborative data collection, includes participatory approaches to interpretation, and participative parity in making claims from said research.

It is important to emphasise that co-production and transdisciplinary research approaches do not offer a panacea to challenging knowledge hierarchies. Instead, knowledge hierarchies can be both challenged and reproduced through such approaches. Decisions need to be made on what issues to concentrate on (Latour, 2004); how to approach such challenges, and how to act on, interpret, and disseminate research findings; all of which can challenge certain knowledge hierarchies while reinforcing others (Weber et al., 2017). In order to arrive at any negotiated consensus, difficult decisions need to be made that will inevitably lead to the sacrifice of different interests, desires, and goals (Jentoft and Chuenpagdee, 2009; Weber and Stevenson, 2017; Aguado et al., 2021). It behoves society to ensure this sacrifice is not only or even primarily asked of the most marginalised and vulnerable communities. Co-production and transdisciplinary research offer an approach to navigate such sacrifices through transparent processes and collective decisions based on diverse knowledges so that decisions made are genuinely shared and consensual among stakeholders rather than imposed and unilateral. This requires keen and attentive vigilance to how power, access and participation are realised in decision making. As the above case studies demonstrate, this is an iterative and continual process that requires substantial time, trust, and resources (Gutiérrez et al., 2011; d’Armengol et al., 2018; Etiegni et al., 2020; Maxwell et al., 2020; Cooke et al., 2021; Reid et al., 2021; Silver et al., 2022).

There are several theoretical and practical obstacles in pursuing research that adequately breaks down knowledge hierarchies in marine biodiversity knowledge production and governance. As the case studies and above discussion shows, this includes aspects such as how biodiversity is understood and conceptualised, knowledge co-production processes that consider power structures and historical inequalities in representation and influence, and more practical limitations of funding, time and institutional support. In this section we delve deeper into a discussion on the importance of considering discomfort, conflict and failure in marine biodiversity research to ensure that projects, even those with the best intentions, cease to reinforce or perpetuate existing knowledge hierarchies.

Idahosa and Bradbury (2020) emphasise the role of discomfort in limiting the risk of perpetuating colonial or Western knowledge hegemonies in knowledge production processes, arguing that we as researchers can and should ‘utilise emotions of paralysis, discomfort and contradiction towards positive social change’. By critically reflecting on our own positionalities and how these may influence our behaviours and attitudes towards people and ways of knowing different from our own, we can better ensure that discomfort becomes productive instead of paralysing (Idahosa and Bradbury, 2020; Leavy, 2020). In practice, embracing discomfort and reflexivity means acknowledging how our inherent biases shape the research process and experience, and can involve questions such as ‘am I the right person to be researching this?’, ‘what knowledges am I promoting or elevating through my arguments?’, ‘am I claiming to represent anyone else but myself through my work?’, ‘am I comfortable with acknowledging that I am wrong?’ ‘Have claims been validated by participants including knowledge holders and co-researchers, before finalising them?’ These questions support a reflection of whether we are, consciously or unconsciously, devaluing different knowledge systems through our research. For example, trying to integrate the spectrum of values held within societies into marine biodiversity governance that continue to devalue, disregard or refute approaches to co-existence and rights to self-governance (Rivers et al., 2023).

Co-production and transdisciplinary research approaches ultimately emphasise a coming together of different knowledges, which attempt to address ongoing tensions and power imbalances in order to arrive at some form of negotiated consensus between stakeholders that can address the knowledge-action gap within ocean biodiversity governance (Gómez and Köpsel, 2022; Strand et al., 2022a). One of the greater challenges of transdisciplinary research, therefore, is navigating various ontologies and epistemologies (Raymond et al., 2010; Moon et al., 2021), where ontologies refer to ‘how one perceives reality and the world’ and epistemologies refer to ‘how one values something to be true or valid’ (Strand et al., 2022b). Even within this group of authors, we have diverging disciplinary and contextual backgrounds that influence how we perceive and understand knowledge hierarchies in biodiversity research (see Supplementary information ), and this can often result in conflict and disagreement. Despite the apparent need for conflict resolution mechanisms and training in navigating disagreement in research, there is limited support in academic institutions to prepare researchers for these realities, and institutional and academic systems do not adequately promote this reflection.

However, as we experienced in our own discussions in theorising and co-writing this paper, we should not try to ignore or distort the tensions, conflict, and emotions that arise in the process of such co-production. Doing so ignores how working with the multi-layered tensions surrounding even specific concepts – such as the very idea of ‘knowledge hierarchies’ – can reveal the implicit ambiguities and uncertainties at the core of complex issues. These have too often been ignored or brushed over in the past in the name of efficiency or to project findings that appear more definitive or objective than they are in reality; exactly the process through which knowledge hierarchies are often constructed in the first place (Knowles, 2023). Instead, such tensions need to be mediated while ambiguities and uncertainties need to be openly acknowledged, accepted, and disseminated by all involved in knowledge co-production (Galafassi et al., 2018; Strand et al., 2022a). This moves away from claims to universal truths to instead acknowledge that imperfect decisions must be made on imperfect data, mediating tensions and unequal power relations through engagement of stakeholders and knowledge holders in data collection, analysis, and decision making (Manuel-Navarrete et al., 2021; Moon et al., 2021; Trisos et al., 2021). This accepts that there are only “clumsy solutions” to “wicked problems,” but that such “clumsy solutions” are all the more clumsy if they are not generated through (honest) dialogue between a diverse group of knowledge holders interpreting a wide range of data from distinctive perspectives, worldviews, and values (Weber et al., 2017; Temper et al., 2019; Moon et al., 2021; McGarry, 2022). Ultimately, we echo Haraway’s (2016b) call to “stay with the trouble” and resist the temptation of easy solutions or escapist fantasies in the face of complex, multi-faceted problems. Adopting an iterative, ongoing ethic of care, can foster meaningful and robust responses that support comfort with such reflection and promote approaches to biodiversity governance that is grounded in empathy.

Closely linked with the importance of embracing discomfort and conflict and making explicit the fallibility of science, is the need to consider failure in marine biodiversity research. Failure is a critical aspect of any research endeavour, but scientists are often discouraged or even stigmatised from openly discussing this in their work (Cvitanovic et al., 2022). Assessing conservation project publications, Catalano et al. (2019) find that it is rare to find reports of failure, despite the opportunities these could create in providing a repository of lessons learned for future research. Similarly, Ounanian (2021) discusses the ‘the collective tendency in (social) science to conceal the imperfections of field research’ and aims to portray the often-overlooked failure and redesign of marine social science processes. The limited attention to failure can also be extrapolated from the case studies presented in this paper (Section 4), and the authors feel we are usually encouraged to report on our successes and impact but rarely asked to expand on our failures and shortcomings. Reflecting on the importance of normalising and learning from failure in marine social science, Cvitanovic et al. (2022) provide several considerations and actionable steps to better navigate and overcome failure. This includes communicating uncertainties prior to fieldwork, sharing emotions during research collaborations and even ‘phone a friend’ to discuss feelings of stress and pressure, and debrief with collaborators as a group to process aspects that may have gone wrong or not according to plan (Cvitanovic et al., 2022). Enabling and even promoting humility within scientific practice through an openness to failure, not only strengthens science but also invites question and critique from those outside of the ‘black box’, a process essential to achieve appropriate, fair, and effective integration of knowledge.