Advertisement

Sustainability Science

, Volume 14, Issue 3, pp 843–856 | Cite as

Local knowledge, global ambitions: IPBES and the advent of multi-scale models and scenarios

  • Noam ObermeisterEmail author
Original Article
Part of the following topical collections:
  1. Concepts, Methodology, and Knowledge Management for Sustainability Science

Abstract

In 2016, the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) published its first methodological assessment report on scenarios and models, identifying important gaps in the literature. IPBES has since then moved onto Phase 2, namely a commitment to build on the assessment findings to catalyse the development of the next generation of multi-scale models and scenarios for biodiversity and ecosystem services. Part of that commitment involves the inclusion of Indigenous and Local Knowledge (ILK) in those models and scenarios. IPBES is both an institution (with its governance structure, work programme, deliverables, and so on) and a network (with its member states, authors, stakeholders, and readership). Within that network, the methodological assessment report can be said to be ‘performative’, ergo playing a significant role in shaping engagement and research pathways in the years to come. Within the social sciences, this paper marks a first attempt at evaluating some of the potential challenges of Phase 2—with specific regard to the inclusion of ILK—and strives to generate more engagement from social scientists and humanities scholars on this issue. I combine in-depth expert interviews with document analysis and focus on the ideas of ‘scale translation’ and the translation of ILK into quantitative data—which I contend are likely to be the most contentious and arduous aspects of ‘integration’. I conclude that while IPBES is on track for leading the research community away from IPCC-type global, panoptic models and scenarios, a more honest and genuine dialogue between natural scientists, social scientists, and ILK holders is still required—so as to better communicate what may be (scientifically) feasible and (politically) acceptable.

Keywords

Models Scenarios Biodiversity and ecosystem services IPBES Indigenous and local knowledge Transdisciplinarity 

Introduction

‘In terms of scenarios and models, [indigenous and local] knowledge is crucial in order to accommodate fundamental aspects of day-to-day life and cultural complexes that also encompass language, systems of classification, resource-use practices, social interactions, ritual and spirituality’ (IPBES 2016a: 26). So reads one of the concluding statements of the first chapter of the Methodological assessment report on scenarios and models of biodiversity and ecosystem services. The Intergovernmental Science-Policy Platform for Biodiversity and Ecosystem Services (IPBES) approved this assessment at its fourth plenary in 2016, as part of Deliverable 3c of the work programme (2014–2018) (IPBES 2016a). The Methodological assessment report (MAR) marked the first coordinated, global attempt at unifying diffuse and divergent practices of modelling and scenario building for biodiversity and ecosystem services (BES). In its ambition to create an ‘institutional epistemology inclusive of diverse ontologies and different ways of “knowing” biodiversity’ (Borie et al. 2015: 14) and move beyond assessments, IPBES has committed to catalyse the development of the next generation of multi-scale models and scenarios for BES (Phase 2)—including the integration of Indigenous and Local Knowledge (ILK) at all scales and in all aspects of its work (Lundquist et al. 2017; Stenseke and Larigauderie 2017; Thaman et al. 2013; UNEP 2013).

Global environmental assessments (GEAs), such as IPBES, are important sites of study for at least three main reasons: (i) they are both political and scientific in nature; (ii) they strive to exert significant amounts of epistemic authority; and (iii) due to the political and non-generative nature of their work, they have been sites where inclusivity and knowledge integration have increasingly become requisites (Beck et al. 2014; Brosius 2006; Gustafsson and Lidskog 2018; Jabour and Flachsland 2017; Turnhout et al. 2016). GEAs have also been described as ‘sites of knowledge production’ which can help us understand ‘how relations between science and policy, between “local and global”, and between different epistemic cultures, are being negotiated and ordered’ (Borie et al. 2015: 6). GEAs are sites of knowledge dissemination and circulation. Indeed, IPBES can be conceptualised as an institution—with its staff (Secretariat), governing bodies (Bureau, Multidisciplinary Expert Panel), direct stakeholders (Member States) and its products (mostly assessments to date)—but also as a network: a loosely defined community of experts, members and indirect stakeholders. I share the view, then, that IPBES’s outputs are ‘performative’ (Lofmarck and Lidskog 2017) and hence—provided they deem influential—are likely to (re)draw and (re)order pathways of future research within the IPBES network and beyond.

From the experience of the Intergovernmental Panel on Climate Change (IPCC), we have witnessed how models and scenarios can be powerful epistemic devices that define desirable or undesirable ‘putative futures’ and, in so doing, restrict the scope of actions that can be taken to promote or avoid them (Mahony and Hulme 2016). It is precisely for these reasons that Beck and Mahony (2017) recently called on the IPCC to take greater responsibility for the ‘politics of anticipation’ that their scenarios produce. In the case of climate change—with the advent of satellite and remote sensing technology—increasingly global, integrated models such as General Circulation Models (GCMs) have offered a ‘global view […] conducive to notions of “planetary management” and the centralisation of power’ (Liftin 1998, p. 210). Global climate models and the institutions which have endorsed them (e.g. the IPCC) have been central to the reconceptualisation of climate as a global, ‘ontologically unitary system’ (Miller 2007). They have given credibility to a particular politico-scientific narrative: climate change is a global problem that requires global solutions (Pearce et al. 2018).

In view of these concerns, I propose the term ‘global kinds of models and scenarios’ (based on ‘global kinds of knowledge’, Hulme 2010) to describe models and scenarios that depict such a global, panoptic gaze, and do not allow for a diversity of more local socio-ecological perspectives. IPBES’s vision and mandate point to a divorce from such practices, yet the practice of making multi-scale BES models and scenarios is still very much ‘science in the making’ (Kok et al. 2017; Lundquist et al. 2017; Rosa et al. 2017). While there is a wealth of literature on methods for participative modelling and scenario-building (e.g. Oteros-Rozas et al. 2015), on the one hand, and on the integration of non-scientific knowledge systems in scientific assessments (e.g. Tengo et al. 2017), on the other hand, there is very little common understanding on how these two worlds interact. IPBES will need to ‘identify effective participatory tools and processes that can bridge diverse knowledge systems in scenario processes’ (Kok et al. 2017: 180). This includes finding ways to integrate ILK beyond the local—something that seems paradoxical given the place-based nature of ILK (Lofmarck and Lidskog 2017). I deem it necessary, here, to highlight the distinction between the current and future work of IPBES assessments (which is unlikely to involve new models and scenarios), and IPBES’s ambition to mobilise and encourage scientists to develop multi-scale BES models and scenarios, inclusive of ILK—which I will be focusing on. To this end, a workshop was recently held in Auckland, New Zealand, in September 2017 (cf. Lundquist et al. 2017 for the report).

At this point, I should also clarify what I mean by ‘knowledge integration’. The term ‘integration’ is certainly not without its problems. Scholars have chiefly been concerned that knowledge integration is too often understood as the ‘integration of indigenous and local knowledge into science, in accordance with scientific principles, criteria and validation processes’ (Thaman et al. 2013: 57), implying an asymmetrical relationship of power (Agrawal 1995; Klenk and Meehan 2015) and often leading to the scientisation—and hence the compartmentalisation and depoliticisation—of ILK (Brosius 2006; Filer 2009; Lofmarck and Lidskog 2017; Sutherland et al. 2013). Here, I suggest that knowledge integration should be understood as a gradient from asymmetrical incorporation to genuine co-production (i.e. conversation and cross-pollination on an equal standing). In that sense, despite some of its negative connotations, integration may be used more freely to describe the activity of bridging, weaving, or collating non-scientific knowledge systems with scientific knowledge, at different scales. For ILK, I adopt the IPBES definition: ‘social and ecological knowledge practices and beliefs pertaining to the relationship of living beings, including people, with one another and with their environments. Such knowledge can provide information, methods, theory and practice for sustainable ecosystem management’ (IPBES 2018).

In the following section, I contend that knowledge integration in multi-scale models and scenarios inevitably involves translation between knowledge systems and across scales. By analysing ‘what is lost in translation at these border crossings’ (Fujimura 2011: 65), we can begin to evaluate whether IPBES is likely to succeed in paving the way for truly multi-scale models and scenarios. Specifically, the paper aims to (i) discuss the problems of translation with specific reference to the quantitative/qualitative and local/global divides and (ii) explore some of the wider, relevant issues with knowledge integration in both research and assessment contexts, as well as offering a view on knowledge brokering. In the discussion and conclusions, I review some the interviewees’ ideas in greater depth and propose that, in order to sustain the journey away from ‘global kinds of models and scenarios’, IPBES and its wider community of practitioners will need to foster a more open and honest dialogue on what can be achieved, and to manage expectations accordingly.

Conceptual background

Taken in its most literal sense, translation entails the translation from one language to another. Far from being straightforward, this can be a complex issue in GEAs and, more often than not, non-English speakers and literature get excluded from the assessment (Kovacs and Pataki 2016; Thaman et al. 2013). Translation is also used more broadly to describe ‘interactions between knowledge systems […] to enable mutual comprehension of the shared knowledge’ (Tengo et al. 2017: 18). Here, translation is taken to mean the process by which a particular knowledge claim, system, or culture is modified in order to become comprehensible and legible for another knowledge system. As Latour (1987, 1999) pointed out translation can also act as an ordering device bridging the gap between ‘world’ and ‘words’ through a complex chain (or sequence) of transformations and references. Significant attention must then be paid to the ways in which knowledge ‘circulates’ (Goldman and Turner 2011), to the role of ‘language, metaphors, and analogies’ (Fujimura 2011: 70), to the various epistemological transactions (or trade-offs) involved and, most importantly, to what gets lost in that process. In the case of ILK and BES models and scenarios, translation may entail, for example, the translation of qualitative claims into quantitative pathways, or the translation of ‘worldviews’ into particular model parameters (e.g. De Vries and Petersen 2009). Such translation may very well lead to the scientisation of ILK (Lofmarck and Lidskog 2017)—an outcome that is likely to be dissatisfactory for some of the ILK holders and social scientists involved in IPBES (Stenseke 2016).

The other aspect of translation that requires attention here is the translation of local knowledge into global models and scenarios. In this case, translation is contingent on the mobility of ILK: ‘to become part of the assessment, all knowledge forms have to travel away from their original context and this travel implies a translation which affects different knowledge forms unevenly’ (Lofmarck and Lidskog 2017: 25). With models and scenarios, that may well involve the abduction and/or abstraction of ‘deeply contextual’ knowledge (Lofmarck and Lidskog 2017: 25). As Turnhout et al. (2016) explain, ‘this process of scaling-up consists of multiple steps of translation during which fragments of knowledge get modified and gain new meanings’ (p. 65). To describe this process, I borrow the term ‘scale translation’ from the Guide on the production and integration of assessments from and across all scales (IPBES 2016b). In its original sense, scale translation is a scientific challenge which may involve either upscaling or downscaling. It is particularly challenging due to the fact that the weight and importance of individual variables will change at different scales (IPBES 2016b). In the case of integrating ILK, I suggest that scale translation will also include the upscaling of ILK to the regional or global level.

Methods for data collection and analysis

The main data source for this paper was in-depth, semi-structured expert interviews. Experts can be seen as ‘gatekeepers of knowledge’ in two ways: they possess intrinsic knowledge on their area of expertise and they have knowledge of the daily intricacies of the institutions and epistemic communities within which they work (Bogner et al. 2009; Bogner and Menz 2009). Experts can act as helpful proxies in reconstructing the socio-political context of the sites they interact with. In that case, ‘attention is drawn to the embeddedness of the expert in circumstances and milieus’ (Meuser and Nagel 2009: 25). As such, the expert interview can generate high quality data on both the topic at hand and the institutions in which they are embedded. For this project, most of the interviewees were selected based on their direct or indirect involvement with Deliverable 3c. Whilst the interviewees have varying familiarity with participatory models and scenarios work, all of them have worked with ILK and are active members of the IPBES network. I conducted 10 semi-structured in-depth interviews lasting an hour or longer. Of the interviewees, 6 are natural scientists, 4 are social scientists; 6 were working specifically on models and scenarios at the time of the interview; 3 are members of the ILK taskforce; and 3 were lead authors of the Methodological assessment report (MAR) (cf. Table 1 for further details).
Table 1

List of interviewees with their respective disciplinary background and role in IPBES

Interviewee

Disciplinary background

Gender

Role and membership in IPBES

Member of phase 2 group

At New-Zealand workshop

A

STS scholar

F

Lead author of the GA

No

No

B

Ecologist

M

Member of ILK taskforce

No

No

C

Ecologist, ex-modeller

M

Contributor to ECA

No

No

D

Land-use expert

F

Lead author of the LDRA & member of TSU on Models and Scenarios

Yes

No

E

Ecologist, modeller

M

Lead author of the MAR & member of TSU on Models and Scenarios

Yes

Yes

F

Sustainability scientist

M

Lead author of the MAR

No

Yes

G

Human geographer

F

Member of ILK taskforce

No

No

H

Political scientist

F

Member of scenario and model group

Yes

Yes

I

Anthropologist

M

Lead author of the GA & member of ILK taskforce

No

No

J

Marine ecologist, modeller

F

Lead author of the MAR

Yes

Yes

GA global assessment, ECA Europe and Central Asia Assessment, LDRA land degradation and restoration assessment, TSU technical support unit, MAR methodological assessment report on scenarios and models

To supplement the interview data, I analysed two IPBES documents of relevance to the topic at hand: the full Methodological Assessment Report on Scenarios and Models for Biodiversity and Ecosystem Services and the Guide on production and integration of assessments from and across all scales (working document). I take the view that ‘the available documents follow strict protocols of recording and cross-referencing, which create a “web of texts” that is considered largely impenetrable to those outside the established processes’ (Montana 2016: 8)—hence the importance of interviews. Nevertheless, ‘the textual output of IPBES is performative in the sense that it will likely be of key influence for the continued processes’ (Lofmarck and Lidskog 2017: 25). Moreover, documents in international politic–scientific contexts also reflect and drive particular practices and political outcomes (Mahony and Hulme 2016). By combining the textual output of IPBES and the data from the interviews, I was able to get a better sense of the direction in which IPBES was headed and identify some of the key challenges ahead. The approach to the data was largely inductive. Key themes emerged from an inclusive qualitative coding exercise. These key themes were then broadly organised into the following sections. The analysis was written with the intent of putting the different interviewees in conversations with each other—reflecting the thread that I followed through the coding and in order to engineer the kind of cross-disciplinary conversation which has largely been absent to date.

Analysis

ILK in models and scenarios: scale, complexity, and translation

From modelling the natural world to ‘blackboxing’ the social world

In the words of an ex-modeller: a ‘model is always an abstraction, a representation, a very simple representation of the real world’ (Interviewee E). There are different families of models with varying potential for the integration of ILK (Interviewee G). The MAR, for instance, characterises ‘expert-based modelling’—where ILK holders, as experts, draw on their experience to establish relationships in the model—as the preferred site for knowledge integration (IPBES 2016a). This is presented in contrast to Integrated Assessment Models (IAMs)—which couple social and ecological dynamics—where integrating ILK is a much greater challenge (IPBES 2016a). While scenarios are often linked to models, scenarios can be ‘qualitative interpretations of nature as opposed to quantitative interpretations of nature’ (Interviewee B). Their main purpose is to encapsulate different plausible futures; they are ‘angles’ that reflect and mediate our desires, values, and relationships with nature (Interviewee E). For those reasons, scenarios seem to offer greater potential for including multiple knowledge systems and worldviews (Interviewees E and G). Ecologist and member of the ILK Taskforce, Interviewee B, insisted on this distinction:

‘I think it is important to distinguish between models and scenarios. I cannot see how ILK can easily be incorporated in a model - which, by its very nature, is using almost mathematically-derived elements and you cannot really do that with most ILK. Some may be able to do it, but it’s marginal. Scenarios, on the other hand, I think you can have the ability to build in qualitative as well as quantitative work. So, the models would actually be one aspect of what would feed into a scenario’ (Interviewee B).

In the case of both models and scenarios, aspects of ILK inevitably get stripped away or get lost in the process (Interviewees A and E). Models are perceptions of the future inextricably attached to the modellers building them (Interviewee C). As models become more encompassing and complex, all form of agency disappears and the modellers, themselves, begin to lose an understanding of their own models (Interviewee C). The initial theoretical assumptions underpinning the design of the algorithms, such as assumptions of rational behaviour, get dissimulated (Interviewee A). For instance, in many cases, requirements of internal consistency, inherent to modelling, are carried over to scenario-building activities. That may well be reasonable in the case of biophysical processes, but becomes far more problematic in the case of qualitative storylines:

‘Who’s internally consistent? I certainly am not. When I talk about my story of the future: what I want, what I fear, and what I think - all of those statements put together make a scenario, but it is certainly not going to be internally consistent’ (Interviewee C).

For interviewee H, a political scientist involved in scenario work, these kinds of issues raise important questions around legitimacy and value implications in scenario-building processes. She is mainly concerned with the selection of a few storylines which are supposed to be cross-scale and representative of societies’ desired futures (Interviewee H). In a similar vein, Interviewee F brought up the challenge of modelling what he called ‘radical futures’ (futures that offer alternatives to the status quo) and associated transitions. Hence, while biophysical models can present strong narratives about the natural world, irrespective of place (e.g. physical laws), the task becomes much harder in trying to ‘blackbox’ social dynamics (Interviewee C). Yet because biophysical changes are often tied to the social contexts in which they occur, the need to bridge the social and ecological dynamics remains. At the global level, efforts to combine social and ecological dynamics in singular models have been relatively unsuccessful. For Interviewee C, global IAM models are generally very poor and the modellers building them often have too much faith in them. For Interviewee F, these debates have been around for a while. For him, the real challenge and novelty in IPBES is in the ambition to have cross-scale integration.

The current reliance on climate science methodologies

In meeting this ambition, one of the issues that the BES community is facing is its reliance on methodologies considered best practice in the climate sciences (e.g. Shared Socioeconomic Pathways), but are often inadequate for BES (Interviewees D and F). These methods have been designed with specific aims in mind. Some were built for modelling energy transitions, for example, and they may not translate well in the case of evaluating ecosystem services or biodiversity conservation (Interviewee F). Moreover, while IPCC scenarios do translate well locally, they still remain very top-down. As regards the IPBES mandate to have models and scenarios that work at different scales of decision making, IPCC-type scenarios will not be sufficiently bottom-up (Interviewee J). Moreover, while climate change is often conceptualised as a global problem that requires global solutions, biodiversity issues can be seen as global, but will almost always require very local solutions (Interviewee B). The global characteristic of CO2 is unique:

‘Carbon dioxide is a global thing: you burn CO2 in Sweden and Indonesia and a year later it doesn’t matter where it was from. If you are sucking carbon dioxide in the Arctic or in Somalia, it doesn’t matter for the world, but it does matter for biodiversity. If you cut down trees in Indonesia, that’s different from cutting it down in Sweden. It’s going to affect different species; it’s going to have different effects on different ecosystems; it’s going to affect different people. […] Biodiversity issues are much more of an aggregation problem than a global problem - whereas carbon is really this global cycle problem. Or it is local stuff but it gets mixed away very quickly’ (Interviewee F).

If IPBES is truly committed to including ILK in models and scenarios, it will need to move away from a default reliance on climate sciences practices (Interviewee D), but with the current lack of funding and in-kind support, stimulating the creation of completely different methodologies will be challenging (Interviewee F).

The issue of scale in BES models and scenarios

IPBES has recognised that the spatial resolutions of models and scenarios have often been inadequate. If feasible, a multi-scale approach is what the organisation is striving for, as outlined in its guiding document (IPBES 2016b). The challenge then will be to match the scale of future assessments with both the scales of the drivers of environmental change and the various scales of the ecosystems under pressure from these drivers (IPBES 2016b). The issue of scale was identified as one of the main issues to be discussed in the workshop held in Auckland (Interviewee J). Currently, there is no scientific consensus on how to move beyond the local scale—where most of the development of participatory processes for model- and scenario-building has occurred (Interviewees C, D, E, and H). Some of the tension comes from a top-down/bottom-up divide (Interviewee C). On the one hand, local models and scenarios can be aggregated, collated, and compared to say something about relationships at national or regional scales (Interviewee D and E). On the other hand, a top-down approach can facilitate comparison across scales and may render issues visible, such as climate change, that are otherwise invisible in local models (Interviewee E).

The issue of upscaling becomes more complex with the involvement of ILK—which is inherently local to begin with. In the reductionist framing of ILK ‘as data’, there are questions about its relevance at the global scale. A lot of the local, granular data disappears when aggregated (IPBES 2016b) and ILK may even be irrelevant in global models and scenarios:

‘For global modelling, if you go into the waves you won’t see the sea. […] You want to see trends. If we want to make a global map - even with some high-resolution scale - it’s not that a specific cell reflects any truth, it’s just that you are trying to identify patterns on a global level. In that sense, what people are thinking locally (even if it’s ILK communities or if it’s private parties) is irrelevant. You should just be trying to see: “is biodiversity going up or down in the coming 100 years?”’ (Interviewee D).

One way around this is to have bottom-up stakeholder processes that encompass a wider diversity of narratives, compare those with the global models, and start thinking about where they match and do not match (Interviewee C). Regardless of the scale of analysis, if the models and scenarios are going to involve ILK, ‘you actually do need to talk with the local people and confirm local values’ (Interviewee J), and there is no way around grounding them in particular contexts (Interviewee H). Interviewee B suggested that one of the aspects of ILK which may be scalable is the knowledge of ‘ecosystem processes’:

‘If you strip away worrying about species and if you also actually stop thinking about ecosystem services - which again are much more specific and place-localised (an ecosystem service of a floodplain is not the same as an ecosystem service for carbon sequestration or the production of food) - [and start thinking about] ecosystem functions [that] are at a different level (all ecosystems have very specific functions which are common, globally), […] that could provide the Rosetta Stone which would enable everybody to talk about their particular issues’ (Interviewee B).

Here, Interviewee B is suggesting that ecosystem processes or functions could act as a ‘boundary object’ (cf. Star and Griesemer 1989). Interviewee G disagreed, arguing that there is a misperception that scientists need to upscale ILK, whereas, in the day-to-day, indigenous people and local communities are engaging at the global level and upscaling their knowledge themselves:

‘It becomes almost like the content of people’s knowledge is not important. I don’t think that’s true. People have knowledge and understandings and insights that we don’t have. I’m reluctant to just say it’s about process. But what I am saying is that this question of how do you scale up from a local and indigenous knowledge system to a global ILK system, my answer, to date, is: indigenous and local communities are showing us how to do that every day. So, we work with them. We don’t try to scale it up ourselves. We work with them on their methods for scaling it up’ (Interviewee G).

Moreover, Interviewee G suggested that ILK is multi-scalar anyway: indigenous people and local communities have an understanding of more-than-local pressures and ‘they are living in modernity’. They are dealing with the impacts of industrialisation every day of their lives and they have their own models and their own understandings of priorities’ (Interviewee G). She explained that the team working on the classification of Nature’s Benefits to People (now called Nature’s Contributions to People) has been working ‘really hard’ on including these indigenous models (Interviewee G). The question, then, revolves around whether the integration (and upscaling) of ILK in models and scenarios work involves collating the scientific and indigenous models or whether it necessarily involves some form of translation, boundary-crossing.

The quantitative/qualitative divide and the issue of translation

The MAR identifies one of the main challenges of mobilising ILK in models and scenarios work as ‘translating qualitative data into quantitative inputs’ (IPBES 2016a: 90). To date, there is no straightforward way of translating a cultural or spiritual claim, for example, into a model or a scenario—although you can put a ‘tag’ on culturally important sites and give them weight in the scenario-building process (Interviewee E). A lot of work has been going into ‘fuzzy data sets’ or ‘fuzzy maps’ which offer ways of translating qualitative words into quantitative ‘density functions’ (a statistical tool that determines the probability of a variable falling within a range of values)—which can then be inputted into a model (Interviewee C). However, according to Interviewee C—whose PhD student has been working on these issues—most of the attempts to do this so far have failed. It is also a ‘technical fix’ that does not get around the problem of presenting a diversity of worldviews (Interviewee C). Moreover, these tools are only useful for a very narrow overlap of quantitative and qualitative variables (cf. Fig. 1, below).
Fig. 1

Venn diagram of the ‘space of translation’ (from Interviewee C)

In Fig. 1, ‘qualitative data’ can refer to either ILK or to the various qualitative storylines feeding into scenarios. The overlap of the quantitative models and qualitative scenarios can consist of socio-ecological variables, such as population growth, but will seldom involve the stakeholders’ values and narratives (Interviewee C); an omission that IPBES is committed to address (Pascual et al. 2017). For Interviewee C, quantification can only happen in that intersection, which I have termed ‘space of translation’ (Interviewee C). For these reasons, Interviewee C’s approach to participative scenario-building involves an initial ‘creative’ qualitative strand of work (directly with the stakeholders)—to draw out the various stories and narratives—before asking stakeholders to evaluate different models and scenarios at the end of the process. Upstream, ILK can help improve ‘cultural and societal layers’ in mapping exercises (Interviewee J). The difficulties, then, are that everything often ends up having value for different people and scientists tend to struggle with non-quantifiable layers (Interviewee J).

Upstream integration of ILK also brings with it the problem of ‘validation’. Interviewee B argued that this something ‘that most scientists are either openly, or at least covertly, dismissive of in terms of ILK’ (Interviewee B). For him, the fact that ILK will need to be validated by modellers, in any given modelling exercise, means that ILK should not be included in quantitative BES models or scenarios. Instead, he argued that the criteria of validation of ILK are determined by ‘place-habitation’ and time length. ILK is ‘continuously evolving through the interaction of grounded experiences and different types of knowledge (written, oral, tacit, practical, and scientific) that are empirically tested, applied and validated by indigenous peoples and local communities’ (IPBES 2016b: 82). With regard to this issue, the IPBES guiding document explicitly stipulates: ‘ILK holders need to ensure that the inclusion and interpretation of their knowledge in assessments is robust and appropriate in terms of their own validation methods’ (IPBES 2016b: 83). For Interviewee A, an STS scholar, this is crucial, because there is an asymmetry of information—between the modellers and the ILK holders—on what is done to their knowledge when it goes into a model:

‘You would have to have an understanding of what you would need to say in terms of what you are going to do to this knowledge. And I think that that understanding is incomplete, because it’s mostly implicit. […] It’s very possible to integrate ILK, but we have to recognise that as soon as you do that, the product of the integration cannot actually be traced back to its ingredients; those have changed. So rather than integration, it is translation. I think that integration is a misleading term. You do violence to knowledge. You do violence to all knowledge as soon as you integrate it. It’s not just true for ILK; it’s also true for other sources of knowledge: as soon as you put it into a model it becomes something else. You can do it, but what are you doing?’ (Interviewee A).

For Interviewees A and G, you will not get consent from ILK holders to quantify, strip away parts of their knowledge, upscale it, and ‘disentangle knowing and acting’. You will not get them to formally consent on paper, but if there is a common understanding that any form of translation (or quantification) will inevitably strip away certain claims, values, and political identities, that could lead to some pragmatism (Interviewees A and H). Interviewee A suggested that these questions could and should be addressed by the ILK Taskforce. As I have attempted to show, efforts to reconcile scientific practice with political demands are contingent on translating knowledge in a way that is acceptable to the different parties involved. This throws up wider issues with knowledge integration in research and assessment contexts which I explore in greater detail in the following section.

Wider issues with bringing ILK and science together in research and assessment contexts

Terminology, epistemology, and interdisciplinary differences

There is a long history of integrating non-scientific knowledge systems into science (e.g. citizen or amateur science) (Interviewee A). Integration can have many meanings and ‘when people are using that word they are thinking about different things’ (Interviewee I), but for Interviewee B, ‘integration’ is a term that carries a lot of ‘cultural baggage’. Alternative terminology was discussed at length in the ILK Taskforce (Interviewee B and G):

‘We’ve talked about weaving on the ILK taskforce. We’ve talked about bridging. We’ve talked about synergy. We’ve talked about just about everything except integration. I don’t think we’ve got quite the right word yet. I think integration can be fine as long as integration is negotiated. So it’s not a matter of science representing indigenous knowledge, because it hasn’t proven very good at that’ (Interviewee G).

Indigenous and Local Knowledge (ILK) as an umbrella term is not without its problems either. For Interviewee I, it is too encompassing and tends to be used in ways that suggest that it is homogenous, whereas ILK as a monolithic entity is neither homogenous in terms of knowledge, nor in terms of values (Interviewee D). Indeed, different indigenous communities have different political and cultural claims that may even come into conflict with one another (Interviewee G). More problematically, indigenous knowledge is often perceived as ‘magical’ or ‘mystical’, which often supersedes or dissimulates the broader political implications of indigenous knowledge (Interviewee F). While indigenous and local communities certainly do have intrinsic, traditional knowledge of natural resources, the overemphasis (or romanticisation) of this facet of ILK tends to underplay both the more complex, hybrid nature of ILK, and the associated (political) demands for understanding and managing natural resources differently (Interviewee G). Even within the IPBES ILK Taskforce, local knowledge has been systematically overlooked (Interviewees B and F):

‘We’ve spent most of the time talking about “I”, and LK really has had a look-in in the discussions and that’s really a pity, because IK is kind of deified (a bit) and LK is utterly disregarded; I mean “how can a farmer possibly help?”. Well, yes they can if they and their families have been living on that land for a continuous number of generations. Yes they can, because they have the same process of accumulation of knowledge as indigenous people do’ (Interviewee B).

This kind of romanticisation does not only apply to ILK. In what Interviewee I—an anthropologist and lead author of the Global Assessment—called the ‘ontological framing’, knowledge systems often get portrayed stereotypically:

‘I think both sides have a tendency to romanticise issues. […] You have ways of framing science that are very stereotypical and ways of framing ILK that are very stereotypical. So, those are two kinds of narratives coming together in that space. […] They’re both trying very hard and trying to have good intention, but in the process you have misrepresentations of what representation is, what participation is, what integration is. […] I think it ends up promoting romanticisation, promoting separation, [and] thinking about integration as a magic bullet; in fact, underestimating how much integration there is out there, how much people on the ground do that all the time and do not see those separations of knowledge in such a marked way. […] I’m just trying to step back a little bit. The conceptualisation of this whole issue [and] this wave of ontological thinking ends up stressing differences and problems more than appreciating other ways of where they converge, where it’s different, and so forth’ (Interviewee I).

Such a framing can overemphasise a dichotomy between ILK and science and may be conducive to ideas of integration whereby ILK is seen as knowledge that needs to be validated by scientists (Interviewee B). For Interviewee G, ‘what ILK systems have pursued has been holistic knowledge, when science has become very reductionist’. For interviewee E, a modeller, ILK is chiefly experiential. It can help broaden the scope of scientific work, but if it is to be included in models and scenarios it will inevitably need to be processed by scientists (Interviewee E). For Interviewee G, there is a misperception that the way ILK holders and scientists see the world is vastly different. Preconceived ideas should be cast aside and indigenous people and local communities should be given the ‘opportunity to have their voices heard’ (Interviewee G). Regardless of those differences, what ILK and science both bring to the table are understandings of what may be safeguarding or degrading biodiversity and ecosystems (Interviewee G).

Elsewhere, I have shown these sorts of disagreements emerge, in part, from different epistemological standpoints between natural scientists and social scientists, and that they may have adverse effects on knowledge integration (Obermeister 2017). For instance, Interviewee C told me that ‘hard-core environmental biophysical modellers’ and ‘hard-core qualitative social scientists’ have vastly different languages and, hence, seldom talk to each other. Moreover, they have different understandings of what constitutes ‘bad science’—which can entail that some social scientists would rather see all the global models disappear (Interviewee C) and some natural scientists get nervous at the prospect of working with social scientists (Interviewee B). There is also a tendency for these natural scientists to be far more ecocentric than their social science peers—which can lead to further value conflicts (Interviewee E). While Interviewee C spoke about the divide between logical positivism and constructivism, for Interviewee I, it primarily boils down to the ongoing dichotomy between ‘idiosyncratic’ knowledge (only context matters) and ‘generalisable’ knowledge (certain theories can be universalised). Interviewee C gave the example of the purpose of case studies: while ecologists may use case studies to say something about patterns and trends beyond the initial context, the social scientists will use case studies ‘to show irreducible complexity’ (Interviewee C). In moving towards a transdisciplinary research or assessment setting that is conducive to opening up the space for non-scientific, lay knowledge, I have argued that fostering a dialogue and reconciling these ‘geographies of knowledge’ is imperative (Obermeister 2017). For Interviewee I, this involves a more robust, collective understanding that GEAs inevitably contain both a ‘factual space’ and a ‘political space’. In the words of Interviewee A, ‘political logics and scientific logics don’t necessarily align’. I explore this idea further in the following section.

The depoliticisation of knowledge and the politicisation of issues

The translation of ILK is not the only issue that makes scientific sense, but can be politically untenable. While the MAR and the guiding document do not take issue with ‘representation’, most of the interviewees pointed to the various dilemmas that come with it. The idea that a natural scientist may be able to represent, in part, the breadth of knowledge in their field is less problematic than in the case of an ILK holder or a social scientist representing a plurality of perspectives (Montana 2017). For Interviewee H, such representation can be detrimental to ‘generative dialogue’:

‘You cannot represent the globe and all its diversity with 60 people. It raises a number of issues that are, for social scientists, much bigger in the sense of: how do you create legitimacy? [With] the people who are there, what is the quality of dialogue? How can that dialogue be valuable for those not in the room? I do think that is possible, but it really depends on a truly high quality of […] “generative dialogue” - where you are not there to represent a particular perspective or a particular interest group, because then you are not open to change your mind or open to new ideas, but you really have a dialogue where there is much more listening than talking. […] They [natural scientists] talk about: “we have to have that person, because they represent these issues and those issues”. […] You do need these kinds of representatives, but I do want those representatives to leave representation [behind]’ (Interviewee H).

When applied to a scientific logic of inclusion, representation can lead to the type of depoliticisation of knowledge decried by Filer (2009) in the case of the Millennium Ecosystem Assessment. As Brosius (2006) points out, the representation of ILK by social scientists or ILK experts (Tengo et al. 2017) can go a long way in underplaying some of the political reasons for ILK holders to be participating in these GEAs in the first place (e.g. demands for co-management of natural resources). Whilst the representation of ILK by scholars can be disempowering and disrespectful, it may also be empowering: some indigenous (and non-indigenous) scholars have been ‘at the forefront of decolonising methodologies’ (Interviewee G). Ultimately, though, ILK holders do need to be representing themselves (Interviewee A).

In doing so, ILK holders often bring issues to the table that had not necessarily been considered by scientists (Interviewees G and J). For instance, ILK can bring an understanding of people’s contribution to nature (as opposed to nature’s contribution to people) (Interviewee B). They can also politicise certain issues by making them visible. Interviewee G provided the example of the Asia–Pacific assessment where ILK holders brought up the issue of nuclear fuel waste, which had hitherto been ‘out of sight, out of mind’. That contribution changed the nature and focus of the assessment (Interviewee G). In the case of models and scenarios, such politicisation can take the form of non-scientists having a say on what kinds of models and scenarios are relevant to particular decision contexts (Interviewee H). This may sometimes be at odds with some scientists’ expectations that they will ‘walk into the room and present a scenario or ten different options and have people pick’ (Interviewee J). Interviewee I pointed out that, even though IPBES has done more than its predecessors in politicising issues around biodiversity and ecosystem services (e.g. its explicit focus on ‘values’), it will need to get better at politicising some issues, such as climate mitigation, which often get depoliticised in the ‘political economy of the process’ (Interviewee I). For Interviewee B, recognising that the products coming out the assessments are inherently political is not about being policy prescriptive; it is about getting better at communicating with policy makers. Scientists with experience in the policy world already share a sense that whatever evidence they put forward, it will not necessarily change policy (Interviewee J). Even in the case of models and scenarios, the products will need to travel from the factual space into the political space (where narratives inform the outcomes) (Interviewee I). As Jasanoff (2010) claimed: ‘environmental knowledge achieves robustness through continual interaction—or conversation—between fact-finding and meaning-making’ (p. 16). In the next section, I argue therefore that knowledge needs to be brokered both within and between these two spaces.

Knowledge brokering and the ‘data-knowledge’ interface

Whereas the MAR does not refer directly to knowledge brokering, the guiding document states: ‘activating ILK networks can help identify inherent solutions to cross-scale issues, through processes like knowledge brokering and collaboration’ (IPBES 2016b: 83). The term ‘knowledge broker’ is often too encompassing (Turnhout et al. 2013). Hence, Turnhout et al. (2013) make a helpful distinction between knowledge brokering as ‘bridging’ and knowledge brokering as ‘facilitating’. Whilst ‘bridging requires knowledge brokers to mediate and translate between the domains’, facilitation is more focused on finding appropriate processes of interaction between different knowledge systems (Turnhout et al. 2013: 361). A couple of the interviewees highlighted the importance of objective, professional facilitation in participative models and scenarios work, so as to make sure that stakeholders are safeguarded against being used as ‘data-holders’ and to guarantee that there is a neutral party with no vested interest that stakeholders can trust (Interviewees C and J). When it comes to bridging knowledge systems, a few of the interviewees suggested that social scientists could play an important role (Interviewee B and C)—although it was also pointed out that natural scientists with experience in local fieldwork, such as ecologists, could equally act as knowledge brokers (Interviewees C, E, F, and I). For Interviewee G, social scientists are not the knowledge brokers, but they can provide ideas on how to broker knowledge:

‘I don’t think social scientists should be knowledge brokers. If you want to be a knowledge broker then that doesn’t require a PhD […] Looking at what the tools, methods, and processes are to allow that dialogue to happen in an effective way, that’s something that social scientists can help with. […] There is a science of knowledge brokering which social scientists can bring to the table. The broker is someone else. The broker could be the ecologist’ (Interviewee G).

In that sense, there was a relative consensus among interviewees that knowledge brokering is more about experience, tools, and process than a role grounded in a particular intellectual tradition or affiliation. However, for Interviewee C, the knowledge brokers do have to learn the language of social science (as well as being sufficiently scientifically literate). In the context of IPBES, then, knowledge brokering involves both brokering within and between (Interviewee I). On the one hand, knowledge brokering is central to interdisciplinary work and a common language needs to be developed within the scientific community (Interviewee I), as well as between the scientific community and the ILK holders. On the other hand, knowledge needs to be brokered between scientists and policy makers. For Interviewee B, these are two separate spaces of knowledge brokering:

‘We talked a lot, in the literature, in the last decade or so, about science–policy interfaces, but for me there’s a data-knowledge interface and knowledge-policy interface. You have to cross one interface before you get into the other’ (Interviewee B).

For Interviewee B, it is in the knowledge-policy interface that social scientists can help translate and communicate the concerns of both scientists and ILK holders to policy makers. Hence, in the assessment process, the natural sciences and the social sciences remain as two different strands of work (Interviewee D), but in the bridging of the assessments and policy, social scientists can play a central role. With regard to BES models and scenarios, the MAR recurrently identifies one of the key issues as the need to make sure that ‘scenarios and models of drivers’ are ‘specifically tailored to the needs of different policy or decision contexts’ (IPBES 2016a: 107). The MAR is explicit about the need to match ‘methodological approaches’ with ‘particular policy or decision-making process’ (IPBES 2016a: 18). In agreement with Interviewee B, I contend that this is the principal area of models and scenarios work where there is the greatest potential for social scientists to act as effective knowledge brokers.

Discussion and conclusion

Despite efforts to address finer scales, GEAs preceding IPBES—and especially the IPCC—have legitimised a ‘view from nowhere’ (Shapin 1998) through ‘global kinds of models and scenarios’. In an ambition to break away from these largely top-down, panoptic understandings of climate and environmental change, IPBES has committed to support and catalyse the development of multi-scale models and scenarios. Their novelty is summarised in the report for the IPBES visioning workshop, held in September 2017 (Auckland, New Zealand):

‘The new IPBES scenarios and modelling framework will shift traditional ways of forecasting impacts of society on nature to more integrative, nature-centred visions and pathways for the future of nature that are relevant for conservation policies and practice. […] Importantly, they will integrate the social-ecological feedback loops across drivers, biodiversity, ecosystems, ecosystem services, and human wellbeing, and incorporate multiple systems of knowledge’ (Lundquist et al. 2017: 12).

With regard to scenarios in the short term, this may include a reconceptualisation and extension of IPCC global scenarios (i.e. Shared Socioeconomic Pathways and Representative Concentration Pathways) by the BES community (Crossman et al. 2018; Lundquist et al. 2017; Rosa et al. 2017). However, ultimately, new Nature Futures scenarios are expected to be produced by 2019 (before the second IPBES work programme) (Crossman et al. 2018). An important challenge identified in the literature—and echoing some of the interviewees’ concerns—is the need for ‘comparable metrics for biodiversity and ecosystem services’ to ‘harmonise outputs from models’ (Rosa et al. 2017). In the case of ILK, one such metric was suggested by Interviewee B who proposed that indigenous and local understandings of ecosystem processes could be scalable and compatible with scientific understandings—providing in his words: ‘the Rosetta Stone which would enable everybody to talk about their particular issues’ (Interviewee B). Another interviewee spoke of fuzzy datasets, fuzzy maps and drew a Venn diagram (Fig. 1) to illustrate the site of intersection between quantitative and qualitative data, that I named the ‘space of translation’. However, as he pointed out, this is only a technical fix and runs the risk of dissimulating values and worldviews. Instead, Interviewee G proposed that the process of collating and scaling up could start with ‘indigenous models’, ergo allowing indigenous and local communities to determine the common metrics and variables rather than the modellers themselves.

These interviewees are in fact offering options for a common metric (or language) which could act as a ‘boundary object’, defined by Star and Griesemer (1989) as ‘objects which are both plastic enough to adapt to local needs and the constraints of the several parties employing them, yet robust enough to maintain a common identity across sites’ (p. 393). Conceptually, boundary objects emerge in part from Latour’s (1987, 1999) idea of translation as a channelling, stabilising and ordering device across boundaries, rallying allies around a common understanding. With the subject at hand, to ensure that multi-scale models and scenarios can effectively act as boundary objects, translation must happen across scales and between knowledge systems. As I have shown through the analysis, the need for translation not only applies to ILK and scientific knowledge, but also between different research and policy practices at all stages of model and scenario development (from the input, through the processing, to the outputs). This includes translation across the natural and social worlds (see “From modelling the natural world to ‘blackboxing’ the social world”), from methods in climate change research to BES research (see “The current reliance on climate science methodologies” in section), across scales and geographies (see “The issue of scale in BES models and scenarios”), across the qualitative/quantitative divide (and the issues of validation) (see “The quantitative/qualitative divide and the issue of translation”), between terminologies, epistemologies and disciplines (see “Terminology, epistemology, and interdisciplinary differences”), and between the ‘factual space’ and the ‘political space’ (see “The depoliticisation of knowledge and the politicisation of issues”).

As illustrated by the idea of boundary objects, translation is also a delicate balancing act whereby different parties need a common view of what should and should not be translated, into what ‘language’ and for what purposes. For instance, in the act of reconciling the scales of environmental and climate change with the scale of ILK, there is a tension between a bottom-up and top-down approach for multi-scale models and scenarios. A bottom-up approach can be translated vertically (scaled up) to reveal interactions and trends beyond the context within which they originate. A top-down approach can be translated horizontally (across geographies) and then vertically (with greater granularity, resolution), but only one way (by ‘zooming in’). Both approaches, however, make certain socio-ecological and socio-political realities visible and others invisible, favouring particular narratives over others. While bottom-up models and scenarios are preferable for the inclusion ILK and local worldviews (Crossman et al. 2018; Kok et al. 2017), they may not adequately capture the impacts and trickle-down effects of climate change. Such epistemological transactions are inevitable, leading practitioners to call for ‘a bottom-up/top-down approach [which] would build on many local scenarios, stakeholder networks and local research capacities as well as place these in a global context, focusing on the interactions between local trajectories and global dynamics’ (Crossman et al. 2018: 44).

In including ILK, all the interviewees expressed concerns about the issues of representation, selection, validation, romanticisation (of knowledge) and depoliticisation—pointing to the difficulty of translating knowledge and criteria for its evaluation across different knowledge systems. Natural scientists may be seen to represent a ‘consensus view’ in their field, but the same may not necessarily be said for social scientists, humanities scholars and ILK holders. While ILK holders are actively encouraged to validate their own knowledge with their own criteria, scientists and modellers do end up having to apply some process of selection and validation further down the line. Asymmetries of information with regard to what happens to knowledge when it is fed into a statistical model or a qualitative scenario persist—with one interviewee suggesting that expert modellers may not even fully understand what happens to their own knowledge. Political motivations and narratives expressed by ILK holders may be too ‘unconformable’ for scientists who, for the most part, continue to uphold the porous boundary between science and politics.

These issues and the questions they raise are already well documented in the social science literature; whether that involves discussing alternatives to ‘integration’ (e.g. Klenk and Meehan 2015; Sutherland et al. 2013; Tengo et al. 2017), demystifying indigenous and local knowledge (e.g. Agrawal 1995; Filer 2009; Houde 2007), or breaking down the boundary between science and politics in science–policy interfaces, with the coproductionist idiom for example (e.g. Jasanoff 2010). When it comes to social scientists’ involvement with multi-scale models and scenarios, I suggest that their quantitative inclination should not be perceived as a threat, but rather as a ‘stimulating scientific challenge’ (Stenseke 2016: 126). Moreover, if ‘values’ are going to be at the centre of IPBES’s work, knowledge integration will need to involve the bridging of both knowledge systems and values (Pascual et al. 2017; Stenseke and Larigauderie 2017)—an area where social scientists may be of assistance. In particular, scenarios could offer a site for weaving ILK and science by bringing qualitative and quantitative strands of work together (De Vries and Petersen 2009; Stenseke and Larigauderie 2017). For example, this could be done through ‘visioning’ exercises to identify desirable or undesirable policy pathways (Lundquist et al. 2017; Rosa et al. 2017). While the contribution of the social sciences and humanities certainly does not limit itself to ‘process design’, in their review, Vadrot et al. (2018) recently highlighted that the social sciences and humanities can assist in developing ‘techniques for deliberation, participation and pluralistic (valuation) approaches for modelling, predicting and assessing risk’ as well as ‘methods to ensure participation and ownership such as participatory scenarios analysis’ (p. 9).

By offering ‘concepts for mediating, organising and facilitating interfaces between different actor groups and fields’ (Vadrot et al. 2018: 9), social scientists and humanities scholars can strengthen what interviewee G called the ‘science of knowledge brokering’. However, the social scientists or humanities scholars need not necessarily be the knowledge brokers. In fact, most interviewees agreed that the knowledge broker could or should be the ecologist, the professional facilitator or the ILK holder. For Peterson et al. (2018), IPBES should be the knowledge broker ‘across multiple research-practice interfaces and […] translate among a diversity of perspectives, actors, methods, and values’ (p. 39). As a first step in that direction, IPBES and its network could start by reframing integration as the assemblage of a ‘variety of types of evidence that each present a piece of the puzzle’ (Interviewee I). For Interviewee C that might entail the presentation of narratives and scenarios from different scales and different perspectives, side by side.

All in all, with its models and scenarios work, IPBES (specifically the 3c Group) and affiliated researchers will need to get better at managing the expectations of the different stakeholders involved through sustained knowledge brokering. On the one hand, growing pressures to produce ‘better science’—which must be evermore policy-relevant and inclusive—may create standards too high for scientists to meet. On the other hand, in research and assessment contexts, the criteria of validation for the physical and natural sciences cannot be brought to bear on the social sciences, the humanities, and on non-scientific knowledge systems. There are also dangers that if the push for integration is too strong, certain authors may integrate disrespectfully or the benefits of knowledge integration may come into direct conflict with the need to meet other objectives. Hence, an open and honest dialogue will be key to success.

Specifically, scientists will need to get better at communicating what they can and cannot realistically achieve, as well as spelling out what will happen to knowledge when it gets integrated. Social scientists, government stakeholders, and ILK holders will, accordingly, need to provide an indication of what may be appropriate and what may be unacceptable. Towards its second work programme, IPBES could begin with making sure that there are strategies for bringing ILK and models and scenarios together in future assessments, as well as equipping the scientific community with guidelines for doing so in research contexts. Having only here scratched the surface, more engagement and research in the social sciences and humanities is needed in order to fully gauge and understand the expectations of ILK holders and government stakeholders, respectively. Such engagement is crucial; not only to ensure that the inclusion of knowledge outside of academia is orchestrated in a respectful and democratic way, but also to support IPBES on its journey away from the strictly ‘global kinds of knowledge’ upheld by its predecessors.

Notes

Acknowledgments

I am first of all grateful to my interviewees for being so generous with their time and for engaging so enthusiastically. I would like to thank Dr. Seth Gustafson for excellent guidance and for his patience. I would like to thank the anonymous reviewers for their extensive and constructive feedback. I would also like to thank Jonathan, Clara and Elliot for being thorough readers and for their insightful feedback. Finally, I would like to thank my friends and fellow UCL alumni for their ongoing support.

References

  1. Agrawal A (1995) Dismantling the divide between indigenous and scientific knowledge. Dev Change 26(3):413–439CrossRefGoogle Scholar
  2. Beck S, Mahony M (2017) The IPCC and the politics of anticipation. Nat Clim Change 7:311–313CrossRefGoogle Scholar
  3. Beck S, Maud B, Chilvers J, Esguerra A, Heubach K, Hulme M, Lidskog R, Lövbrand E, Marquard E, Miller CA, Nadim T, Neßhöver C, Settele J, Turnhout E, Vasileiadou E, Görg C (2014) Towards a reflexive turn in the governance of global environmental expertise: the cases of the IPCC and the IPBES. Gaia 23(2):80–87CrossRefGoogle Scholar
  4. Bogner A, Menz W (2009) The theory-generating expert interview: epistemological interest, forms of knowledge, interaction. In: Bogner A, Littig B, Menz W (eds) Interviewing experts. Palgrave Macmillan, New York, pp 43–80CrossRefGoogle Scholar
  5. Bogner A, Littig B, Menz W (2009) Introduction: expert interviews—an introduction to a new methodological debate. In: Bogner A, Littig B, Menz W (eds) Interviewing experts. Palgrave Macmillan, New York, pp 1–16CrossRefGoogle Scholar
  6. Borie M, Mahony M, Hulme M (2015) Somewhere between everywhere and nowhere: the institutional epistemologies of IPBES and the IPCC. Unpublished paper presented at the STEPS Annual Conference, Resource Politics, 7-9 September, University of Sussex, UKGoogle Scholar
  7. Brosius PJ (2006) What counts as local knowledge in global environmental assessments and conventions? In: Reid WV, Berkes F, Wilbanks TJ, Capistrano D (eds) Bridging scales and knowledge systems: concepts and applications in ecosystem assessment. A contribution to the millennium ecosystem assessment. Island Press, Washington, pp 129–144Google Scholar
  8. Crossman ND, Banerjee O, Brander L, Verburg P, Hauck J (2018) Global socio-economic impacts of future changes in biodiversity and ecosystem services: state of play and approaches for new modelling. Report prepared for WWF-UK.  https://doi.org/10.13140/rg.2.2.26613.68329
  9. De Vries BJM, Petersen AC (2009) Conceptualizing sustainable development: an assessment methodology connecting values, knowledge, worldviews and scenarios. Ecol Econ 68:1006–1019CrossRefGoogle Scholar
  10. Filer C (2009) A bridge too far: the knowledge problem in the millennium ecosystem assessment. In: Carrier JG, West P (eds) Virtualism, governance and practice: vision and execution in environmental conservation. Berghahn Books, New York, pp 84–111Google Scholar
  11. Fujimura JH (2011) Technobiologicbpal imaginaries: how do systems biologists know nature? In: Goldman MJ, Nadasdy P, Turner MD (eds) Knowing nature: conversations at the intersection of political ecology and science studies. The University of Chicago Press, Chicago, pp 65–80Google Scholar
  12. Goldman MJ, Turner MD (2011) Introduction. In: Goldman MJ, Nadasdy P, Turner MD (eds) Knowing nature: conversations at the intersection of political ecology and science studies. The University of Chicago Press, Chicago, pp 1–23CrossRefGoogle Scholar
  13. Gustafsson KM, Lidskog R (2018) Organizing international experts: IPBES’s efforts to gain epistemic authority. Environ Sociol: 1–12Google Scholar
  14. Houde N (2007) The six faces of traditional ecological knowledge: challenges and opportunities for Canadian co-management arrangements. Ecol Soc 12(2):34CrossRefGoogle Scholar
  15. Hulme M (2010) Problems with making and governing global kinds of knowledge. Global Environ Change 20:558–564CrossRefGoogle Scholar
  16. IPBES (2016a) The methodological assessment report on scenarios and models of biodiversity and ecosystem services. Secretariat of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, BonnGoogle Scholar
  17. IPBES (2016b) Guide on production and integration of assessments from and across all scales. Secretariat of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, BonnGoogle Scholar
  18. IPBES (2018) IPBES Core Glossary. Available from https://www.ipbes.net/glossary. Accessed 14 Apr 2018
  19. Jabour J, Flachsland C (2017) 40 years of global environmental assessments: a retrospective analysis. Environ Sci Policy 77:193–202CrossRefGoogle Scholar
  20. Jasanoff S (2010) A new climate for society. Theory Cult Soc 27(2–3):1–21Google Scholar
  21. Klenk N, Meehan K (2015) Climate change and transdisciplinary science: problematizing the integration imperative. Environ Sci Policy 54:160–167CrossRefGoogle Scholar
  22. Kok MTJ, Kok K, Peterson GD, Hill R, Agard J, Carpenter SR (2017) Biodiversity and ecosystem services require IPBES to take novel approach to scenarios. Sustain Sci 12(1):177–181CrossRefGoogle Scholar
  23. Kovacs EK, Pataki G (2016) The participation of experts and knowledges in the Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES). Environ Sci Policy 57:131–139CrossRefGoogle Scholar
  24. Latour B (1987) Science in action: how to follow scientists and engineers through society. Open University Press, Milton KeynesGoogle Scholar
  25. Latour B (1999) Pandora’s hope: essays on the reality of science studies. Harvard University Press, CambridgeGoogle Scholar
  26. Liftin KT (1998) Satellites and sovereign knowledge: remote sensing of the global environment. In: Liftin KT (ed) The greening of sovereignty in world politics. MIT Press, Cambridge, pp 193–222Google Scholar
  27. Lofmarck E, Lidskog R (2017) Bumping against the boundary: IPBES and the knowledge divide. Environ Sci Policy 69:22–28CrossRefGoogle Scholar
  28. Lundquist CJ, Pereira HM, Alkemade R, den Belder E, Carvalho Ribeiro S, Davies K, Greenaway A, Hauck J, Karlsson-Vinkhuyzen S, Kim H, King N, Lazarova T, Pereira L, Peterson G, Ravera F, van den Brink T, Argumedo A, Arida C, Armenteras D, Ausseil AG, Baptiste B, Belanger J, Bingham K, Bowden-Kerby A, Cao M, Carino J, van Damme PA, Devivo R, Dickson F, Dushimumuremyi JP, Ferrier S, Flores-Díaz A, Foley M, Garcia Marquez J, Giraldo-Perez P, Greenhaigh S, Hamilton DJ, Hardison P, Hicks G, Hughey K, Kahui-McConnell R, Karuri-Sebina G, De Kock M, Leadley P, Lemaitre F, Maltseva E, de Mattos Scaramuzza CA, Metwally M, Nelson W, Ngo H, Neumann C, Norrie C, Perry J, Quintana R, Rodriguez Osuna VE, Roehrl C, Seager J, Sharpe H, Shortland T, Shulbaeva P, Sumaila UR, Takahashi Y, Titeux N, Tiwari S, Trisos C, Ursache A, Wheatley A, Wilson D, Wood S, van Wyk E, Yue TX, Zulfikar D, Brake M, Leigh D, Lindgren-Streicher P (2017) Visions for nature and nature’s contributions to people for the 21st century. NIWA Science and Technology Series Report No. 83, NIWA: New ZealandGoogle Scholar
  29. Mahony M, Hulme M (2016) Epistemic geographies of climate change: science, space and politics. Prog Hum Geogr 42:1–30Google Scholar
  30. Meuser M, Nagel U (2009) The expert interview and changes in knowledge production. In: Bogner A, Littig B, Menz W (eds) Interviewing experts. Palgrave Macmillan, New York, pp 17–42CrossRefGoogle Scholar
  31. Miller CA (2007) Democratization, international knowledge institutions, and global governance. Governance 20(2):325–357CrossRefGoogle Scholar
  32. Montana J (2016) How IPBES works: the functions, structures and processes of the Intergovernmental Platform on Biodiversity and Ecosystem Services. C-EENRG Working Papers 2016-2, Cambridge Centre for Environment, Energy and Natural Resource Governance, University of Cambridge, pp 1–23Google Scholar
  33. Montana J (2017) Accommodating consensus and diversity in environmental knowledge production: achieving closure through typologies in IPBES. Environ Sci Policy 68:20–27CrossRefGoogle Scholar
  34. Obermeister N (2017) From dichotomy to duality: addressing interdisciplinary epistemological barriers to inclusive knowledge governance in global environmental assessments. Environ Sci Policy 68:80–86CrossRefGoogle Scholar
  35. Oteros-Rozas E, Martín-López B, Daw T, Bohensky EL, Butler J, Hill R, Martin-Ortega J, Quinlan A, Ravera F, Ruiz- Mallén I, Thyresson M, Mistry J, Palomo I, Peterson GD, Plieninger T, Waylen KA, Beach D, Bohnet IC, Hamann M, Hanspach J, Hubacek K, Lavorel S, Vilardy S (2015) Participatory scenario planning in place-based social-ecological research: insights and experiences from 23 case studies. Ecol Soc 20(4):32CrossRefGoogle Scholar
  36. Pascual U, Balvanera P, Díaz S, Pataki G, Roth E, Stenseke M, Watson RT, Başak Dessane E, Islar M, Kelemen E, Maris V, Quaas M, Subramanian SM, Wittmer H, Adlan A, Ahn S, Al-Hafedh YS, Amankwah E, Asah ST, Berry P, Bilgin A, Breslow SJ, Bullock C, Cáceres D, Daly-Hassen H, Figueroa E, Golden CD, Gómez-Baggethun E, González-Jiménez D, Houdet J, Keune H, Kumar R, Ma K, May PH, Mead A, O’Farrell P, Pandit R, Pengue W, Pichis-Madruga R, Popa F, Preston S, Pacheco-Balanza D, Saarikoski H, Strassburg BB, van den Belt M, Verma M, Wickson F, Yagi N (2017) Valuing nature’s contributions to people: the IPBES approach. Curr Opin Environ Sustain 26–27:7–16CrossRefGoogle Scholar
  37. Pearce W, Mahony M, Raman S (2018) Science advice for global challenges: learning from trade-offs in the IPCC. Environ Sci Policy 80:125–131CrossRefGoogle Scholar
  38. Peterson GD, Harmackova ZV, Meacham M, Queiroz C, Jiménez Aceituno A, Kuiper JJ, Malmborg K, Sitas NE, Bennett EM (2018) Welcoming different perspectives in IPBES: “Nature’s contributions to people” and “Ecosystem services”. Ecol Soc 23(1):39CrossRefGoogle Scholar
  39. Rosa IMD, Pereira HM, Ferrier S, Alkemade R, Acosta LA, Akcakaya R, den Belder E, Fazel AM, Fujimori S, Harfoot M, Harhash KA, Harrison PA, Hauck J, Hendriks RJJ, Hernández G, Jetz W, Karlsson-Vinkhuyzen SI, Kim H, King N, Kok MTJ, Kolomytsev GO, Lazarova T, Leadley P, Lundquist CJ, Márquez JG, Meyer C, Navarro LM, Nesshöver C, Ngo HT, Ninan KN, Palomo MG, Pereira LM, Peterson GD, Pichs R, Popp A, Purvis A, Ravera F, Rondinini C, Sathyapalan J, Schipper AM, Seppelt R, Settele J, Sitas N, van Vuuren D (2017) Multiscale scenarios for nature futures. Nat Ecol Evol 1:1416–1419CrossRefGoogle Scholar
  40. Shapin S (1998) Placing the view from nowhere: historical and sociological problems in the location of science. Trans Inst Br Geogrphers NS 23(1):5–12CrossRefGoogle Scholar
  41. Star SL, Griesemer J (1989) Institutional ecology, “translations” and boundary objects: amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–39. Soc Stud Sci 19:387–420CrossRefGoogle Scholar
  42. Stenseke M (2016) The intergovernmental science-policy platform on biodiversity and ecosystem services and the challenge of integrating social sciences and humanities. Bull Geogr Soc Econ Ser 33:119–129Google Scholar
  43. Stenseke M, Larigauderie A (2017) The role, importance and challenges of social sciences and humanities in the work of the intergovernmental science-policy platform on biodiversity and ecosystem services (IPBES). Innov Eur J Soc Sci Res 31:S10–S14CrossRefGoogle Scholar
  44. Sutherland WJ, Gardner TA, Haider LJ, Dicks LV (2013) How can local and traditional knowledge be effectively incorporated into international assessments? Oryx 48(1):1–2CrossRefGoogle Scholar
  45. Tengo M, Hill R, Malmer P, Raymond CM, Spierenburg M, Danielsen F, Elmqvist T, Folke C (2017) Weaving knowledge systems in IPBES, CBD and beyond—lessons learned for sustainability. Curr Opin Environ Sustain 26–27:17–25CrossRefGoogle Scholar
  46. Thaman R, Lyver P, Mpande R, Perez E, Cariño J, Takeuchi K (eds) (2013) The contribution of indigenous and local knowledge systems to IPBES: building synergies with science. UNESCO, IPBES Expert Meeting Report, UNESCO/UNU, Paris, pp 1–84Google Scholar
  47. Turnhout E, Stuiver M, Klostermann J, Harms B, Leeuwis C (2013) New roles of science in society: different repertoires of knowledge brokering. Sci Public Policy 40:354–365CrossRefGoogle Scholar
  48. Turnhout E, Dewulf A, Hulme M (2016) What does policy-relevant global environmental knowledge do? The cases of climate and biodiversity. Curr Opin Environ Sustain 18:65–72CrossRefGoogle Scholar
  49. UNEP (2013) Consideration of initial elements: recognizing indigenous and local knowledge and building synergies with science. Information document 5 (Annex), plenary of the intergovernmental science-policy platform on biodiversity and ecosystem services, First Session, pp 1–6Google Scholar
  50. Vadrot ABM, Rankovic A, Lapeyre R, Aubert P-M, Laurans Y (2018) Why are social sciences and humanities needed in the works of IPBES? A systematic review of the literature. Innov Eur J Soc Sci Res 31(sup1):S78–S100CrossRefGoogle Scholar

Copyright information

© Springer Japan KK, part of Springer Nature 2018

Authors and Affiliations

  1. 1.LondonUK

Personalised recommendations