Evidence-based-policy-based-evidence on forests, trees and agroforestry

By Meine van Noordwijk

The Centre’s chief science advisor argues that researchers need to address more complex issues of policy and evidence if they want to have a real impact.

Interest in evidence-based policy alternates with periods where policy-based evidence dominates. However, it is naive to expect that the former can exist in isolation. Historically, efforts to compile and augment evidence have been carried out and/or been paid for by those who accumulated wealth and power. The evidence available is thus coloured and not neutral.

Existing evidence on ‘forest’ is based on a long tradition of distinguishing ‘forest’ (F) from ‘non-forests’ (NF). Properties of at least some members of the F class are supposed to represent all, and can be contrasted with those of at least some members of the NF class. The difference may be taken as ‘evidence’ for the continued policy relevance of F.

However, there are many types of F, many types of NF and at least some NF matches at least some properties of the ‘ideotype’ of F (for example, agroforests that match the biodiversity of natural forests); also, at least some F matches at least some properties of the ideotype of NF (for example, plantation forestry as opposed to tree–crop agriculture). The existence of a difference in mean between F and NF cannot, in that case, be used as evidence for categorical policy decisions regarding F and NF. Rather, we may progress faster by using more detailed classification in the F–NF continuum, figure out which properties matter for policies and compile evidence accordingly.

The ‘forest’ versus’ non-forest’ distinction exists in two quite separate realms: 1) an ecological/biophysical one where the degree of dominance of woody perennials in vegetation is associated with many properties, ecosystem functions and services; and 2) a social/political/institutional one where forests were distinguished from village lands and put under the control of a local lord or king, a role later taken over by government.

There is evidence for both types of F–NF distinctions, but it isn’t always clean and clear which is which. Government-reported F data, for example, as a basis for global forest assessments, have been marred by inconsistencies of interpretation. Trees outside forest sometimes occur in dense stands that would be, based on biophysical criteria, classified as F; but they are not under the institutional control of F agencies, and their stakeholders/managers want to keep it that way.

Agroforestry exists on the F–NF interface of both the ecological/biophysical and the social/political/institutional perspectives. Rather than carving out an agroforestry niche that has boundaries to worry about with both the F and the NF worlds, we should be concerned by the disfunctionality of the F–NF dichotomy and argue for a more evidence-based approach to the ways landscapes (with a mosaic of F and NF elements, plus agroforestry) function, what this means for all stakeholders and for the way decisions are made. Once we understand the current complexity, entry points for change from the status quo can be identified and coalitions formed to influence change. More than a decade ago this approach was termed Negotiation Support Systems and has gradually found traction[1].

Clark et al (2011) introduced simple, two-way tables to describe the opportunities for ‘linking knowledge with action’ for natural resources management and characterized efforts across the six lower cells: K1A0 (Disciplinary research aimed at general enlightenment),KnA0 (Interdisciplinary, multiple knowledge systems research aimed at general enlightenment), K1A1 (Decision Support Systems from single disciplinary perspective), KnA1 (Decision Support Systems from interdisciplinary perspective), K1An (joint fact finding, political boundary work), KnAn (Negotiation Support Systems). Evidence-based policy initiatives generally target the central cell of the table, aiming for synthesized science that can be agreed on by everybody as a sound basis for decisions in the best interest of all. As spelled out by Clark et al (2011), the criteria for ‘good science’ focus on credibility in the first column, to credibility plus salience in the second and credibility plus salience plus legitimacy in the third.












Whether we like it or not, most of the contested forest–agriculture mosaics of the developing world are in the messy KnAn cell, where all claims to ‘knowledge’ are seen as politically constructed, where the knowledge = po­wer and power = knowledge concepts thrive and where efforts at enhancing evidence-based policy come across as naive and highly political by seeming to ignore politics.

The reformed CGIAR is driven by funding opportunities of donors who want to avoid K1A0 and KnA0 forms of ‘pure’ science and ensure action, outcomes and impacts as a direct and tractable result of research outputs and new knowledge. The presumed clarity of impact pathways of an A1 world with a dominant decision maker, however, falls apart in the more complex reality of An systems, where political outcomes are less predictable and direct efforts to influence them are beyond the ethical reach of international research organizations. Yet, if research funding is channelled to the nice short-term success stories of K1A1, we may continue to ignore the more important but messy Kn An issues that will matter most in the medium to longer terms.

We need to break the positive feedback loop between evidence-based policy and policy-based evidence and take a deep dive into multiple-actors, multiple-knowledge and multiple-stakes landscapes, where forests, trees and agroforestry concepts are vehicles of debate about what kind of landscapes, and world, people want to live in.



[1] Van Noordwijk M, Tomich TP, Verbist B. 2001. Negotiation support models for integrated natural resource management in tropical forest margins. Conservation Ecology 5(2): 21. Available from http://www.consecol.org/vol5/iss2/art21.

Further reading

Van Noordwijk M, Catacutan DC, Clark WC, 2011. Linking scientific knowledge with policy action in natural resource management. ASB Policy Brief 19. Nairobi: ASB Partnership for the Tropical Forest Margins.

Clark WC, Tomich TP, van Noordwijk M, Guston D, Catacutan D, Dickson NM, McNie E. 2011. Boundary work for sustainable development: natural resource management at the Consultative Group on International Agricultural Research (CGIAR). Proceedings of the National Academy of Sciences. doi:10.1073/pnas.0900231108

Petrokofsky G, Kanamaru H, Achard F, Goetz SJ, Joosten H, Holmgren P, Lehtonen A, Menton M, Pullin AS, Wattenbach M. 2012. Comparison of methods for measuring and assessing carbon stocks and carbon stock changes in terrestrial carbon pools. How do the accuracy and precision of current methods compare? A systematic review protocol. Environmental Evidence 1:6. doi:10.1186/2047-2382-1-6.




This post relates to the CGIAR Research Program on Forests, Trees and Agroforestry

Rob Finlayson

Robert Finlayson is the Southeast Asia program's regional communications specialist. As well as writing stories for the Centre's website, he devises and supervises strategies for projects and the countries in the Southeast Asia region, including scripting and producing videos, supervising editors and translators and also assisting with resource mobilization.

You may also like...