It is widely acknowledged that conservation cannot be undertaken without the support and participation of local people, and that livelihood concerns and future development goals need to be at the centre of any viable conservation strategy that involves people (Pimbert & Pretty Reference Pimbert and Pretty1995; Hulme & Murphree Reference Hulme and Murphree2001; Arnold Reference Arnold2002; McShane Reference McShane2003; Barrett et al. Reference Barrett, Lee and McPeak2005). Nevertheless, for several years a contentious exchange has been waging regarding the legitimacy of local peoples involvement in conservation efforts and protected areas (PAs) (Oates Reference Oates1999; Terborgh Reference Terborgh1999; Pimm et al. Reference Pimm, Ayres, Balmford, Branch, Brandon, Brooks, Bustamante, Costanza, Cowling, Curran, Dobson, Farber, da Fonseca, Gascon, Kitching, McNeely, Lovejoy, Mittermeier, Myers, Patz, Raffle, Rapport, Raven, Roberts, Rodrıguez, Rylands, Tucker, Safina, Samper, Stiassny, Supriatna, Wall and Wilcove2001; Wilkie et al. Reference Wilkie, Morelli, Demmer, Starkey, Telfer and Steil2006) and the impacts of PAs on local communities (Adams Reference Adams2004; Chapin Reference Chapin2004; Hutton et al. Reference Hutton, Adams and Murombedzi2005; Brockington & Igoe Reference Brockington and Igoe2006; Cernea & Schmidt-Soltau Reference Cernea and Schmidt-Soltau2006). Yet, despite scores of publications supporting different view points, the global debate continues to be highly dichotomous, with repeated cycles of ‘irresistible dialectic’ (Redford & Painter Reference Redford and Painter2006, p. 3) pitting strict conservation against human well-being. As in most disagreements, this exchange is characterized in great part by polarized one-sided presentation of arguments by conservation biologists and social scientists, where both sides selectively use information to support their viewpoints.

This debate could be regarded as a regular scientific squabble with minor impacts on the world. However, it is important to acknowledge the power that such conceptual debates have in shaping policies, institutional programmes and funding streams for conservation and development efforts. Therefore, we take the opportunity in this comment to examine some aspects of this debate in further detail, and to suggest a way of examining and evaluating conservation and development outcomes.

Protectionism and poverty

The dominant colonial approach to conservation was the establishment of PAs from which people were essentially excluded or resettled, often forcibly (Adams Reference Adams2004; Hutton et al. Reference Hutton, Adams and Murombedzi2005), and there are still many ‘protectionist’ views in the conservation field (Gartlan Reference Gartlan, Eves, Hardin and Rupp1998; Oates Reference Oates1999; Terborgh Reference Terborgh1999; Terborgh et al. Reference Terborgh, Van Schaik, Davenport and Rao2002). The PA network is extensive; there are presently more than 100 000 PAs, constituting 11.5% of the terrestrial surface of the Earth (Naughton-Treves et al. Reference Naughton-Treves, Holland and Brandon2005). However, considerable social conflicts result from conservation initiatives and, despite protection measures, in most areas human impacts continue relatively unabated and local non-compliance is customary (Robbins et al. Reference Robbins, McSweeney, Waite and Rice2006). At the same time, tropical forest landscapes are often marked by dire poverty. For example, in the vast Tri-National de la Sangha area of Cameroon, Congo and the Central Africa Republic, an estimated 75% of the population live below the poverty line of US$ 1 per day (Sayer et al. Reference Sayer, Campbell, Petheram, Aldrich, Perez, Endamana, Nzooh Dongmo, Defo, Mariki, Doggart and Burgess2007). Undertaking conservation in such landscapes is not tenable if it further contributes to marginalization and impoverishment and disregards the development needs of local people, as has been previously argued (Cernea & Schmidt-Soltau Reference Cernea and Schmidt-Soltau2006; Campese et al. Reference Campese, Borrini-Feyerabend, de Cordova, Guigner and Oviedo2007). There are valid ethical and biological arguments for biodiversity conservation at the regional and global levels, but it is clearly unreasonable to expect the rural poor to incur the opportunity costs, restrictions and potential harm that are often imposed by global conservation programmes (Arjunan et al. Reference Arjunan, Holmes, Puyravaud and Davidar2006).

Integrated conservation and development

In recognition of the diverse problems associated with protectionism, traditional means of conserving biodiversity have made way in the past twenty years to a more participatory and people-centred approach to what became known as integrated conservation and development projects (ICDPs; McShane & Wells Reference McShane and Wells2004). ICDPs were ‘designed’ to reduce poverty and improve incomes levels, nutrition, health care and education, as well as to conserve biodiversity (Christensen Reference Christensen2004). Despite virtually no field testing, and flush with funding that had traditionally gone to development projects, ICDPs quickly became the standard approach in attempting to combine the conservation of biodiversity with community development in PAs (Western & Wright Reference Western and Wright1994; Wells et al. Reference Wells, McShane, Dublin, O'Connor, Redford, McShane and Wells2004).

However, a number of reviews have suggested that ICDPs have not reconciled conservation and development agendas (Alpert Reference Alpert1996; Chapin Reference Chapin2004; Christensen Reference Christensen2004). While positions have varied tremendously, both conservationists and social scientists have harshly criticized ICDPs, contributing to their declining popularity and a disenchantment with community-based approaches (Neumann Reference Neumann1998; Agrawal & Gibson Reference Agrawal and Gibson1999; Wells et al. Reference Wells, McShane, Dublin, O'Connor, Redford, McShane and Wells2004). While conservationists have complained that community-based approaches do not provide enough protection for endangered species and environments and fail to achieve conservation goals (Kremen et al. Reference Kremen, Merenlander and Murphy1994; Oates Reference Oates1999), social scientists have argued that ICDPs function in the same moulds as conventional development projects and result in highly inequitable outcomes that severely limit local people's rights (Schmidt-Soltau Reference Schmidt-Soltau2004) potentially becoming poverty traps for local communities (Dove Reference Dove1993). Chapin (Reference Chapin2004, p. 20) suggested that one key reason for this was that ‘on the ground, ICDPs were generally paternalistic, lacking in expertise and one-sided: driven by the agenda of the conservationists with little indigenous input’. The large non-governmental organizations (NGOs) involved in biodiversity conservation and, by implication, the nation states they support in land-use management, found themselves under scrutiny as never before (Chapin Reference Chapin2004; Romero & Andrade Reference Romero and Andrade2004). While large NGOs have become increasingly powerful and outspoken in the last decade owing to their access to financial capital, they have also been targets of strident critique and have needed increasingly to demonstrate that they operate with greater equity and include local stakeholders in a meaningful way (Romero & Andrade Reference Romero and Andrade2004; Alcorn & Royo Reference Alcorn and Royo2007).

Several authors question the inherent assumptions within ICDPs, which focus on local people as the agents of environmental destruction while disregarding larger-scale extraction of resources by loggers, mining, cattle ranching and industrial agriculture (Dove Reference Dove1993; Nepstad et al. Reference Nepstad, Capobianco, Barros, Carvalho, Moutinho, Lopes, Lefebvre and Ernst2000; Christensen Reference Christensen2004; Sayer & Campbell Reference Sayer and Campbell2004). This failure of ICDP managers to address the larger issue surrounding conservation in the tropics by focusing only on local, and arguably minor, threats has contributed to their lack of success (Christensen Reference Christensen2004) and further alienates local constituent support for conservation activities (Schmidt-Soltau Reference Schmidt-Soltau2004). However, many of these authors also underplay the impact of human use on forest environments at a local scale, adhering to notions of local conservation agency (for example Cernea & Schmidt-Soltau Reference Cernea and Schmidt-Soltau2006).

Polarized debates

Despite the clear paradigm shift to attempt to take human welfare into account, considerable debate has recently resurfaced regarding the impacts of conservation on human welfare and human rights (Cernea & Schmidt-Soltau Reference Cernea and Schmidt-Soltau2006; Brockington & Igoe Reference Brockington and Igoe2006; Campese et al. Reference Campese, Borrini-Feyerabend, de Cordova, Guigner and Oviedo2007). Such polarized views are hard to reconcile. For example, it has recently been reported that there are up to 170 000 ‘conservation refugees’ in Central Africa alone (Cernea & Schmidt-Soltau Reference Cernea and Schmidt-Soltau2006), yet conservationists working in the region emphatically refute these figures (Maisels et al. Reference Maisels, Sunderland, Curran, von Liebenstein, Oates, Usongo, Dunn, Asaha, Balinga, Defo, Telfer, Redford and Fearn2007). Likewise, a global review of ‘conservation evictions’ presented a table of demographic figures from the 13 newly-created national parks of Gabon (Brockington & Igoe Reference Brockington and Igoe2006). While the Wildlife Conservation Society stated there were no people settled within the boundaries of these parks an independent researcher was quoted as suggesting there were almost 7000 people living within these new PAs (Brockington & Igoe Reference Brockington and Igoe2006).

A similar dichotomy has characterized heated debate over the legitimacy of local people in Amazonian PAs, where extractive reserves have been used as an example of both the failure and success of people-based conservation. In a resurgence of the protectionist paradigm, several authors (Terborgh Reference Terborgh1999, Reference Terborgh2000; Redford & Sanderson Reference Redford and Sanderson2000) have challenged the conservation capacity and resource-use rights of local communities, focusing on a local scale and arguing for more extreme measures because of smallholder threats to biodiversity, namely wildlife (Pimm et al. Reference Pimm, Ayres, Balmford, Branch, Brandon, Brooks, Bustamante, Costanza, Cowling, Curran, Dobson, Farber, da Fonseca, Gascon, Kitching, McNeely, Lovejoy, Mittermeier, Myers, Patz, Raffle, Rapport, Raven, Roberts, Rodrıguez, Rylands, Tucker, Safina, Samper, Stiassny, Supriatna, Wall and Wilcove2001), while ignoring larger-scale threats to conservation. Defenders of local people's involvement in conservation argue on a larger scale that people-based PAs fulfil their conservation objectives by preserving forest cover and serving as deforestation buffers (Schwartzmann et al. Reference Schwartzmann, Moreira and Nepstad2000a, Reference Schwartzmann, Nepstad and Moreirab; Zimmerman et al. Reference Zimmerman, Peres, Malcolm and Turner2001; Almeida Reference Almeida2002; Nepstad et al. Reference Nepstad, Schwartzman, Bamberger, Santilli, Ray, Schlesinger, Lefebvre, Alencar, Prinz, Fiske and Rolla2006). At the national level, the criticism of extractive reserves becomes even harsher and almost solely based on hearsay; Brazilian conservationists polemically write about the ‘unbearable lightness of extractive reserves’ (Costa Reference Costa2004) that ‘suffocate the forest’ (Correa Reference Correa2004), without having actually having set foot in an extractive reserve to collect evidence. These examples underline how polemic argumentation and often anecdotal evidence are selectively used by both sides of the debate to support a hypothesis or ethical standpoint without sufficient on-the-ground verification.

Given the huge volume of articles in the conservation arena significant advances should have been made, yet there is a continued reluctance for social and biological scientists to engage in a meaningful way, with few exceptions (Brosius & Russell Reference Brosius and Russell2003; Adams et al. Reference Adams, Aveling, Brockington, Dickson, Elliot, Hutton, Roe, Vira and Woolmer2004; Agrawal & Redford Reference Agrawal and Redford2006), despite encouragement towards interdisciplinary approaches to conservation and development (see Campbell Reference Campbell2005). A rich body of literature has been produced on techniques and approaches for integrated natural resources management, participatory resource management and evaluation of conservation success (see Sayer & Campbell Reference Sayer and Campbell2004). Mulder (Reference Mulder2007, p. 904) suggested that social scientists had the tools and expertise to ‘predict the kinds of inequities that so frequently result from conservation projects and make recommendations for wiser and forward looking interventions’. Yet, these tools are not used in most projects, as conservation biologists generally do not encourage their integration into project planning and implementation.

Which way forward?

As researchers who drive the debates which influence intervention strategies, we have a responsibility to contribute to this subject by moving beyond polemic, anecdotal evidence and prolonged argument, towards a common framework in which all parties can examine and appreciate the multiple perspectives of conservation and development efforts. For this, careful field-based, in-depth and multi-disciplinary research are needed to provide a contextualized analysis of conservation scenarios from multiple and balanced perspectives (Chan et al. Reference Chan, Pringle, Ranganathan, Boggs, Chan, Ehrlich, Haff, Heller, Al-Khafaji and Macmynowski2007; see also Sayer et al. Reference Sayer, Campbell, Petheram, Aldrich, Perez, Endamana, Nzooh Dongmo, Defo, Mariki, Doggart and Burgess2007). We make four suggestions for further research development.

Firstly, systematic comparison of a large(r) number of cases is required. While case studies are important to the understanding of the complexity of each situation and the issues involved, they provide a limited basis for broader generalizations regarding the success and failure of integrated conservation and development strategies. To move beyond this, more systematic comparative information needs to be gathered on different conservation and livelihood outcomes in a larger number of different types of PAs (for example Agrawal & Redford Reference Agrawal and Redford2006; Nepstad et al. Reference Nepstad, Schwartzman, Bamberger, Santilli, Ray, Schlesinger, Lefebvre, Alencar, Prinz, Fiske and Rolla2006).

Secondly, to evaluate external versus internal threats to conservation and human well-being, it is important to examine PA trajectories within their historical and spatial context. Conservation and livelihood outcomes often depend most on contextual factors (such as a peripheral or central location, proximity to cattle ranches, oil palm plantations or duration of migration process). Blanket statements about ICDPs or particular types of PAs working or not working are therefore problematic. Analyses of the structure and design of PA efforts need to be disentangled from location and context factors.

Thirdly, in order to achieve a better spatial contextual analysis and a broader comparative basis, more remote sensing analyses should be conducted to assess how effective PAs are at conserving forest cover and functioning as deforestation buffers against external and internal threats to conservation (see Schwartzman & Zimmerman Reference Schwartzman and Zimmerman2005; Nepstad et al. Reference Nepstad, Schwartzman, Bamberger, Santilli, Ray, Schlesinger, Lefebvre, Alencar, Prinz, Fiske and Rolla2006). In association with this, in-depth ecological, ethnographic and economic analyses of the most debated aspects, such as impacts on game populations, displacement of local inhabitants and opportunity costs of local inhabitants are needed. Here again, comparing realities inside PAs to those immediately outside PAs is essential if PA impacts are to be accurately evaluated.

Finally, analysis of ICDP and PA success needs to move beyond either looking at conservation outcomes or livelihood benefits as competing criteria for evaluation. PA analysis needs to encompass a series of criteria that address a number of dimensions of conservation and development. For example, the Centre for International Forestry Research (CIFOR) has developed a toolbox aimed at assessing environmental and development outcomes in conservation landscapes (Sayer et al. Reference Sayer, Campbell, Petheram, Aldrich, Perez, Endamana, Nzooh Dongmo, Defo, Mariki, Doggart and Burgess2007; see also http://www.cifor.cgiar.org/conservation/_ref/home/index.htm), using outcome tracking methods and visioning tools that require considerable input from a wide range of stakeholders within a project, or landscape.

With such multi-disciplinary and wide-ranging contributions, hopefully the debate can be based on more solid common ground. It is only simultaneously considering these dimensions, that the costs and benefits, losers and winners, and the trade-offs of particular conservation approaches can be made explicit and more honestly negotiated in academia and conservation practice.