The agricultural matrix and a future paradigm for conservation.

Land-use policies, especially in tropical zones, are frequently seen as facing three insoluble problems: (1) low or inefficient agricultural production, which leads to (2) clearing of natural habitat for more farmland, which leads to (3) loss of biodiversity (Ausebel 1996; Glokany 1998; Trewavas 2001). The solution to this three-part dilemma is often seen by going backward, so to speak (1) loss of biodiversity can be stemmed if (2) farmers stop clearing natural habitat, which can be done if (3) agricultural production is intensified on land already devoted to agricultural production. The seemingly inescapable conclusion is that intensifying agriculture already underway is the main solution to the problem of biodiversity loss (Green et al. 2005). We argue that this seemingly impeccable framework is wrong and that a more realistic framing suggests a view of conservation that is distinct from more traditional forms. In our view this more-realistic framing must acknowledge the conservation consequences of the agricultural matrix in a landscape framework. Here, we focus on the issue of land-use policies, especially on the ecological relationship between natural areas and the surrounding agricultural matrix. The way the problem is frequently framed, as a zero-sum game where competition for land is divided between production and conservation, is wrong, as noted by Angelsen and Kaimowitz (2001 and chapters therein). The problem with this framing is that it tacitly assumes a closed system. What normally happens in practice is that when agriculture is intensified in one corner of the world, that very intensification acts as a magnet to attract more producers to the area, which actually leads to more habitat conversion, as demonstrated by Angelson and Kaimowitz (2001). The simple three-part logical sequence is theoret-