完形填空 There is much discussion in the literature these days about how one "defines" sustainability, sustainable development, and related concepts. Many argue that the concept is useless because it cannot be "adequately defined". 46) Most of this discussion is misdirected because it attempts to cast the problem as definitional, when in fact it is a problem of prediction, and it fails to take into account the many time and space scales over which the concept of sustainability must apply. Defining sustainability is actually quite easy: A sustainable system is one that survives for some specified (finite) time. The problem is that one knows one has a sustainable system only after the fact. Thus, what usually pass for definitions of sustainability are actually predictions of what set of conditions will actually lead to a sustainable system. 　　47) For example, keeping harvest rates of potentially renewable resources such as trees and wildlife below rates of natural renewal should, one could argue, lead to a sustainable natural resource extraction system. But that is a prediction, not a definition. Usually there is so much uncertainty in our ability to estimate natural rates of renewal and our ability to observe and regulate harvest rates that a simple prediction such as this is always highly suspect. 　　Likewise, sustainable economic development can be observed only after the fact. 48)Most "definitions" of sustainable development include elements of a sustainable scale of the economy relative to its ecological life support system; a fair distribution of resources and opportunities between current and future generations, as well as between individuals in the current generation; and an efficient allocation of resources that adequately accounts for natural capital. Note that these three components of sustainable economic development are actually "predictors" of sustainability and not really elements of a definition. Like all predictions, they are uncertain and are subject to much discussion and disagreement. 　　49) The second problem is that when one says a system has achieved sustainability, one does not mean an infinite life span, but rather a life span that is consistent with the system''s expected time and space scale. We expect an organism to have a longer life span, the species to have an even longer life span, and the planet to last much longer. But no system is expected to have an infinite life span. 　　The real problems are not so much defining the goal as predicting what policies will lead to its achievement. Here there is ample room for, and need of, vigorous discussion, debate, analysis, and modeling in order to determine which policies have the best chance of achieving the goal of sustainability. 50) Given the huge uncertainties involved, it is of particular importance in this regard to select policies that are precautionary, that is, they do not take unnecessary risks that could decrease the chances for sustainability they do not rely on hoped-for technological fixes for their successes.