【正文】 ly applies in these kinds of cases as soon as the administrative provision has been violated, even if no actual harm or threat of harm to the environment occurs. If the criminal provision requires or presumes environmental harm or the threat of such harm, the statute is not of the Abstract Endangerment variety. Although Abstract Endangerment crimes focus on vindicating administrative values, punishing the administrative violation indirectly furthers ecological values in two ways. First, an entity that follows administrative rules is less likely to harm the environment. More to the point, if administrative rules are followed, the regulatory agency can monitor the entity’s operations to ensure that harm is less likely to occur. Nevertheless, although environmental values are implicated by Abstract Endangerment crimes, the overlap with such values is inplete. For one thing, an entity in pliance with all administrative rules can still cause environmental “harm.” Consider that, for each parameter—air, water, soil— the administrative agency will set a baseline of “acceptable” contact between a pollutant and the environment. This baseline will reflect a promise among such considerations as the pollutant’s effect on the environment, society’s need for the polluting activity, and the existence (and cost) of technology that can mitigate the damage. Thus, pliance with the baseline does not mean a lack of environmental “harm.” However, the disconnect between administrative and environmental values can go the other way as well. That is, an entity that violates administrative rules may not be causing environmental harm. Take, for example, an entity that transports hazardous waste without the required paperwork. 3 This violation harms administrative norms first and foremost. But there is no emission, and thus no environmental harm: as regards environmental values, the crime is inchoate. Regardless of which aspect of the disconnect one views, the failure of Abstract Endangerment crimes to focus on an activity’s impact on the environment makes this model less than pletely effective in protecting ecological values. B. Model II: Concrete Endangerment Crimes with Administrative Predicates The second model is that of Concrete Endangerment Crimes with Administrative Predicates (“Concrete Endangerment”). As with the first model, the activity in question must take place in an unlawful way by engaging in the activity without a required permit or other authorization or by violating conditions in a statute, regulation, or However, the characteristic of unlawfulness may be integrated in different ways. Some of the crimes in this model include as an element the fact of violating regulatory law. Others provide the defense that the activity was Even though the legal technique is different, the unlawfulness of the discharge remains a factor in criminality. C. Model III: Concrete Harm Crimes with Administrative Predicates The third model is Concrete Harm Crimes with Administrative Predicates (“Concrete Harm”). As noted above, this model has been added to Faure’s and Visser’s original approach. Statutes fitting this model are similar to Concrete Endangerment crimes in that they require proof that the actor violated an administrative rule. However, these crimes go beyond threats and require proof of actual environmental harm. The identification of crimes fitting Model III can be tricky, as it depends upon the definition of “environmental harm.” As we will show in Part 1 of this section, some statutory definitions focus on the environment directly, but others adopt an anthropocentric definition of “environmental harm.” These latter seem based on the premise that emissions or releases that threaten or harm human health, safety, or other interests must of necessity also harm the environment. To the extent that this interpretation is accurate, the “knowing endangerment” provisions of the CWA and RCRA are examples of Model III crimes. In addition, Europe provides a number of examples of Concrete Harm crimes. Because we are working toward a graduated punishment approach to environmental crimes, it makes sense to us to differentiate Concrete Harm statutes from the crimes in Model II. Concrete Harm crimes require proof of actual harm, and so it is logical for such statutes to impose higher penalties than those requiring merely a threat of harm. When the government can prove both an administrative violation and actual environmental harm, the authorization and imposition of increased punishment would be expected to further the vindication of environmental values through increased deterrence and However, Concrete Harm statutes run into conceptual and proof problems that can frustrate these goals. For one thing, the concept of environmental “harm” is difficult to define. A second problem involves causation. We will explore these in turn. 1. Defining Environmental “Harm” A traditional way of measuring environmental harm is to look through the lens of harm to human beings instead of focusing on the environment itself. One variation focus