【正文】 F ENVIRONMENTAL HARM 2 Criminal statutes address specific social harms. The act element of a criminal statute articulates the social harm at which the crime is directed. The mentalstate element articulates the attitude a defendant must have had toward the social harm in order to be criminally While mental state is a crucial element in determining criminal liability, the analysis in this article focuses exclusively on the act element of environmental crimes. Focusing on the act element in a 1995 article, Michael Faure and Marjolein Visser proposed and examined four models of environmental crimes. First is Abstract Endangerment, a model criminalizing disobedience to administrative rules and requirements perse . Second is Concrete Endangerment Crimes with Administrative Predicates (“Concrete Endangerment”). Concrete Endangerment crimes involve behavior that both violates regulatory law and poses a threat of harm to the environment。 thus, on the surface, at least, these crimes target two social harms. Crimes in the third model, Serious Environmental Pollution, punish very serious environmental harm even if the activity at issue was not otherwise unlawful。 even a Concrete Endangerment crime is off limits if the threat is viewed in the short term. But this result seems wrong. Environmental harm has, in fact, occurred, as eliminating the wetland certainly has dire shortterm consequences for the flora and fauna in the ecosystem. If we are to use the criminal sanction to address this situation through anything other than Abstract Endangerment crimes, we will have to articulate what we mean by harm to the environment without using human values as a surrogate. A legislature that wants to address a full spectrum of environmental harm must articulate a standard that reflects purely environmental values. The statute must require proof of something more than mere contact between a pollutant and the environment。 makes excavations or heaps, creates, alters or removes bodies of water。 clears a forest。 however this provision adds that the activity must “thereby interfere not insubstantially” with the interest in question, making the definition less useful than it might have been. Openended definitions of environmental harm may be attractive to legislators, but they also create problems. One problem arises if the statute allows the fact finder to equate “harm” with any negative change in the quality of water, air, or soil no matter how minor. This is because any contact between a pollutant and the environment is likely to cause some minor negative change in the latter. Such an interpretation of “harm” conflates Model II, involving a threat of “harm” and Model III, which requires actual “harm.” A coherent environmental criminal scheme should guide fact finders in distinguishing between these degrees of seriousness. 2. The Issue of Causation Concrete harm crimes require the prosecution to prove that the defendant’s behavior caused environmental harm, however that is defined. Serious Environmental Harm crimes—some of those in Model IV, to be discussed below—carry the same requirement. Proving causation is not particularly difficult in the case of a single polluting event that results in clear damage. However, the requirement could present a challenge to prosecutors in other situations, reducing the number of cases in which these resultdefined crimes are useful as tools of environmental protection. Causation problems are likely to be mon in prosecutions requiring proof of actual harm. This is because in many situations a defendant may be able to show that one or more 5 additional actors independently emitted pollutants into the same water, soil, or air, either previously, simultaneously, or subsequently to the defendant’s own actions. Of course, similar causation problems also arise in the context of traditional crimes with result elements, most notably in the prosecution of homicides. This being so, any jurisdiction with a developed criminal law will have worked out solutions to at least the most mon of these problems, and there is likely to be informative scholarly material on any issues still outstanding. The main challenge for pollution crimes is to translate the existing solutions from the vocabulary of homicide (or assault, or other more traditional resultdefined crimes) to that of environmental harm. It is valid to ask whether there is really a need to undertake the task of developing a body of causation law in the context of environmental crimes. Why not be content to use Abstract and Concrete Endangerment crimes, which do not require proof that the defendant caused a concrete result? The answer is related to our suggestion of a graduated punishment approach, discussed in Section III. Under such an approach, Concrete Harm crimes would be punished more severely than Abstract or Concrete Endangerment crimes (and Serious Environmental Harm crimes might carry even more severe punishment). Where proof of causation is difficult or impossible, conviction for the lower offense would have to suffice. In