For those going into the legal profession, and for those curious about the logic of legal procedures, this book offers and excellent introduction. Its length will suit those readers who want a quick but accurate overview of this topic. It might also satisfy the needs of a reader in a field of endeavor somewhat removed from the legal profession, namely, that of artificial intelligence. Artificial and computational intelligence has been applied to the legal profession with the goal of creating automated legal reasoning machines. My interest in the book was somewhat different when I first read it years ago, but now has been reactivated from the standpoint of artificial intelligence. The author describes his book as an attempt to give a general description of the process of legal reasoning in case law and in statutorial and constitutional interpretation. He emphasizes right at the beginning that the law should not be viewed as a known system of rules that are applied by a judge, that legal rules are never clear, and that a requirement for such clarity would make society impossible. Ambiguity in the rules he says, allows collective participation to resolve the ambiguity. Such a characterization of legal rules by Levi prohibits an axiomatic or formal approach to legal reasoning, and this will make the problem of creating automated legal reasoners much more difficult. Interestingly, Levi quotes Aristotle in asserting that the pattern of legal reasoning consists of reasoning by example, and that it follows a three-step process: With the doctrine of precedent assumed throughout, a proposition describing a particular case is made into law and then this rule is applied to a situation that is similar to these. Thus: 1. The cases are shown to be similar. 2. The rule of law in the first case is announced 3. This rule is then applied to the second case. It is the finding of similarity in both cases that would entail, in the context of a legal reasoning machine, the use of data mining techniques coupled with a formalization of the Aristotelian "reasoning by example" that is discussed by Levi. An interesting part of Levi's discussion on the determination of similarity (or difference) is that this determination is dependent on the judge. When a statute does not exist, and case law is being considered, the judge is free to dismiss teh facts that the prior judge may have considered important. The doctrine of "ratio decidendi" in legal philosophy, and has been the subject of intense investigation. A formal or computational model of ratio decidendi must come to terms with this tension between precedent and judicial discretion. In addition, Levi argues, the rules are discovered in the process of determining the similarities (or differences). This dynamism in legal reasoning is not too different from what happens in scientific research, for in the latter new laws (rules) are discovered in the process of interpreting new data that has been acqu... [ Read More (0.3k in body) ]