Is There Too Much Criminal Law?

Is there too much criminal law? Are there too many overlapping criminal statutes, covering too much conduct, resulting in sentences that are too long? Douglas Husak says yes, and in this splendid book offers an original and persuasive explanation for why that is so. His argument for a “minimalist” conception of criminal law takes an elegant form. Chapters 2 and 3, which contain Husak’s theory as to what kind of conduct can legitimately be subject to criminal sanctions, comprise the heart of the book: Chapter 2 considers what he calls his “internal constraints” on criminalization, those derived from within the criminal law itself, while Chapter 3 offers a discussion of what he calls “external constraints,” those that depend on a “normative theory imported from outside the criminal law itself.” (P. 55.) The first and last chapters function as bookends for the argument in the middle: Chapter 1 consists of a freestanding indictment of our bloated, overextended system of criminal justice, while Chapter 4 offers a critique of several alternative theories of criminalization: law and economics, utilitarianism, and legal moralism. There is so much in this book that is smart and insightful that one is tempted just to sit back and admire it. That would probably not make for a very interesting review, however. Although I will highlight a few aspects of the book that seem to me particularly successful (including Husak’s potent critique of the law-andeconomics approach to criminalization), my main focus will be on those areas where I see fault in his intricate argument. My criticisms center for the most part on formal rather than substantive aspects of his theory—not on the conclusions that he draws, most of which I agree with—but rather on the path that he takes to those conclusions. In particular, I take issue with the notion that the first four normative constraints on criminalization that he identifies—even if they are the right ones— can properly be found within the contours of criminal law itself. In addition, I offer several possible revisions to his theory (two constraints that seem to me redundant, another that might have been included but was not) and consider quibbles on several further points.