of the Senate Committee on the Judiciary in January of this year and appearing in the Congressional Record on March 12, 1975, at page E 1059. The subject of that statement-Section 1103, "Instigating Overthrow or Destruction of Government," Senate bill S. 1-has now been revised in the following way: (i) The phrase "as speedily as circumstances permit" is stricken from the description of the intent required as a part of the offense, in paragraph (a). (ii) The addition of the phrase "imminent lawless" is added to paragraph (a) (1), so that the behavior constituting the offense is defined to be inciting others to engage in imminent lawless conduct that fits a certain description. (iii) The phrase "then or at some future time" is stricken from the language of paragraph (a) (1), so that the offense is: inciting others to engage in imminent lawless conduct that would facilitate the forcible overthrow or destruction of the government. (iv) A phrase is deleted from paragrph (a) (2), and a third paragraph (a) (3) is added to the definition of the offense, so that membership in an organization knowing it has a certain purpose as opposed to membership in and of itself— constitutes the offense. (v) The grading of the offense is changed, in paragraph (b), to allow that "knowing membership" is a Class D felony. I submit that the indicated changes do very little to safeguard the political rights of American citizens; or-to say the same thing in different words-raise no difficulty for the government's ability to "get" the radical opposition. The deletion of the phrase "as speedily as circumstances permit" makes it that much easier to pin the required revolution intent upon a person. A person now becomes liable if he dreams of some revolutionary project to be carried out by future generations many centuries hence. The concept of "imminent lawless conduct" is beautiful. The authors of this change did not want to confine the offense to incitement of illegal conduct, or violent conduct, or conduct violative of one or more provisions of the criminal code or they would have said so. Instead, they fastened onto the phrase "imminent lawless conduct”—meaning presumably (for the phrase is quite obscure) conduct that verges on the illegal, conduct that perhaps might be construed as illegal, conduct that is almost but not quite illegal. Or perhaps the reference is to conduct that is not yet illegal but is about to become illegal. It might seem that the ordinary citizen receives some protection by the elimination of the phrase "then or at some future time." But note that the subjunctive mood of the verb is retained: it is conduct that would facilitate, etc., and not conduct that actually does facilitate. And there is no reference to the persons or the group for whom the conduct "would facilitate" violent overthrow. There is no requirement that such persons exist at all-no requirement that someone be actually engaged in overthrowing the government by violent means, or even planning to overthrow the government by violent means. And the verb is still "to facilitate"-not "to cause," or "to bring about," or "to enhance the likelihood of," verbs which are much to specific to be helpful in a sedition statute. It remains as true of the revised Section 1103 as of the original, that incitement to conduct as diverse as opposing gun-control legislation or calling for an end to political surveillance renders one liable to fifteen years in prison and a $100,000 fine. The changes in paragraph (a) (1) do not tie the government's hands at all. Neither do the changes in paragraph (a)(2) and the addition of (a)(3) offer any significant protection to the ordinary citizen. It is required, as a part of the offense, that the group have the proscribed incitement as a purpose; but it is not required that it be the group's sole purpose, or even a major purpose, or an important purpose. The requirement of "knowing" membership is only a slight impairment of the government's power to "get" its political enemies under the sedition statute; we know from recent history that a person can be put in jail for belonging to a group “knowing" that its purpose is X, even though he denies that its purpose is X, and all the members deny that its purpose is X, and the actual conduct of the group indicates opposition to X. All that is required is a certain repressive atmosphere. I observe that nothing in Section 1103 prevents indicting and prosecuting a person for "conspiring" to commit the offense defined therein, or for "soliciting" someone else to commit the offense. It remains easy enough to get a person for his words. Taking the standpoint of the ordinary citizen, as I must, I can perceive Section 1103 only as a monstrous engine of destruction aimed directly at the concept of the sovereignty of the people. S. 1 should never be allowed to pass the Committee on the Judiciary, let alone the Senate, containing such language. Strike Section and 1103 with the language of the Brown Commission's recommendation-the 1103 entirely; it serves no useful purpose. Alternatively, replace Sections 1102 Brown Commission's Section 1103, "Armed Insurrection." OFFICE OF THE DISTRICT ATTORNEY, COUNTY OF WESTCHESTER, Hon. JOHN L. MCCLELLAN, White Plains, N.Y., June 25, 1975. Chairman, Subcommittee on Criminal Laws and Procedures, Senate Committee on the Judiciary, Dirksen Office Building, Washington, D.C. DEAR SENATOR MCCLELLAN: I am writing to you in my capacity as Chairman of the Committee on Federal Legislation of the National District Attorneys Association. The Association, through its Board of Directors, has followed with keen and active interest the legislative development of the proposed codification of federal criminal laws as now embodied in S. 1, 94th Congress, "The Criminal Justice Reform Act of 1975". Our interest and study began with the Brown Commission draft in 1971. The Association, representing some 6,000 state and local prosecutors throughout the United States, was primarily and deeply concerned with provisions defining jurisdiction in early drafts of the legislation which in our judgment constituted unreasonable and excessive expansion of federal enforcement powers. Traditionally and constitutionally enforcement of criminal statutes has been the function and primary responsibility of state and local governments. Federal enforcement jurisdiction was limited to those areas where federal authority to act was clear, either from a constitutional or territorial basis. Nevertheless, the "Brown Commission" draft contained provisions for such jurisdictional encroachments as discretionary constraint in exercise of concurrent jurisdiction, "piggy-back” jurisdiction and prohibition of multiple prosecutions. In response to objections raised by state and local authorities these provisions were modified. The National District Attorneys Association was among those interested groups which submitted opinions and testimony in this regard. Ancillary jurisdiction has been limited substantially in S. 1 of 1975. We find the jurisdictional provisions generally to be reasonable, support Chapter 2, S. 1 in principle and oppose any further revisions which would incorporate the objectionable jurisdictional provisions of the "Brown Commission" Bill or otherwise expand federal jurisdiction. We point out here our objection to preemptive jurisdiction in the case of election law violations where both state and federal interests are involved [Sec. 205 (b) ]. Provision should be made for concurrent jurisdiction with preemptive federal jurisdiction only on order of the Attorney General of the United States along the lines of Sec. 205 (c). In addition to definitions of jurisdiction state and local prosecutors were also concerned with substantive law definitions which, because of federal impact, might trigger and eventually bring about undesirable amendments to state penal statutes. Here again we note with satisfaction that modifications were made to tighten up and render more precise certain provisions which were cause for concern. This is true, for example, with Chapter 3, Sections 301-303 of S. 1 relating to culpable states of mind. The defense of insanity was another area of some concern. We note with particular approval the present language of Section 522 of S. 1 dealing with the insanity defense. It is hoped that this new approach will have beneficial impact on insanity defense statutes in all the states. Abolition of insanity as a separate defense, except to the extent that a defendant's mental disease or defect precludes a finding of culpable state of mind is an important conceptual breakthrough in this always troublesome area of the law. A Sub-Committee memorandum accurately observed that the approach of Section 522 is to place under criminal sanction persons who would have been found not guilty by reason of insanity under the previously endorsed American Law Institute test. In an era when the concept of civil commitment is crumbling in the face of constitutional attacks and its efficacy is seriously questioned, the Section 522 approach insures that a sentencing judge has the option of imposing jail sentences in appropriate cases. The National District Attorneys Association strongly supports this measure and recommends against any action to restore the "right or wrong" test of the ALI which was adopted in the Brown Commission draft. The majority of prosecutors note with approval the adoption in S. 1 of authorization for imposition of the death sentence in the Class A felonies of murder, treason, sabotage and espionage. A clear statement of support for these provisions by the Congress of the United States will profoundly influence state legislatures and the courts in restoration of the death penalty as an appropriate constitutional punishment and a deterrent to violent crime. We support this provision as a national statement of public policy on the death sentence. In summary, the National District Attorneys Association, with comments noted, supports in principle S. 1, 94th Congress and recommends its enactment. This draft, a product of years of legislative effort and study, in our judgment represents, on the whole, a sound basis for codification of federal penal statutes. Its jurisdictional provisions now appear to define reasonable and workable standards under which both federal and local prosecutors can carry out their respective functions effectively and without significant conflict. The enactment of this legislation will have great beneficial impact on those states which penal law require review, revision and codification and on law enforcement generally. In making this recommendation for adoption of the measure the National District Attorneys Association congratulates the Chairman, members of the SubCommittee and staff who contributed to this great legislative undertaking. The Association also wishes to express its appreciation for the generous readiness on the part of all concerned to hear and fully and fairly consider opinions and recommendations from all responsible sources. Sincerely yours, CARL A. VERGARI, District Attorney of Westchester County, The American Economic Review, Vol. 65, No. 3, June 1975 The Deterrent Effect of Capital Punishment: A Question of Life and Death By ISAAC EHRLICH* Debate over the justness and efficacy of capital punishment may be almost as old as the death penalty itself. Not surprisingly, and as is generally recognized by contemporary writers on this topic, the philosophical and moral arguments for and against the death penalty have remained remarkably unchanged over time (see Thorsten Sellin (1959, p. 17), and (H. A. Bedau, pp. 120-214). Due in part to its essentially objective nature, one outstanding issue has, however, become the subject of increased attention in recent years and has played a central role in shaping the case against the death penalty. That issue is the deterrent effect of capital punishment, a reexamination of which, in both theory and practice, is the object of this paper. The multifaceted opposition to capital punishment relies partly upon ethical and aesthetic considerations. It arises also from recognition of the risks of errors of justice inherent in a legal system, errors occasionally aggravated by political, cultural, and personal corruption under certain social regimes. Such errors, of course, are irreversible upon application of this * University of Chicago and National Bureau of Economic Research. I have benefitted from comments and suggestions from Gary Becker, Harold Demsetz, Lawrence Fisher, John Gould, Richard Posner, George Stigler, and Arnold Zellner. I am particularly indebted to Randall Mark for useful assistance and suggestions and to Walter Vandaele and Dan Galai for helpful computational assistance and suggestions. This paper is a reduced version of a more complete and detailed draft (see the author 1973b). Financial support for this study was provided by a grant to the NBER from the National Science Foundation, but the paper is rot an official NBER publication since it has not been reviewed by the board of directors. 397 form of punishment. But the question of deterrence is separable from subjective preferences among alternative penal modes and can be studied independently of any such preferences. Of course, the verification or estimation of the magnitude of the deterrent effect of the death penalty--the determination of the expected tradeoff between the execution of a murderer and the lives of potential victims it may help save -can, in turn, influence evaluation of its overall desirability as a social instrument even if that evaluation is largely subjective. Recent applications of economic theory have presented some analytical considerations and empirical evidence that support the notion that offenders respond to incentives and, in particular, that punishment and law enforcement deter the commission of specific crimes. Curiously, two of the most effective opponents of capital punishment, Beccaria in the 18th century and Sellin in recent years, have never to my knowledge questioned analytically the validity of the deterrent effect of punishment in general. Beccaria even recognizes explicitly the probable existence of such a general effect. What has been questioned by these scholars is the existence of a differential deterrent effect of the death penalty over and above its most common practical alternative, life imprisonment (see Beccaria, pp. 115–17). Sellin has presented extensive statistical data that he and others have interpreted to imply, by and large, the absence of such an effect (see Sellin (1959, 1967)). Whether, in fact, the death penalty constitutes a more severe punishment than 398 THE AMERICAN ECONOMIC REVIEW life imprisonment for the average potential offender cannot be settled on purely logical grounds, although crime control legislation, ancient and modern, clearly answers this question affirmatively. Observation that convicted offenders almost universally seek and welcome the commutation of a death sentence to life imprisonment is consistent with the intuitive ranking of the death penalty as the harshest of all punishments. Still, one may argue that the differential deterrent effect of capital punishment on the incentive to commit murder may be offset by the added incentive it may create for those who actually commit this crime to eliminate policemen and witnesses who can bring about their apprehension and subsequent conviction and execution. The existence of the differential deterrent effect of capital punishment is ultimately an empirical matter. It cannot, however, be studied effectively without thorough consideration of related theoretical issues. The crucial empirical question concerns the kind of statistical test to devise in order to accept or reject the relevant null hypothesis. Since the inquiry concerns a hypothetical deterrent effect, the null hypothesis should be constructed in a form that permits testing of the relevant set of behavioral relations implied by a general theory of deterrence. That includes the deterrent effects of law enforcement activities in general. Moreover, even if a negative effect of capital punishment on the rate of murder is established through systematic empirical research, there still remains the question of the existence of a pure deterrent effect distinct from a potential preventive or incapacitating effect associated with this form of punishment. An effect of the second type might be expected since execution eliminates categorically the possibility of recidivism. Contrary to previous observations, this JUNE 1975 investigation, although by no means definitive, does indicate the existence of a pure deterrent effect of capital punishment. In fact, the empirical analysis suggests that on the average the tradeoff between the execution of an offender and the lives of potential victims it might have saved was of the order of magnitude of 1 for 8 for the period 1933-67 in the United States. Two related arguments are offered in this context of which only the second will be elaborated in this paper. First, it may be argued that the statistical methods used by Sellin and others to infer the nonexistence of the deterrent effect of capital punishment do not provide an acceptable test of such an effect and consequently do not warrant such inferences. Second, it is argued that the application of the economic approach to criminality and the identification of relevant determinants of murder and their empirical counterparts permit a more systematic test of the existence of a differential deterrent effect of capital punishment. The theoretical approach, emphasizing the interaction between offense and defense—the supply of and the (negative) social demand for murder—is developed in Section I. Section II is devoted to the empirical implementation of the model. Some implications of the empirical evidence are then presented and discussed in Section III. I. An Economic Approach to Murder and Defense Against Murder A. Factors Influencing Acts of Murder and Other Crimes Against Persons The basic propositions underlying the approach to murder and other crimes against the person are 1) that these crimes are committed largely as a result of hate, jealousy, and other interpersonal conflicts involving pecuniary and nonpecuniary motives or as a by-product of crimes against property; and 2) that the propensity to perpetrate such crimes is in |