that during the debate in the British House of Commons in 1956, Sir Patrick Spens, a former Chief Justice of India, said, "I am absolutely convinced I know-that fear of violent death is a deterrent, and no statistics, no argument whatever will convince me that it is not." Well, people like that you cannot debate with. Their minds are closed. Then there is the other recurring argument that it is easy to number the failures, but we cannot number the successes. No one can ever know how many people have refrained from murder because of the fear of being hanged. I don't quite understand that argument. To me it is a red herring. If no one knows and no one can know, it is a subject that should not be debated, because no one can debate it. But on the other hand, the way the statement is phrased, it assumes the existence of deterrence which is the very thing that needs to be proved. And therefore I have no sympathy with that kind of argumentation. Now, I think it is possible, by the use of the statistics that we have, to test whether or not the death penalty influences murder rate. I would be the first to admit that it is difficult to know how many capital murders there are among the murders that are known or are committed, because the very definition of murder in the first degree, which is punishable by death, requires the presence of certain items of knowledge which one cannot ascertain without catching the offender, in most instances. But we have to go on the assumption that the ratio of capital murder to general murder remains fairly constant from year to year. If we study the rate of willful homicide, we are also likely to be showing what has happened to capital murder over a period of time. Now, if a State has the death penalty for murder, and then abolishes it, those who are freed from fear of execution should presumably, some of them, be added to those who might otherwise also commit murder, and the murder rate or the homicide rate should rise. Well, all studies that have been made of comparable States, and I want to stress comparability here, because I don't think it is quite right to compare the abolition States with the nonabolition States without reference to the type of States involved. Our Southern States all have higher homicide rates, the highest homicide rates in the Nation will be found in some of the Southern States, and they all have the death penalty, while those States that have abolished the death penalty, most of them have relatively low homicide rates. So it would be natural that the abolition States would therefore have lower rates than the death penalty States, if they were all lumped together into two categories. But if we compare the abolition States with the adjoining death penalty States, and therefore comparable States within the same cultural region and reasonably similar in population and so on, then we can find no difference between the homicide rate of the abolition States and the death penalty States, but they are practically identical, and the trends over a period of years are identical. I have made diagrams of the homocide rates of groups of States compared in this manner, and I would challenge anyone who did not know what particular graph or line belonged to the abolition State and the ones that belonged to the death penalty States to pick out the aboli tion State. It cannot be done. Therefore, there is no unique power that the death penalty has to affect the homocide rates. The police have been mentioned. Police authorities are generally opposed to abolition. They believe in the death penalty except in the States that have abolished the death penalty. There the majority of the police chiefs do not believe that the existence of the death penalty protects the police. The vast majority of the police chiefs, not all, in the death penalty States defend the death penalty, because they believe it affords a protection. Now, a few years ago I acquired from the FBI Xeroxed copies of all of the offense reports on policemen killed during the period of 1961, 1962, and 1963, and we made a very careful analysis of those police homicides. When we compared, on the same basis that I have mentioned earlier, the abolition States with adjoining death penalty States, we found that the rate of policemen killed during those 3 years per 10,000 police officers was the same in the abolition States and death penalty States. There is no evidence whatsoever to suggest that the police are correct in their assumption that the death penalty gives them special protection. There is the problem of prison homocides. As you know, there are two or three States in the Union that have abolished the death penalty for ordinary murder, but have retained it for the killing of prison personnel in the belief that the existence of the death penalty will afford greater protection for prison personnel, inmates or staff. Well, we have never had very much information about that except general statements made by wardens about the character of prisoners who have been committed to prison for murder. But last year I made a study, in which all of the State penitentiaries and Federal institutions cooperated, in a study of the killings that occurred in these institutions, in all but five States, and they were very small, except for Mississippi, but Maine, for instance, and Idaho would not be likely to have many events of that type. Now, that information is rather interesting from many points of view. There were 61 homicide victims during the year 1965, eight staff members and 53 inmates in assaults committed by prisoners. Nine inmates were killed by unknown persons, all in jurisdictions having the death penalty. Fifty-two persons were killed by 59 known prisoners, and the 8 staff members were all killed in death penalty States. Who were these killers who were known? Fifty-nine prison killers were known. Sixteen of them were serving a sentence for murder of the first or the second degree, four of them in abolition States. Nineteen were serving terms for robbery. Altogether 43 were in prison for an aggressive crime against persons, 16 for property offenses. If we assume that the cases of inmates killed by unknown offenders involved nine perpetrators, then, of the offenders, were prisoners in death penalty States, and 11 in abolition States. Therefore it is true that sometimes a killing by someone who has been committed for murder occurs in prison, and there is no way of preventing it, because first of all the State courts or the courts have decided that these persons should not be sentenced to death, but should be sentenced to life in prison. Executive authorities may have com muted sentences, on the assumption that life imprisonment would be a more proper punishment than execution. And when you consider the unnatural character of prison institutions, unisexual institutions, throwing together sometimes thousands of men, usually men in close confinement, within walls, exposing them to all sorts of situations in which personal aggression might be invited or might result, it is perhaps amazing that there are no more such events, no more such assaults committed in prison. After all, such assaults do occur in freedom by people who are not confined in that manner, because of circumstances in which they find themselves. Now, what about the question of the length of sentence, the length of time spent in institutions. Here again statistics may take a little beating if they are not properly dealt with. How long is a life sentence? Well, unless you segregate at least those who die in prison, having been sentenced to life, from those who are able to be discharged from prison, you are going to lower the rate. The study that we made some years ago on the basis of five or six States showed that about 20 percent of the prisoners sentenced for murder to life die in the institution, and they had only served on the average about 5 years. At the present time those who are paroled from a life sentence in Pennsylvania serve about 20 years. In California, so far as I remember, if my memory does not deceive me, it is more around 10 or 11 years. But I don't have sufficient information to go beyond these two items of data, unless I were to do some hunting and maybe submit it to the committee later if the committee so wishes. Senator HART. If such data is developed, it will be added to the record. Mr. SELLIN. Now, with regard to the protection of the public, I think there are a few items that I have included in this paper because I was confining myself to a certain period of time, but I think it is very telling, take particularly New York. I have cited data from California. In New York, from July 1930 through 1961, 63 prisoners convicted of first degree murder were paroled in New York. Sixty-one of them had originally been sentenced to death, and their sentences commuted. Their average age at parole was 51 years. Fifty-six had no previous felony convictions. This I imagine also suggests to you that the parole board had used some pretty fair selective process. Three of the 63 became delinquent. Two of them were returned to prison for technical violations, and one after conviction for burglary. And you notice what I have cited about the situation in Ohio. From 1945 through 1965, there are even more recent figures, 273 first degree murderers were paroled in Ohio. Fifteen became parole violators. But only two of them were returned to prison after a conviction of a new crime, one for robbery and one for an attempt to rob. In other words, while a second homicide sometimes occurs, I know a few instances, they are very rare, and if you take the information simply on the basis of the statement in a statistical table, you may be fooled. The most recent case of that kind that I know of was a man who was convicted of murder in the first degree, sentenced to death, the sentence commuted. He served a long time in prison, was paroled, and after a time, I forget whether it was 6 months or more, was again charged with murder in the first degree, and again sentenced. I forget whether it was to death or imprisonment. This man had never killed anybody. He had never even had a gun or a weapon. This was a felony murder situation, where he was with two or three others holding up a gas station or something like that, and in the process someone was killed by one of the other members of the gang, but all of them under the law would have to be sentenced to death. So, as I say, statistics may sometimes deceive unless one knows the facts. Now, if I may, I am not going to bother to read anything except the last paragraph of my paper, if you will permit me. In its recent report "The Challenge of Crime in a Free Society," the National Crime Commission appointed by President Johnson 2 years ago was unanimous in its opinion that if a State imposed death penalties but does not execute them, it should remove capital punishment from its statutes. There are encouraging signs that this will progressively happen, because the death penalty has become impractical in the light of presentday experience. Look at the 400-odd persons on death row, the time being lengthened more and more between commitment after sentence and the possibility of execution or at least final decision. We are throwing the burden which the legislature should assume on the appellate courts and executive, and compelling them to face the issue, and that they are declining now is obvious. They are deferring, permitting appeals, upholding appeals, demanding new trials, et cetera, resulting in the fact that no one gets executed, and if anyone is executed after so many years of waiting, I consider it immoral. I regard the Chessman execution as one of the most immoral acts that any State could perpetrate, after 11 or 12 years of waiting. So it is impractical, in the light of present day experience, it is outmoded in the light of modern penology, it is useless as a deterrent, so haphazard in its execution as to be grossly inequitable, and so much like a primitive rite that it should have no place in a modern penal code. I hope that the Federal Congress will remove the death penalty from its statutes, and thereby approve what is perfectly obvious, the modern trend even in the United States, and thereby also spur the movement of abolition in the States of the Union. Thank you very much, Senator Hart, for being patient. Senator HART. Thank you, Professor. You, as well as you have, inasmuch as you are the concluding witness for this set of hearings, called attention to something that Mr. Bennett emphasized, but because it is so lacking in emotion is apt to get overlooked. As you approach your conclusion you reminded us again that the whole process of administering justice is bedeviled by the existence of the death penalty, the burden shifting to the executive, endless appellate procedures that are involved, 400-odd on death row and two executed. In the study of individuals, there is the excitement that attaches and we get swept away by the description of executions. But also, and I think a very persuasive reason for us to eliminate capital punishment is the fact that it snarls up the administration of justice, and to the degree that that can be improved, we have contributed to the advance in an area where our security is affected just as greatly really in the long haul as whether we execute two more next year under State orders. Thank you. Mr. Paisley. Mr. PAISLEY. Professor, the fact that States like Michigan and other States that have abolished capital punishment, that is, capital offenses, murders have not increased is about as convincing evidence as you can get that it really does not deter; capital punishment does not deter crime. Perhaps I didn't make myself clear. The fact that Michigan has abolished it for years, and there are no more murders in Michigan than there are in other States, that is about the best evidence you can get, isn't it, that capital punishment is not desirable, is not necessary. It has been brought out but I want to emphasize it. Mr. SELLIN. The nature of the offense, I am now speaking of murder, the nature of that crime is such that the problem of deterrence is likely to be on its face practically nil. Most victims are wives, husbands, paramours, drinking acquaintances, and there are only a relatively small number of victims that occur in the so-called felony murder type. The last report of the Uniform Crime Reports, for instance, indicates that of the homicides they knew of for 1965, only about 15 percent were actually known to be in the felony murder category, in other words associated usually with robbery. Otherwise the victims were all in the categories that I mentioned, and the murders occur under an emotional strain or a situation in which the thought of consequences does not appear. The one who thinks that he is so skillful that he is going to get away with it, no punishment affects him, because nobody is ever going to discover him. That is what he thinks, and there is no deterrent effect there possible. As Professor Radzinowicz said, this whole problem of deterrence is a very difficult one to investigate, especially in connection with certain other offenses, but I think that when it comes to the taking of a human life, which runs up against the strongest barriers of norms and mores that we have in our society, we can imagine that the circumstances are very special for this type of crime, and that therefore deterrence cannot operate as it might be able to operate in certain other types of offenses. Mr. PAISLEY. Thank you very much. Senator HART. As we close this set of hearings, I would suggest and I would ask counsel to review the great number of letters that other members of the committee and I have received, and those which he feels add, either because of the background of the writer or the subject matter contained, to the testimony that we have had, both for and |