Edwin M. Borchard, the author of the first landmark study of wrongful conviction in America, was a professor at Yale Law School.

Perhaps the major source of these tragic error is an identification of the accused by the victim of a crime of violence. This mistake was practically alone responsible for twenty-nine of these convictions...

There are a few cases, including cases of alleged rape -- a type intentionally omitted from this collection -- where the issue of guilt turns mainly upon the veracity of the prosecuting witness and of the accused. But little scientific study has yet been made of the problem of lying witnesses or of defective powers of observation and of the effect of suggestion in criminal cases.

In thirteen of the cases no crime at all was committed, and in three other cases the commission of any crime is doubtful. In the balance of the cases, a crime appears to have been committed, but it was not committed by the accused.

Erroneous convictions on circumstantial evidence exclusively, that is, in the absence of identification and perjury, are not many, yet enough to be disturbing. Of the eleven cases of this type here recorded, eight involve charges of murder in the first degree and convictions of murder in the first or second degree...

Cases of circumstantial evidence into which entered a mistaken identification are fifteen in number. Cases of circumstantial evidence in which perjury was an ingredient are eleven in number. Cases in which the perjury of prosecuting or other witnesses, taking advantage of circumstantial evidence, natural or manufactures, was the main factor in the conviction are not inconsiderable -- fifteen. Among them are four for murder in which the alleged "murdered" person later turned up alive and well. In fourteen cases the victim was "framed" by hostile witnesses.

There is not much that the prosecuting or judicial machinery can do to prevent some of these particular miscarriages of justice. In many of the cases just mentioned, the prosecuting attorney was obliged to take the evidence as presented to him, including the uncontrollable perjury of vengeful witnesses, and lay it before the jury without realization of its worthlessness. Yet in only a few of the cases can it be said that no fault, carelessness, or overzealousness can be charged to the prosecution. In this group we may, by liberality, list sixteen cases. in a very considerable number, the zealousness of the police or private detectives, or the gross negligence of the police in overlooking or even suppressing evidence of innocence, or the prosecution's overzealousness was the operative factor in causing the erroneous conviction. Such lapses from the impartial enforcement of the law are hardly excusable. Yet, without making any claim to generalization, it is common knowledge that the prosecuting technique in the United States is to regard a conviction as a personal victory calculated to enhance the prestige of the prosecutor. Except in the few cases where evidence is consciously suppressed or manufactured, bad faith is not necessarily attributable to the police or prosecution; it is the environment in which they live, with an undiscriminating public clamor for them to stamp out crime and make short shrift of suspects, which often serves to induce them to pin a crime upon a person accused.

Especially is this the case where the accused has been previously convicted or has had unfortunate experiences with the criminal law. Evidence of a prior conviction is often fatal to an accused person, and the prejudice it is calculated to create in the minds of the jury is doubtless one of the main reasons why prosecutors insist upon such evidence. Baffled by crime waves, the result of conditions the public seems unwilling to recognize and face, there has been a growth in public impatience not only with the technicalities of the law, by which so many guilty criminals escape conviction, but also with the normal processes of the law, designed to safeguard the accused against injustice...

In several of the cases in this collection a species of third degree or undue influence produced "confessions" from the accused -- confessions which constituted a material factor in their convictions. Notable among these are the cases of Johnson in Wisconsin and Stielow in New York, both men of poor intelligence, influenced by promises of forthcoming advantage to themselves. The confessions in the Boorn and Butler and Yelder cases are explainable only on psychological grounds and have no special interest. The Lyons confession, coming after the conviction, is in the main amusing. While confessions may often seem conclusive, they must be carefully examined. Persons charged with crime are not infrequently of defective or inferior intelligence, and, even without the use of formal third-degree methods, the influence of a stronger mind upon a weaker often produces, by persuasion or suggestion, the desired result. State's Attorney Homer Cummings, in the Connecticut case of Israel, who under pressure confessed the murder of a priest, but whom Mr. Cummings refused to try, said:

It was the opinion of the physicians that any confession made by the accused was totally without value, and they were of the opinion also that if they carried through to the accused a continuous and fatiguing line of interrogation, accusation and suggestion, in due course he would be reduced to such a mental state that he would admit practically anything that his interrogators desired. They further stated that this was a common phenomenon with certain types of people, and that where such people were subjected to interrogatories, accusations or suggestions from persons of stronger will, the lesser mind will ultimately succumb and accept the conclusions of the more powerful intellect.

Public opinion is often as much to blame as the prosecutor or other circumstances for miscarriages of justice. Criminal trials take place under conditions with respect to which public interest and passions are easily aroused. In fourteen of the cases in this collection in which the frightful mistake committed might have been avoidable, public opinion was excited by the crime and moved by revenge to demand its sacrifice, a demand to which prosecutors and juries are not impervious. This can by no means be deemed an argument for the abolition of the jury, for judges alone might be equally suscepible to community opinion. But it is a fact not to be overlooked.

In eight of the murder cases recorded here, no crime was committed at all. Nobody was murdered. The convictions rested upon perjury or circumstantial evidence and were later shown to have been without foundation. In six of the cases, the person alleged to have been murdered turn up hale and hearty some time after the supposed murder had entered upon his sentence in the penitentiary. In several of the cases the convicted prisoner, later proved innocent, was saved from hanging or electrocution by a hairbreadth. Only by rare good fortune were some of the sentences of hanging and electrocution commuted to life imprisonment or indictments for first-degree murder modified by verdicts of second-degree murder, so that the error could still be corrected. How many wrongfully convicted persons have actually been executed, it is impossible to say. But that these cases offer a convincing argument for the abolition of the death penalty, certainly in cases of convictions on circumstantial evidence, can hardly be gainsaid.

The unreliability of so-called "expert" evidence is disclosed by eight striking cases. It seems clear that there is a necessity for publicly employed impartial experts who have no more interest in the case than the judge. Wisconsin has such a statute providing for experts to be appointed by the court, after notice and hearing.

The errors here recorded were uncovered in various ways:

1. In the case of peculiarly original crimes, by the fact that they continued to be perpetrated after the wrong person was in custody.

2. By the corroborated confessions of guilty third persons or of prosecuting witnesses.

3. By the substantiated confessions of one or more accomplices in a joint crime, by which all the guilty participants were named or identified and thus the wrongfully accused person proved innocent.

4. By sheer good luck, by which the police or the prosecutor of the governor discovered that the wrong man was caught and convicted. How many unfortunate victims of error have no such luck, it is impossible to say, but there are probably many.

After discovery of clear error, police, prosecutor, and court have in many cases shown commendable zeal in undoing the wrong...

The expense frequently incurred in proving a person's innocence, as is manifest in the Andrews, Kimball, Stielow, and other cases, and the fact that relatively few convicted persons can assume such a burden, makes it seem desirable that, for certain cases in which the suspicion of error is strong, and independent public investigation committee be established, which shall become part of the judicial machinery, and which may be invoked at the request of the trial judge or the governor or the board of pardons, as the case may be. The haphazard investigations now often instigated are hardly adequate or efficient. The expense to the state would not be great and the opportunity of correcting error considerable.