Introduction: 200 Years of Mistaken Identification
Siegfried L. Sporer
University of Aberdeen
Guenter Koehnken
University of Kiel
Roy S. Malpass
University of Texas, El Paso
Why do police officers, investigators, prosecutors, judges, and others with an interest in eliciting accurate memory-based testimony need to inform themselves of the research literature in experimental psychology that speaks to the question of witness memory? The answer is straightforward, from the perspective of a simple costâbenefit analysis. As with so many matters in the administration of public funds, effectiveness holds important rewards. Those who investigate crimes, who must decide which line of investigation to pursue and which line to postpone or set aside, necessarily make judgments about the likely guilt of suspects based on the information at hand. If they can make these judgments with a high degree of accuracy we all will be better off. For many cases, eyewitness identification is an important component of evidence, prosecution, and plea negotiation. If witness identification is correctly implemented investigators and prosecutors can make their judgments effectively, and focus their resources more efficiently. A major component of effectiveness requires avoiding expending scarce resources on erroneous prosecutions. These are counterproductive not only because of their injustice, but also because the person who should be the focus of those resources is not being pursued. It is in everyone's interest to make the best use of the memory of witnesses: to preserve it without changing it; to render it maximally accessible; to provide an environment in which witnesses feel free to report their recollections; to accurately assess the probable validity of the witness's report, regardless of the witness's certainty or doubts about its accuracy.
Eyewitness evidence is hardly needed in many cases. But in a substantial proportion of cases the evidence given by witnesses is essentially the only evidence linking a suspect to a crime. When this is true, the dangers of misallocation of resources is at its greatest, as is the need for information and assistance from research in the areas of scientific psychology concerned with memory and testimony. That mistaken identification and its attendant consequences (erroneous conviction, unpursued offender) have occurredâand continue to occurâare exemplified by many books, articles, and famous cases over a span of many years.
THE CASE OF THE SERGEANT LESURQUES
Probably the most famous case is that of the Sergeant Joseph Lesurques at the end of the 18th century (for detailed accounts and further references to this case, see Hirschberg, 1960; Sello, 1911). In 1796, a postal coach traveling from Paris in the direction to Lyon was robbed by five men. One of the robbers sat as a passenger inside the coach, and stabbed the two postal officers in the back with a dagger, supposedly when the four armed riders approached the coach. During the same day, before the robbery, the four robbers had been observed repeatedly riding on the same road, eating at a tavern and afterward having coffee at a coffeehouse nearby. One of them had even asked the tavern owner to sew one of his spurs to a cord attached to his boot. Later on, the same spur as well as one of the swords one of the robbers had temporarily forgotten in the stable of the tavern were found at the scene of the crime. Apparently, the robbers had been quite carefree about their appearances, and had been observed by a variety of people. Consequently, quite a few witnesses from the respective villages were summoned to give evidence. During the years 1796 until 1804, five successive trials took place in which 10 persons were accused of the crime, 7 of whom were found guilty and executed. At the first trial, 3 robbers were found guilty and executed and 1 was sentenced to 25 years as a galley slave for receiving stolen goods. At the second trial, another man was identified by a postal officer, sentenced and executed after a full confession.
In this context, the fate of Lesurques is quite remarkable. Lesurques, 35, was married, had three children, and had a substantial income after having been honorably discharged as a sergeant from the army. A friend of his, Guesnot, had been otherwise involved in this case, and after clearly demonstrating his innocence was supposed to pick up his papers at the courthouse. Guesnot met up with Lesurques and talked him into accompanying him to the courthouse. Outside the office of the justice of the peace, several witnesses were expectantly waiting to be questioned. Two of them, both servants at the tavern, claimed to recognize the two men as the robbers. Despite Guesnot's and Lesurques' pleas of innocence, the two maids held up their accusations with full certainty. Five additional witnesses also identified the two men as the robbers.
Guesnot was able to clear himself of the charges through a series of alibi witnesses. So could have Lesurques, had not a peculiar incident swayed the judge and the jury to believe the contrary. Considering the public attention the robbery had received, the criminal justice officials involved in its prosecution were eager to find the criminals. This seemed to have been particularly true of the presiding judge at the trial who was not willing to accept Lesurques' alibi. When Lesurques stated that the morning of the robbery he had been with the goldsmith Legrand and the owner of a jewelry store, Aldenhof, the judge summoned Aldenhof to fetch the business diary book where the transaction of a silver spoon between the two parties was supposedly documented. When the judge noted that the entry of the date in question had been tampered with, no one would believe Lesurques' statements, and his fate was inevitable. Lesurques was convicted and executed, and his family was expropriated. According to various historical sources (e.g., Hirschberg, 1960; Sello, 1911; see Sporer, 1984, for review), in the following years not only Lesurques but 13 suspects were âidentified,â and 7 men (including Lesurques) were tried and executed as the robbersâa remarkable feat, considering that there were only 5 robbers observed at the incident! Some 70 years later, Lesurques' family still argued before French courts of law to clear their name. Although the family had been able to regain some of the property the courts had confiscated, the appellate courts were never willing to fully admit their mistake and to unequivocally establish Lesurques' innocence.
MISTAKEN IDENTIFICATIONS AS A MAJOR SOURCE OF MISCARRIAGES OF JUSTICE
Although the case of Lesurques has been described as a warning example in many major textbooks on evidence, miscarriages of justice as a consequence of mistaken identifications have been rather prevalent in the history of criminal law. For example, even in the most recent case of John Demjanjuk, whom many prosecutors and other parties involved regarded for a long time as âIvan the Terrible,â and who was considered responsible for the death of 850,000 Jews at the Nazi camp Treblinka, the death sentence ruled by an Israeli district court was set aside by the Israeli Supreme Court. Although the Supreme Court was not willing to call Demjanjuk innocent, the justices were not convinced that Demjanjuk was identical to the camp guard Ivan at Treblinka, despite several witnesses claiming this identity (see Wagenaar, 1988). Several authors described other cases of mistaken identity as deterring examples in their accounts of eyewitness memory (e.g., Loftus, 1979; Shepherd, Ellis, & Davies, 1982; Sporer, 1984; Yarmey, 1979). Lest one thinks that these examples are unhappy exceptions, a recent systematic investigation of various sources of miscarriages of justice has singled out mistaken identification as the number one source of error (Rattner, 1988). Table 1.1 lists these various sources of error documented in Rattner's analysis.
Most of the 205 cases surveyed had severe consequences, due to the serious nature of the crimes charged. Ten percent lead to death sentences. Strikingly, mistaken identification was observed to be the major source of error contributing to the wrongful convictions, being involved in 52% of the cases surveyed. Although mistaken identifications were frequently not the only source of error in these cases, their importance for criminal investigation is clearly underscored by these data.
TABLE 1.1
Distribution of Types of Error Contributing to Wrongful Conviction Type of Error | f (N = 205) | Percent (Relative) | Percent (Adjusted) |
Eyewitness misidentification | 100 | 48.8 | 52.3 |
Perjury by witness | 21 | 10.2 | 11.0 |
Negligence by criminal justice officials | 19 | 9.3 | 9.9 |
Pure error | 16 | 7.8 | 8.4 |
Coerced confession | 16 | 7.8 | 8.4 |
âFrame-upâ | 8 | 3.9 | 4.2 |
Perjury by criminal justice officials | 5 | 2.4 | 2.6 |
Identification by police due to prior criminal record | 3 | 1.5 | 1.6 |
Forensic science errors | 3 | 1.5 | 1.6 |
Other errors (missing data) | 14 | 6.8 | â |
Note. From Rattner (1988, p. 291).
GOALS OF THIS VOLUME
The goal of this volume is to gather evidence from various research domains on eyewitness testimony. For this purpose, we have invited leading experts in this field to summarize the research available in their respective specialties. Although many of the studies discussed in this volume deal with eyewitness identification, it is noteworthy that many of them also touch upon other areas of concern to eyewitness researchers as well. For example, Hammersley and Read (chapter 6) discuss voice recognition by humans and computers, including particularly detailed instructions on conducting voice âline-ups.â Davies (chapter 10) writes of differential aspects of recognition memory in children. Yarmey (chapter 11) summarizes what we know of the eyewitness memory of the elderly, and Chance and Goldstein (chapter 7) analyze the state of our knowledge on the special problem of cross-racial identification.
There are also fairly unique chapters on psychological aspects of facial image reconstruction techniques (Shepherd & Ellis, chapter 5), on person descriptions (Sporer, cha...