THE ANALYSIS OF TRACE EVIDENCE is One of the most useful as well as one of the most disputed corners of forensic investigation: useful because it deals with what the criminal may have left behind or taken away with him from the crime scene; disputed because, though couched in the trappings of science, much of it is based on anecdote or “common sense” and has never been rigorously tested. As Albert Einstein put it, “Common sense is the collection of prejudices acquired by age eighteen.” Sometimes these anecdotes are valid generalizations and sometimes the common sense turns out to be valid science, but there is no way to know which is which without performing validating tests.
When any forensic procedure becomes the old, established way of doing things, and evidence derived from it is admitted in criminal cases without question, error may creep in. And when the procedure in question is based on unsubstantiated science, the error may be severe and damaging. Not all, or even most, trace evidence is invalid, but the possibility is something to be alert to.
Trace evidence consists of those small, possibly microscopic, bits of detritus that the criminal leaves behind at the crime scene or takes with him. It can be collected and analyzed to establish a connection between the criminal and the crime. Among the possibilities are blood (and other body fluids); explosive residue; animal, vegetable, or artificial fibers; glass shards; hairs; paint scrapings; shoe or foot impressions; tire impressions; bite marks (discussed in Chapter 15); markings on bullets or cartridges; lip prints; ear prints; glove prints; bits of food or other debris; and scrapings of anything from which material may have been transferred.
Dr. Edmond Locard (1877–1966) was the author of the important seven-volume Traité de Criminalistique (1923) and the first man to use the term “criminalistics.” A French criminalist, Locard held degrees in both medicine and law, and was a student of Alexandre Lacassagne, the professor of forensic medicine at the University of Lyon. Locard became Lacassagne’s assistant but left in 1910 to establish his own police laboratory. His enterprise eventually became the official laboratory of the technical police for the Prefecture of the Rhône. His skills were so highly regarded that he was in constant demand by police agencies throughout Europe.
In 1910 Locard stated what is now known as Locard’s Exchange Principle: “Any human action, and certainly the violent action of committing a crime, cannot occur without leaving a trace.”
In 1912 Locard had a chance to demonstrate his theory when he was called in to aid the Lyons police judiciaire in the investigation of the murder of a young woman named Marie Latelle. The girl had been strangled in the parlor of the house she shared with her parents, and there were indistinct marks on her neck, which the police suspected might be fingerprints. From the condition of the body, the police inferred that Marie had been killed early the night before. She had a boyfriend, Emile Gourbin, a bank clerk, who was the logical suspect. His relationship with the lovely Marie had been volatile at best; she liked to flirt, and he was madly jealous. He was arrested, but he had a strong alibi: at the time of the murder he was some kilometers away dining with friends. These friends not only supported his alibi, they testified that he had stayed downstairs playing cards until about one in the morning and had then retired to an upstairs bedroom to spend the night.
The police judiciaire called Locard to see if he could help—perhaps he knew of some way to develop the fingerprints on the victim’s neck. Locard examined the body. He peered closely at the neck through a magnifying glass. He was sorry, he told the police, but the marks on the neck were not fingerprints. They were scrape marks, probably from fingernails.
Locard went to the jail and took scrapings from under the suspect’s fingernails. Back in his lab he peered at the material through his microscope. There were skin cells that might have been the victim’s, but they might also have been Gourbin’s. In addition there were tiny grains of something else, something pink. Taking a close look at this something else, Locard found it to be made up of rice powder containing small amounts of zinc oxide, magnesium stearate, and a form of iron oxide known as Venetian red. Slightly darker than scarlet, Venetian red pigment was used in expensive face powders along with rice powder, zinc oxide, and magnesium stearate.
The police questioned all local compounders of face powders—there were no mass-market cosmetics back then—and located one who produced a product identical to the one found under Gourbin’s fingernails. They had, indeed, sold their powder to Marie Latelle.
When confronted with this evidence, Gourbin confessed. To provide himself with an alibi, he had set the clock at his friend’s house an hour and a half ahead. Then he had sneaked out to meet Marie. The assignation had been for quite another purpose, he assured them, but when he and Marie had a fight, he strangled her. Gourbin’s confession was one of the earliest successes of Locard’s use of microscopy and trace evidence.
Locard also contributed to the developing study of dactyloscopy (fingerprint analysis), and invented poroscopy, the study of pore patterns on the papillary ridges of the fingerprints as an aid in identification. In 1914 Locard published the results of a statistical analysis of fingerprint patterns and established rules for comparing prints that still hold true today. In establishing points that match between a suspect print and the exemplar:
—If more than twelve points concur and the fingerprint is sharp, the two prints match.
—If eight to twelve points concur, the case is borderline and the certainty of the match will depend upon the sharpness of the fingerprints; the rarity of its type; the presence of the core of the print and a delta in the comparable part of the print; the visibility of pores; the clarity and obvious agreement of the width of the papillary ridges and valleys, the direction of the lines, and the angles of the bifurcations.
—If fewer than eight points concur, the fingerprints cannot be used for positive identification but only to say that the suspect cannot be excluded.
Locard was one of the founders of the International Academy of Criminalistics in 1929, a model for today’s forensic science organizations.
On the morning of Good Friday, April 10, 1936, thirty-four-year-old Nancy Evans Titterton was raped and murdered in her apartment at 22 Beekman Place in Manhattan. She was found by Theodore Kruger, owner of an upholstery shop, and his assistant, John Fiorenza, who were returning a couch to the apartment. The front door to the apartment was ajar, so they entered and called for Mrs. Titterton. They found her body in the bathroom, lying prone in the empty bathtub.
This was a high-profile crime. Nancy Titterton was a noted author and book reviewer, and the wife of the literary critic Lewis Titterton, the head of the literary rights department at NBC. Assistant Chief Inspector John Lyons was assigned to direct the investigation and given a force of sixty-five detectives in the hope that he could clear the case quickly.
Titterton’s body was naked except for a pair of rolled-down stockings. Her tied-together pajamas had been used to strangle her. She had been raped. A knife that had been wiped clean of fingerprints lay on the bathroom floor. There were signs of a struggle in the bedroom. Ligature marks on the victim’s wrists indicated that she had been tied up before she was raped, but the rope had apparently been cut off and taken away. But then Lyons had a piece of luck: when the medical examiner lifted the body, he uncovered a thirteen-inch-long piece of rope, cleanly cut at both ends. It was similar to a Venetian blind cord, but it had not been removed from any of the blinds in the apartment.
There were several apparent leads. A strange man, described as young, pleasant, and well-spoken, had knocked at a downstairs door twice, the last time at three o’clock in the morning, to inquire after two different women, neither of whom lived in the building. But the man could not be located, and none of the other leads went anywhere. Lyons decided to see if science could help. As the New York Times put it: “[The police] made no concealment of their reliance on laboratory examinations of physical items of evidence to lead them to the solution. Fields of science almost ignored heretofore were entered yesterday, and some of the methods of scientific investigation suggested in the Hauptmann case have been resorted to now.” (Bruno Richard Hauptmann had recently been tried and convicted for the kidnapping and murder of the Lindbergh baby.)
Two items of possible forensic interest were closely examined. The first was the length of rope discovered under the victim. It was found to have a Tampico fiber base, a product made from the Mexican agave rigida plant and rare enough that it might be traceable. The second was a half-inch-long strand of stiff white hair that was found after careful examination of the bedspread with a magnifying glass. Dr. Alexander Gettler of the New York City Toxicological Laboratory, who found the hair, placed it under a microscope and decided it was horsehair. Furniture was often stuffed with horsehair. Lyons got a sample of the hair from the couch the workmen had delivered, and Gettler found the two to be identical. Lyons then had his men check all the other furniture in the apartment and found that none of it was stuffed with anything that matched the hair found on the bedspread. So the hair had come from the couch. Perhaps it had been carried in on the clothing of one of the furniture repair men.
Theodore Kruger, owner of the upholstery shop, told Lyons that Fiorenza, his assistant, had been with him when they had picked up the couch the day before the murder, and had not come to work until almost noon on the day of the murder. When Lyons pulled Fiorenza’s rap sheet he found that in his twenty-four years Fiorenza had been arrested four times for grand theft and had served a two-year jail sentence. There was also a 1934 psychiatric report that diagnosed Fiorenza as delusional and prone to wild fantasies.
The upholstery shop assistant was now certainly a prime suspect, but there was not enough evidence to take to a jury. The needed evidence arrived April 17 when the thirteen-inch Tampico fiber cord found beneath the victim was traced to the Hanover Cordage Company of York, Pennsylvania. Their wholesalers had made many sales in the New York area, and one of them had been to Kruger’s upholstery shop.
The white horsehair might have found its way into the bedroom innocently, though the two men had not entered the bedroom when they attempted to deliver the couch. But the rope found under the body tied Fiorenza to the crime.
Fiorenza was brought in for intensive questioning. And when he heard that the rope had been traced to the shop, he confessed. He claimed to have returned to the apartment that morning convinced that Nancy Titterton had fallen for him during their brief encounter the day before. When she rebuffed him, he became so furious that he tied her up and raped her. There was a certain superficial plausibility to this, as Fiorenza had been diagnosed as delusional. But if so, why had he brought the knife and the cord?
Afterward he had strangled her and left her in the bathtub. He claimed that she was still breathing when he left her. But if he intended to leave her alive, why did he strangle her?
His story engendered neither sympathy nor belief from the jury that convicted him. He was executed on January 22, 1937.
“I hate this ‘crime doesn’t pay’ stuff. Crime in the U.S. is perhaps one of the biggest businesses in the world today.”
—Paul Leland Kirk
In 1929 Paul Leland Kirk (1902–1970) was appointed professor of chemistry at the University of California, Berkeley. Except for the World War II years, he spent his entire career at Berkeley, teaching and developing the techniques of crime-solving that became known as “criminalistics,” a term that Kirk himself popularized. He wanted to establish the field not only as a profession but as a recognized scientific discipline. The clearest explication of his beliefs is a statement he made in 1974:
Wherever he steps, whatever he touches, whatever he leaves, even unconsciously, will serve as a silent witness against him. Not only his fingerprints or his footprints, but his hair, the fibers from his clothes, the glass he breaks, the tool mark he leaves, the paint he scratches, the blood or semen he deposits or collects. All of these and more, bear mute witness against him. This is evidence that does not forget. It is not confused by the excitement of the moment. It is not absent because human witnesses are. It is factual evidence. Physical evidence cannot be wrong, it cannot perjure itself, it cannot be wholly absent. Only human failure to find it, study and understand it, can diminish its value.
Kirk’s training as a microchemist brought him to the field of forensic investigation, but he was best known as a blood-spatter expert and for his analysis of the crime scene in the Sam Sheppard case (see Chapter 11).
The use of blood-spatter evidence was brought to the attention of forensic analysts and the general public with the Sheppard case, but the idea had been around for quite a while. In the 1890s Dr. Eduard Piotrowski of the Institute for Forensic Medicine in Poland experimented with using hammers and hatchets on the heads of rabbits. He published his findings in the 1895 pamphlet “Concerning the Origin, Shape, Direction and Distribution of the Bloodstains Following Head Wounds Caused by Blows.” Piotrowski believed that “It is of the highest importance to the field of forensic medicine to give the fullest attention to bloodstains found at the scene of a crime because they can throw light on a murder and provide an explanation for the essential moments of the incident.”
Biological evidence—blood, saliva, and semen samples—is collected as soon as possible at the crime scene. This is not always as simple as it sounds. Dried blood can be red, brown, black, or even green or yellow. And not all red stains are blood. Semen and especially saliva stains are easy to miss. Careful and close examination of the area with an alternate light source, one that uses a part of the spectrum at which the biological samples are more visible, is essential.
Shoe marks or tire impressions may be either flat (a footprint on a dusty floor) or plastic (a tire track in the mud). They are first photographed, rendering what is called a “latent print.” In the case of a flat print, it is then lifted with tape like a fingerprint. A suspect’s shoe can be compared with the print using side-by-side examination, or with a transparent overlay. A plastic impression will be preserved by making a plaster cast. The same analysts often handle both foot marks and tire marks, and whatever else that comes along for which there is not a specific expert. The examiner will look first for class characteristics: tires with the same tread pattern, shoes with the same distinctive heel and sole markings. Then he or she will look for the “accidentals”—wear patterns, cuts or abrasions that show up in the impression, or substances adhering to the tire or shoe. The class markings will lead the investigator to the make and model of a tire—there are directories that contain this information—but the accidentals will tie the impression to a specific shoe or tire.
Sometimes class characteristics can be very suggestive. There were footprints in the blood at the scene of the Nicole Brown Simpson-Ronald Goldman double murder in Los Angeles. FBI examiner William Bodziak was able to show that they were made by a pair of size twelve Bruno Magli shoes, made exclusively by an Italian company with very limited production. O. J. Simpson, Nicole’s ex-husband, was tried for the murders and acquitted. Although he denied ever owning Bruno Magli shoes, a photograph of him wearing a pair was found. The shoes themselves had disappeared. Finding the shoes and showing that the accidentals were a match would have been a powerful piece of evidence against him. On the other hand, if the accidentals had not matched it would have been easier to believe in his innocence.
In order to make use of footprint or tire track evidence, the examiner needs to be able to show exactly what shoe or tire made the mark. Then detectives need to be able to link a specific person with that mark.
SWGTread, a professional organization for footwear and tire examiners, suggests that their members avoid common terminology such as “consistent with” or “responsible for,” and use the following standardized terms:
—“identification” (definite conclusion of identity)
—“probably made” (very high degree of association)
—“could have made” (significant association of multiple class characteristics)
—“inconclusive” (limited association of some characteristics)
—“probably did not make” (very high degree of nonassociation)
—“elimination” (definite exclusion)
—“unsuitable” (lacks sufficient detail for a meaningful comparison)
Louise M. Robbins (1928–1987) was a professor of anthropology at the University of North Carolina at Greensboro and the author of Footprints: Collection, Analysis and Interpretation. She could discern what others could not and would gladly write it up in a scientific journal or swear to it in a court of law. In 1978, when anthropologist Mary Leakey and her team discovered 3.6-million-year-old hominid footprints in an ancient stream bed at Laetoli in Tanzania, Robbins affirmed that they were human, and further that they were the prints of a man and a woman walking together, and that the woman was pregnant.
In criminal cases where a shoe print was left at the crime scene but there was no shoe to match it to, the prosecution would have Robbins perform a “wear pattern analysis.” As she testified time and again, no two people have identically shaped feet or identical gaits. So if she inspected the inside of the defendant’s shoes, she could match foot to footprint, even if the print was made by a different shoe. Defense attorneys and many of her fellow scientists took to calling her method the “Cinderella analysis.”
In an article published by the Center for Public Integrity, Steve Weinberg wrote:
Prosecutors around the nation used to retain the services of a University of North Carolinz–Greensboro anthropology professor named Louise Robbins, who said she could match crime-scene footprints to the footwear of perpetrators. Few other forensic scientists endorsed the validity of Robbins’ techniques. But prosecutors called on Robbins over and over, banking on the good will of the trial judge to certify her as an expert. Robbins helped convict defendants across the nation until her technique was shown to yield results that were no better than chance would have produced.
Robbins began her Cinderella testifying in 1976, and by 1986 she had been a prosecution witness in ten states and Canada. In 1987 the American Academy of Forensic Sciences had Dr. Robbins’s cases and conclusions reviewed by a panel of 135 anthropologists, forensic scientists, and lawyers. They concluded that her work had no scientific merit. One member of the panel, the law professor Melvin Lewis, called it “complete hogwash.”
Why would judges, who have the responsibility of keeping junk science out of the courtroom, accept the testimony of this glib charlatan? One reason is that she testified for the prosecution. Most criminal court judges, having come up through the prosecutorial ranks, look less critically than they should at prosecutorial evidence.
Just as a specific gun leaves its imprint on a bullet passing through the barrel, so a specific tool—a chisel, screwdriver, or hammer used to pry or chip—will leave its imprint on the deformed surface. And each tool has unique markings that can be compared to markings left behind.
The forensic analyst who specializes in this area of study is known as a firearms and toolmark examiner. These analysts have their own professional organization, the Association of Firearm and Toolmark Examiners (AFTE). Their code of ethics is very strict and covers the many possibilities that might arise in an investigation or a trial, but essentially the examiner is required to be impartial. He is not to take sides, slant testimony, or give more weight to his testimony than the evidence allows.
The three possible conclusions of a toolmark examination are identification—this tool made this mark; exclusion—this tool did not make this mark; and no conclusion—there isn’t enough information to know whether or not this tool made this mark.
As in other fields of forensic analysis, toolmark analysis necessarily suffers from unconscious examiner bias. In a recent case a highly regarded ballistics expert who was trying to match bullet fragments to a particular gun was heard to say on turning in his report, “I really had to work hard to make this match, but I finally made it.” He knew what he was supposed to find, and he worked hard at finding it. What would he have found if he had not known what he was expected to find?
It is surprisingly common to find an ear print at the scene of a crime. Someone who is hiding and pressed against a wall, or a person who is listening at a window, may well leave the print of an ear behind. And there are forensic scientists ready to swear that they can tell one ear from another by its print mark. The scientific community does not agree, but ear-print evidence still makes its way into the courtroom, sometimes with ill results.
On Tuesday, May 7, 1996, Dorothy Wood, a ninety-four-year-old retired health worker, was smothered to death in her house in Huddersfield, West Yorkshire, England. Police believed it was the work of a burglar who had climbed in through a transom window above her bed. The killer had apparently listened at the window before entering, because clear ear prints were found on the glass.
The police had one suspect, a twenty-four-year-old local resident named Mark Dallagher, who had prior arrests for burglary. Dallagher claimed he had spent the entire night with his girlfriend. She could not verify this, however, because she had taken a sleeping pill. The detectives took impressions of his ears and sent them, along with the ear prints from the window, to Cornelius Van der Lugt, a Dutch ear-print expert. Van der Lugt examined the ear prints and reported that they did not match.
Then Van der Lugt changed his mind and decided that they did match after all. Perhaps it was because the prosecutor had decided to charge Dallagher with the murder and needed Van der Lugt’s testimony. The ear evidence was effectively all they had, and no one need know about the first report. By the time the trial began in the Leeds Crown Court, the prosecution had managed to find a jailhouse snitch who would testify that Dallagher had confessed to him while they shared a cell.
The prosecution put two ear-print experts on the stand—Peter Vanezis, the Regis Professor of Forensic Medicine and Science at the University of Glasgow, who said that the ear print might be Dallagher’s but that he couldn’t swear to it, and Cornelius Van der Lugt, who was now 100 percent certain that the print on the window had been left by Dallagher’s left ear. On December 15, 1998, Mark Dallagher was convicted of murder and sentenced to life in prison.
This was a victory for prosecutors and a first in the United Kingdom. One of them said in a BBC interview, “In planning to use the ear-print evidence we sought the advice of experts in order to prove that it could not have belonged to anyone else.”
The verdict was set aside on July 25, 2002, by the Court of Appeals on the grounds that it was “unsafe.” Dallagher’s new lawyers had called on two experts of their own, Dr. Christopher Champod, an analyst with the forensic science service in Britain, and Professor P. J. Von Koppen of the University of Antwerp in Belgium. Both had made a study of ear prints and both, for slightly different reasons, had found that they were not a dependable means of identification. There would have to be another trial.
Ten days into Dallagher’s new trial Van der Lugt’s original report surfaced. In it he had stated that the ear print had definitely not come from Dallagher. The judge stopped the trial and ordered Dallagher released from prison on bail while he pondered what to do. A short while later a DNA analysis of the ear print itself showed that the donor could not have been Dallagher. Judge Sir Stephen Mitchell directed a verdict of not guilty and said to Dallagher, “This most unfortunate saga at long last comes to an end.”
During the same period several convictions in the United States that had depended on latent ear-print analysis, including two at which Van der Lugt had served as the expert witness, were reversed on findings that the technique was unreliable.
Hairs are one of the most common forms of trace evidence found at crime scenes. Like dogs, cats, rabbits, and other animals that might wander across a crime scene, humans are continually shedding hairs. The hair of different species can be reliably differentiated under a microscope, and the hair of two different people might be different enough for exclusion purposes. The practice of bleaching, dyeing, curling, flattening, and otherwise altering hair can further individualize a sample, as can a variety of diseases and parasites. But the error rate for the microscopic analysis of hair is too high to use it for reliably identifying anyone as the donor of a given sample. After DNA analysis became available, the FBI did a study to check their results and found that microscopic examination of hair by their own experts was wrong 11 percent of the time.
DNA can be extracted from the root of a hair, but hair that falls out naturally is usually rootless. So microscopic examination is often still necessary.
On April 19, 1989, a twenty-eight-year-old woman who was jogging in Manhattan’s Central Park was attacked, raped, savagely beaten, and left for dead. When she was found five hours later she had lost 75 percent of her blood, her skull was fractured, and one eye was out of its socket. The immediate suspects were a gang of fourteen- and fifteen-year-old boys who had been running wild through the park, knocking people off bicycles, assaulting people, and otherwise causing havoc for several hours around the time of the attack. They had already been rounded up by the police and were being held in the Central Park precinct house when the jogger was found.
The woman survived, though she was unconscious for twelve days and in and out of delirium for another five weeks. She remembered nothing of the attacks, and even years later the entire five-week period was erased from her memory.
Three hairs found on the clothing of one of the boys and one hair found on another were examined microscopically and found to be consistent with the victim’s hair. On the basis of this and videotaped confessions that were recanted almost immediately, five of the boys were placed on trial and found guilty.
Three months after the jogger was attacked, eighteen-year-old Matias Reyes was arrested for rape. As part of a plea bargain, he confessed to another rape as well as the rape-murder of a pregnant mother. Thirteen years later, in January 2002, after the statute of limitations had run out, he added the attack on the Central Park Jogger to his list of crimes. Recent advances in DNA testing made it possible to check his story, and semen recovered from the jogger’s body was tested. It matched Reyes’s. Further, DNA tests on the hairs recovered from the five boys did not match the victim’s. At the request of Manhattan District Attorney Robert Morgenthau, the convictions against the five boys were dismissed.