A Family Tree Gave Detectives Two Names When They Needed One
A Family Tree Gave Detectives Two Names When They Needed One.
April Tinsley was eight years old when she disappeared in Fort Wayne, Indiana, on Good Friday in 1988. Three days later her body was found beside a road in neighbouring DeKalb County. Her mother, Janet, had reported a child missing on the way home from a friend's house. She would wait thirty years to find out who had taken her daughter.
The investigation is often described as a case solved by a genealogy database. The arrest affidavit records a more revealing sequence. Genealogy identified a family and narrowed the inquiry to two living brothers. Investigators then had to distinguish between them. They did so by obtaining a separate sample from material discarded at a home, comparing it with existing evidence and questioning the man to whom those findings pointed. His guilty plea ultimately supplied a legal resolution the family tree could never have delivered on its own.
What Survived From 1988
April disappeared on 1 April. Her body was found around twenty miles from Fort Wayne. Examination established that she had been sexually assaulted and killed. Biological material recovered from her underwear could later be used to construct a DNA profile of a male contributor. Its value depended initially on something quite ordinary: the item had been recovered and preserved despite nobody knowing how a named suspect would eventually be found.
A profile has no address on it. If an investigator has two suitable samples, a laboratory can compare them. If there is a known individual whose sample matches a trace, that comparison can help establish the trace's origin. In April's case the trace was associated with a young child who had been abducted and attacked, making its source critically important. For years the police had no identified man against whom to make the decisive comparison.
That lack of an identification can be mistaken for a lack of progress in the science. In fact the two are quite different. A laboratory might be able to describe a trace with great precision and still have no route from its profile to a name. Returning to the same sample and measuring the same features more accurately does not necessarily get investigators any closer to a home address or somebody who can be interviewed.
A Claim Appeared On A Barn
In 1990 writing was found on a barn in the area. Its author claimed responsibility for April's death and referred to a missing shoe. The writing was relevant because of what the author professed to know and because the threat suggested another child might be at risk. Taking it seriously did not mean accepting the author's account as proved. People sometimes claim knowledge of notorious crimes they did not commit.
The police therefore had several questions. Did the writer know something that had not been publicly reported? Was the wording compatible with evidence investigators held? Could the writing be attributed to a person, or the location linked to somebody plausibly involved? Even an answer to the first question would not necessarily settle the others. A person can obtain information from someone else and then present it as personal knowledge.
The writer had put the police in the position of responding publicly to a threat without knowing whose assessment of danger they were hearing. Reproducing the threats in full adds little to understanding how officers worked. Their significance became clearer only when later communications supplied something more persuasive than an assertion of responsibility. They supplied material that could be compared in a laboratory.
Why The Messages In 2004 Were Different
In 2004 threatening messages and associated items were discovered in the Fort Wayne area. Some included used condoms. The affidavit describes DNA obtained from those items as consistent with the profile recovered from April's clothing. Sixteen years separated the killing and these messages. The evidence now gave detectives a way to link two periods that an anonymous writer might otherwise have connected only by making a claim.
There is a careful distinction here between what genetics could demonstrate and what the police could reasonably infer. A DNA profile from a condom can point to a biological contributor. It cannot directly reveal who wrote a particular line of a note, who placed an item at an address or what the writer intended to do next. But a match to material from the original investigation makes a completely uninformed hoax much harder to sustain as the entire explanation.
Investigators had acquired a second collection of evidence attributable to the same source. If the older material were limited in quantity or difficult to test, the newer material could provide other opportunities. Above all, the police now had good reason to think somebody connected to the 1988 crime was still active locally years afterwards. They had tightened the relationship between events considerably without yet learning the individual's name.
Think of the case at this point as several entries in a file connected by a clear arrow. The arrow from 1988 to 2004 says that the profiles are consistent with a common contributor. It does not connect either entry to a named person in Fort Wayne. Building the latter connection required an approach that could work even if the man himself had never provided a searchable sample in his own name.
A Face Predicted From DNA Was A Different Kind Of Lead
In 2015, before the family-tree breakthrough, police publicised an image produced with assistance from Parabon NanoLabs. The company described its work as DNA phenotyping. It used genetic information to predict certain features of the unknown contributor's appearance. That made for a recognisable public appeal in an investigation where no recent photograph of a suspect had been identified. It also introduced another use of the word “DNA” that can easily be confused with the testing that led to the arrest.
A predicted image is not a photograph recovered from a scene. Features are estimated with varying degrees of certainty, and many different people can fit the same broad description. Ageing, hairstyle, weight and day-to-day appearance further complicate any attempt at recognition. A member of the public could see something familiar in a prediction and supply a lead. The resemblance alone could not establish that the person they had in mind contributed the sample.
This illustrates three distinct questions scientists and investigators asked about biological material over the course of the case. Could samples collected on different dates be associated with the same source? Could any inherited features of the contributor be estimated well enough to help the public supply information? Could relationships with other people narrow the search to a particular family? A good answer to one of those questions need not be an answer to either of the others.
The sequence also puts the much more precise final comparison in perspective. A face resembling a prediction and a relationship suggested by a family tree both generate candidates. A direct comparison can test a candidate against the original material. That final change in method is easy to miss when a documentary runs images of a predicted face beside a photograph taken after an arrest. The visual resemblance may make the case feel complete. The evidential work came from going back to the biological trace.
For April's family the distinction had a practical consequence. A public appeal in 2015 meant the case was still being worked on and a new source of information might produce a name. A direct comparison in July 2018 meant officers could go to a particular address and ask a particular man to account for a match. The years between those developments were not merely a wait for a computer to produce a sharper picture.
Searching For Relatives
Investigative genetic genealogy depends on the fact that people share inherited DNA with relatives. The question asked of a suitable database can therefore be broader than whether the unknown sample matches its contributor exactly. Are there people whose shared stretches of DNA are consistent with a family relationship? If so, documentary research may establish how their family lines intersect and which descendants could fit the circumstances of the case.
The genealogist's work sits in the gap between a measurement and a set of names. A possible relationship may have more than one explanation. Family trees can contain missing information and errors. Birth, marriage and other public records can help test the connections. Age, place and generation can narrow a search without establishing anyone's guilt. A relative who helps make a link possible need have no knowledge of the crime and may never have met the person being sought.
The sworn affidavit credited genealogist CeCe Moore with research that narrowed this investigation to two surviving brothers. This was an extraordinary change in the scale of the work. Detectives had gone from an unidentified male to a small group they could investigate directly. Yet the size of that group also demonstrates the limitation of the method. Brothers share ancestors. A well-constructed family tree may correctly point to both.
It would have been wholly unjustified to announce two suspects as if each had an equal case to answer in public. What investigators needed was evidence capable of favouring one man over the other. That meant obtaining biological material associated with a specific individual and comparing it with what police already held. The next stage returned them to one of the oldest practices in detective work: observing who actually lived at a particular address.
How Police Connected A Sample To One Household
Detectives observed John D. Miller at a residence in Grabill, Indiana. The affidavit says they checked utility information and carried out surveillance supporting the conclusion that he lived alone. Those details can look like routine administration in an account dominated by new genetic technology. They were part of the reasoning that made the later laboratory result interpretable.
Waste taken from a shared home might contain material from numerous residents, visitors or people whose belongings had been discarded there. Finding a match in that material would give police a reason for further enquiries, but it would not identify which household member had supplied it. Establishing who occupied a residence before collecting discarded items did not eliminate every conceivable alternative. It made the association with an individual substantially clearer.
On 6 July 2018 officers recovered items from the rubbish at Miller's home. They included three used condoms. On 9 July the investigator who swore the affidavit heard from the Indiana State Police laboratory. The profile obtained was consistent with the profiles from the 2004 items and with the material recovered in the investigation into April's death. The three dates now formed a chain spanning thirty years.
Each link did a different job. The original sample was connected with the crime. The 2004 samples linked the later threats with the original contributor. The material collected in 2018 connected that contributor with a household occupied by a known man. Genealogy had been vital in choosing where to look. The laboratory comparison did the work that a family relationship could not.
What Miller Told Detectives
Officers approached Miller in July. He agreed to speak to them and was advised of his rights. Investigators described an interview in which he admitted abducting April, sexually assaulting her, killing her and taking her body to the location where she was found. His account also addressed the communications associated with the case. The child's family had an answer that could be tested against a record extending back to the first search.
An admission is especially informative where parts of it can be checked independently. Here the police went into the interview already possessing the biological comparisons and information about the disposal location, the barn writing and the later material. They did not have to accept every detail a speaker offered in order to see whether the principal account fitted evidence gathered on dates when that speaker had not been named.
The arrest affidavit is a sworn explanation of the basis on which police sought criminal process. It does not have the status of a jury verdict. In many cases an accused person disputes all or part of such an account, and the prosecution must prove its case in contested proceedings. In December 2018 Miller pleaded guilty to murder and child molestation. His admission in court established criminal responsibility independently of the language used in early reports describing a suspect's arrest.
An Allen County judge imposed a total sentence of eighty years. Janet Tinsley had urged the authorities to seek the death penalty. Her daughter's case ended in a guilty plea and imprisonment. Those facts can be recorded without claiming to know what a sentence could possibly mean to a parent who had first gone to the police to ask for help finding a girl walking home.
The most important distinction in this investigation is also the most easily lost in its popular retelling. A name on a tree did not make a man guilty, and a brother's place on the same tree never made him responsible by association. April's case changed when detectives could move from an inherited relationship to individual evidence, and from individual evidence to a corroborated admission. After thirty years, only one man stood before the court to plead guilty to what he had done to her.
Sources
Parabon Snapshot™ DNA Phenotyping Service Reveals New Clues in 27-Year Old Cold Case
Arrest Made In April Tinsley Homicide Case | WBOI - NPR News & Diverse Music in Northeast Indiana
