Thomas Mollett – February 2021
As familial searching is still a relatively new concept and still under much debate and implementation, and since it is applied differently across the world, there is no singular definition for it. But, broadly seen, familial searching revolves around the concept of where an unknown DNA sample is deliberately run through a database in order to find a list of profiles that are genetically similar to the unknown profile and to use this information as an investigative lead to interview family members of the near matches (DNA Project, 2017).
This concept relies on the principle that special inheritance patterns link siblings, parents, children, and other close relatives (Rushton, 2011). A child inherits one set of chromosomes from the father and one from the mother (Goodwin, 2007:20). It can therefore be appreciated that two siblings will share similar alleles and hence have reasonably similar profiles. In the case of identical twins, for example, the profiles are exactly similar.
If an unknown sample is found on, for example, a crime scene and there are no suspects, the first attempt would be to get a direct match, thus a match to somebody on the database. Failing to get such a match, a familial search would deliberately run the profile through the database with the hope of finding a link to someone on the database that may be related to the unknown person (i.e. the source of the crime scene DNA), like a brother or sister, or a parent. The search on the database may yield a list of candidate relatives. Investigators would then locate and interview these candidate relatives in the hope to get information about relatives that may tie in with crime scene information or fit a general profiling description – latter which would be constructed by other information; there may, for example, be some eyewitness accounts that may give some supporting clues, such as hair colour or body type, which may then potentially lead investigators to the perpetrator, or at least to the person who is the source of the crime scene DNA.
Value and Limitations: Debate
Although not formally conducted in all countries, familial searching has been used successfully. In California in the USA, in what is dubbed as the ‘Grim Sleeper’ case – and seen as the first high-profile successful use of the technique – Lonnie Franklin was apprehended after a familial search was conducted. Franklin was linked to 10 murders over a 22 year period after a partial match was discovered via his convicted son’s DNA and DNA from various crime scenes. The familial search was key to sparking the investigation into Franklin who would otherwise not have been identified and apprehended (Grad, 2010).
To explain this: The investigators had DNA from the crime scenes. They did not know to whom the DNA belonged. There was no direct match on the database. After deliberately running the profile through the database, a candidate relative of the perpetrator presented. This candidate relative was a man in prison; that is why his profile was on the database. The investigators then viewed relatives of this man as a person of interest and the investigation then directed focus onto the man’s father. They put the father under surveillance, obtained his discarded pizza, and found that his DNA matched that recovered at a Grim Sleeper crime scene.
Familial searching also has the potential to protect and clear the innocent. Daryl Hunt spent 19 years in prison for the brutal rape and murder of a newspaper editor in 1984, but was freed in 2004 after investigation of a partial match resulted in the confession of another man. It was 10 years after his conviction that DNA indicated that another individual had been involved in the crime, but took another 9 years after the DNA testing a familial search was used to track down the killer. The 1984 sample was run through the North Carolina database a revealed a near match to a convicted felon Anthony Brown. This match indicated that a close relative to Brown rather than Hunt was the real perpetrator. By locating Brown, further investigation led them to Willard Brown, Anthony Brown’s brother. The police obtained Willard Brown’s DNA from a cigarette butt, which confirmed him as the perpetrator – he confessed to the murder. In this case, familial searching ended an injustice that would otherwise not have been corrected (DNA Forensics, 2017).
The intentional search of CODIS for partial matches to an item of evidence offers law enforcement agencies a powerful tool for developing investigative leads, apprehending criminals, revitalising cold cases and exonerating wrongfully convicted individuals (Kim, 2011). However, there are still widespread problems worldwide with the official implementation of familial searching, which raise practical, technical, rights-based, legal and ethical concerns (Rushton, 2011).
When 10 random people were to be stopped and asked their opinion on familial searching, one would probably get 10 different answers. Concerns would mainly revolve around the reliability of the process, privacy and then in particular to what degree DNA is perhaps incorrectly perceived by the public.
Possibly partly due to the ‘CSI effect’ DNA is seen as this ultimate and infallible silver bullet: the impression is created that a perpetrator will always leave DNA behind, and also that whatever DNA is found on a scene, must point to guilt. This may make people afraid to be on databases in case of profiling mistakes. In addition, but coupled to that, the unfortunate reality is that in many countries of the world, and particularly in South Africa, the law enforcement agencies are not always seen as trustworthy. Many high-profile cases have shown up mistakes made by the police, which may have either resulted in a wrongful conviction or a wrongful acquittal. Primarily for these reasons, people are jealous of their DNA (i.e. see it more intimate than e.g. fingerprints) – they don’t simply want it on a database where it can potentially wrongfully implicate them in a crime that they did not commit. These are legitimate concerns. While not all perceptions about the SAPS may be valid, one cannot exclude the possibility of mistakes in the investigative chain and also of malicious intent. Perhaps one of the biggest concerns about familial searches, is that it may over/misdirect an investigation (over-reliance) – thus the investigators start from the “match” and work backward – and only focus on that lead and perhaps neglect other leads and information. This may even lead to malicious investigation and prosecution, essentially based on confirmation bias, where the investigators become so convinced of the guilt of one person that they pursue that person with all their might. They may even for example fabricate evidence to assist their efforts. One may argue that if you are innocent, then you should be able to prove your innocent, but apart from the fact that the onus is not and should not be on the accused (according to South African law) – the might of the State may be too strong to fight. In addition, there may be aspects at play, like prosecutor fallacy – i.e. overstating the strength of the DNA evidence in court. And that since it is a technical field, that judges and jurors do not always understand the evidence. The public is rightfully afraid that these factors may lead to wrongful convictions if their DNA comes into the wrong hands or if there are mistakes along the way.
Other concerns are human rights and privacy. To explore this concern, we need to jump somewhat forward and assume the situation where “innocent” people’s profiles are on databases. What right do we have to involve innocent people in activities, such as criminal activities, in which they had no part? Let’s look at an example: An 80-year-old grandmother, a good citizen, pays her taxes, contributes to society constructively, who has never been involved in any criminal activity. Her profile is on the database because her one son is missing – and she voluntarily submitted her profile on the database in the hope of finding him. Now there is a crime somewhere. DNA from the scene – through a familial search – leads to this old lady – i.e. suggesting that a close relative of her may be the source of the crime scene DNA. Now the question is: Do “we” have the right to barge into her house and demand information that may assist with the investigation? In a perfect world, the answer may seem straightforward – all citizens should simply assist – but a second part to this situation, may prompt us to think differently about it. Whether the mother knows that one of her other sons (even including the missing one) may or not be involved in the crime that the investigators are onto, can we really expect a mother to “split” on a child of her – i.e. to potentially expose a child of hers to arrest and face prosecution? Again, the answer may seem simple, but is it? Will you give up your brother or sister, or your son or daughter up for possible arrest? The law may require that from citizens, but in reality? Especially in a case where you may not even be sure about the relative’s guilt or not, should the pressure be put on a family member to potentially expose the relative to a criminal investigation? Perhaps due to evolutionary programming, family members are generally looking out for each other – and one can imagine that to assist the police to put a family member away for life – even while it is your legal duty to do so – may result in friction in a family, resentment, and feelings of guilt. Whatever the legal position may be, from a reality perspective this must be seen as a potential limitation – albeit resistance to carte blanche admission to databases.
Other concerns: “Living under cloud of suspicion” / “If you refuse, it is a sign of guilt” / Groups that are disproportionally more on databases (Rushton, 2011).
There are other problems and potential limitations of familial searching – among other costs and the development of databases. I will not go into too much detail about that in this essay, as w.r.t. a South African context, this essay focusses more on “who should be on the database, and why?” In order to further debate this issue, let’s briefly look at current South African legislation. Loosely verbatim from 15M #7 of the Government Gazette:
- “. . . Familial searching means a technique whereby a forensic DNA profile derive from i) of a missing person, or ii) obtained from a family member of a missing person, is deliberately searched against the Missing Person and Unidentified Human Remains Index and the Crime Scene Index of the NFDD to obtain a list of forensic profiles that are almost similar to the forensic DNA profile derived from a sample referred to in subparagraph i and ii.
- The Service may use the results of the familial searches referred to in (a) as an investigative lead, by a specially trained police official to i) interview family members of the near matches; or ii) identify unidentified human remains.
- The results of familial searches must be dealt with in a sensitive manner” (Criminal Law (Forensic Procedure) Amendment Act 37 of 2013. 2014).
We can accept that “missing persons” only refers to a person that is not implicated in crime. And that South African legislation only allows searches against the Missing Person and Unidentified Remains Index, but also against the Crime Scene Index – the latter referring to profiles obtained from crime scene samples. Considering that there are six indices in total, including Convicted and Arrestee indices – from a forensic and crime-solving point of view, it would have been good if the latter two indices were included. This is not limited to South Africa alone, but some states in the USA would run against these indices – and in the UK they have no problem adding profiles to the database even on the remotest arrests. They add about 40 000 profiles per year. (Commonwealth countries are silent on familial searching. UK and most of USA and New Zealand have no legislation, NZ only serious cases. Canada expressly forbid it. Dutch, only in specific circumstances.)
So, to debate the issue further from a South African perspective, the question is, when and why and from whom should we include profiles – with familial searching in mind? Should the profile of a man who was arrested for speeding, be put on a searchable index? We have looked at some legitimate concerns and possible limitations earlier, but to critically view this from a South African perspective.
We are living a country where crime is rife. South Africa has one of the highest crime rates in the world. With outstanding high rates of rape and murder. In Cape Town area there has been an increase in murders of 40% in 2016 (Etheridge, 2017). Newspapers are full of murders. It is relentless and the practical outcome of this is that South Africans generally live in fear behind barbed wires and high fences. Everybody complains about crime. The government and police, and perhaps rightly so, are blamed. Of course, crime is a complicated matter. The high unemployment rate and poverty cannot be ignored as contributing factors. However, as science developed we are now presented with an opportunity (in the form of familial searching) to help fight crime. We must remember, familial searching would not only help to find criminals, but the use of it may also act as a deterrent. But there is resistance. Human rights. Invasion of privacy. And so on. Again, these are legitimate concerns, but the question is, what are we willing to sacrifice for a safer environment – or do we want to continue sitting behind our high fences and just complain about crime? Being unsafe is perhaps not a choice of our own – but improving safety is.
However, to juxtaposition all of this, and although not familial, though somewhat similar in concept, the following real-life example does touch on and ask the same ethical questions.
Over a thirty-year period a man killed at least 10 people in Kansas, with letters signed BTK (Bind, Torture, Kill). As police developed a case against Dennis Raider, they obtained a court order for a pap smear his daughter had given years earlier at a university medical clinic. Investigators compared an STR profile of the clinic specimen (thus of the daughter) with DNA from the BTK crime scenes. Through reverse paternity (statistical) testing, the investigators concluded that the daughter (who did not even know about the testing and thus also did not give consent) was most likely the child of the killer. Raider confessed and pled guilty to the murders (Nakashima, 2008).
Obviously, as many would agree, the outcome is great, but how ethical is it?
The way forward?
Ways to overcome these fears of the public is through education. People should know that while DNA can be extremely valuable evidence, it is only part of an investigation. It will be tested through a legal process. They should learn that even respecting Locard’s principle, it is not to say a perpetrator will always specifically leave DNA evidence behind on a scene, and even if they do, that it may not have been collected by the investigators. But more importantly in the context that we are dealing with, DNA on a scene does not mean that the owner of the DNA committed the crime, even that the owner was there – secondary transfer can transfer your DNA to a place that you have not been to. Other evidence, such as circumstantial evidence (i.e. alibi) must also be considered. People should be educated about DNA – but also of the investigative and legal processes. But with this in mind, faith must be restored in our criminal justice and legal systems. The public must trust the system and that there would not be a misappropriation of their DNA. Quality systems must be implemented and followed. The process must be transparent and performed responsibly. Parameters must be defined clearly by legislation, and they must be communicated to the public. Formal mechanisms of oversight must be implemented and an independent body must oversee processes. Science and technology parameters must be set. There must be a high level of discretion applied during investigations, and investigators must be well-trained (Rushton, 2011).
Familial searching must be sold as only an investigative tool exclusively used to create leads and only used when other methods have been exhausted.
Download a free e-copy of Bloody Lies Too here