The workshop on the epistemology of scientific conclusions was winding down with a presentation by a latent fingerprint examiner explaining how he arrived at the decision that a crime scene print came from a particular finger.  His talk included references to likelihood ratios, subjective probabilities and Bayes’ Theorem. The man who had been quietly sitting next to me all day suddenly raised his hand and waived it with a sense of urgency.  He explained that he was a toxicologist and he wondered whether anyone was convicted based on fingerprint evidence alone.  Upon hearing many affirmative answers, he exclaimed, “Seriously?  Oh my God!”  He then turned to me, with a horrified look on his face, and said he assumed that fingerprint evidence was only used to indicate whether further investigation was necessary.  A man across the aisle then demanded to know why fingerprint examiners didn’t just concede that the process of identification was an art, because it certainly wasn’t science.

That was the somewhat glorious end to the workshop to which I had most looked forward at my first conference of the American Academy of Forensic Sciences (AAFS).  The AAFS describes itself as the world’s most prestigious forensic science organization.  Founded in 1948, its over 6,000 members include doctors, lawyers, chemists, engineers, psychologists and those who work in the comparative forensic sciences. Every February, somewhere between three and four thousand people gather at the annual conference to participate in workshops, listen to presentations on the latest developments in various fields, and to network with people who have researched and written about important topics in forensics and criminal justice.  

The theme of this year’s meeting was “Forensic Science Education and Mentorship – Our Path Forward” a recognition of the five year anniversary of the ground breaking 2009 National Academy of Science Report entitled, “Strengthening Forensic Science in the United States: A Path Forward.”   Referring to DNA testing as the gold standard, the NAS report was a scathing rebuke of the other comparative forensic sciences.  According to the president of AAFS, one of the goals of the conference was to assess where the NAS recommendations are being met and where they fall short.

This was certainly one reason I decided to attend.  A couple of years ago, I had a Frye-Mack (Minnesota’s variation on the Frye test) hearing on the admissibility of fingerprints.  The four day hearing included a lengthy discussion on exactly what latent fingerprint examiners should be allowed to say in court.  In other words, how should the court allow them to express the opinion that they have “matched” the crime scene print to my client?  Despite experts Simon Cole and Sandy Zabell pointing out that the phrase, “to a reasonable degree of scientific certainty,” does not exist in science, the judge ruled that this was the appropriate way for latent examiners to phrase their opinions. Given that on cross-examination, the examiner who declared the “match” admitted that her decision was based on no data, having her characterize her confidence level to a reasonable degree of scientific certainty seemed like something with which I could work. But, I wondered, how are latent fingerprint examiners around the country characterizing their opinions today?

The answer is that they are all over the place.  Some examiners are relying on subjective probabilities and Bayesian theory; others don’t know who Thomas Bayes was.  The two most influential professional groups, the International Association of Identification (IAI) and SWGFAST, fail to give much guidance to examiners on how to frame their opinions.  As the latent print examiner put it at the end of his workshop presentation, “We are trying really, really hard.  We just aren’t there yet.”  He meant that, although they are doing research on error rates and the probability of certain patterns occurring, they don’t have the data yet.  My interpretation was, “fake it until you make it.”  The problem, of course, is that they are faking it in a criminal courtroom, where our client’s freedom is at stake.  

I walked away from that workshop feeling vindicated that scientists recognize fingerprint examination for what it is.  But we don’t often have scientists on the bench or on our jury panels. We know that judges are reluctant to keep out evidence, especially fingerprints because that evidence has been coming in for over one hundred years.  A judge who presented during the workshop had a great power point slide that asked, “How many judges does it take to change a light bulb?”  The answer was, “Change?”   So, assuming that judges are inclined to admit comparative forensic science evidence, we can still attack it.

We need to request the forensic examiner’s case file, including all transmittal letters, notes, and phone call logs – absolutely everything in the file.  Then we need to make a thorough discovery request demanding to know the following:

  1. What conclusion the examiner will offer in court?
  2. What does this conclusion mean?
  3. What is the data that supports this conclusion?
  4. How confident is the examiner in this conclusion?
  5. What data supports this level of confidence?
  6. Could the source of the evidence be someone or something else?
  7. How does the examiner calculate the possibility that the source could be someone else? 

     Armed with this information, we can attack whether the examiner is an “expert” who should be allowed to testify in court.  If the expert does testify, we can attempt to limit his or her opinion.  We can ask the court for instructions that more accurately reflect what the examiner is saying.  Five years after the NAS report, the comparative forensic science community has a great deal of work to do.  And if we are doing our jobs, we may just convince a juror to recoil with horror at the thought of a person being convicted on the basis of this type of evidence alone.