by Fred Woodworth *
INTRODUCTION
ELSEWHERE in this issue I stated that the Ken Holt series had had major
impacts on my thinking. One example of such impact is the conclusion I
reached about fingerprints, as set out in the article following this
introduction.
The seed for this article was planted by Sam Epstein in a Ken Holt book, THE
RIDDLE OF THE STONE ELEPHANT, which I first read as a child in 1956. As most
readers of this magazine will immediately agree, fingerprinting is thought to
be an exact science, and the use of fingerprints in identification is a
frequent theme in many mystery and adventure stories.
However, in the Ken Holt book just mentioned, there's a slight murmur of
caution, but frankly I doubt if anyone but me noticed it.
In a chapter called "Whose Fingerprint?," Ken Holt and Sandy Allen want to
identify the person who tore out a key section of the only remaining old
issue in the back files of a smalltown newspaper. After deciding that the
fingerprints of the culpable person are in the dust of a window ledge in the
old newspaper office, they photograph those prints and those of several other
people, and take them in to a large city where they hope that someone with
Global News Service may be able to make a proper comparison. Their
newspaperman contact refers them to a cop, a detective captain who
specializes in such evidence. Now, the following passage touches on the
matter very, very subtly, but from the first moment I ever read it, it seemed
to me that the author was murmuring, so faintly you almost couldn't hear,
that maybe fingerprints weren't quite as cut and dried a form of
identification as everyone else would have us think. Here's the passage from
the book:
"Two different prints on this one," Steiner grunted as he used the magnifying
glass ... "Hmmm." He looked even more carefully as the boys moved closer.
"One of them..."
"Does it match?" Ken asked ...
"Close enough to warrant a better inspection." Steiner crossed the room with
Banner and the two boys close behind him. On a table near the far wall stood
a projector into which he slid both negatives ...
Reaching behind the projector to a switch on the wall, Steiner turned off the
room lights and then flicked the switch of the projector. A rectangle of
light appeared on the far wall where a screen was fastened. Steiner
manipulated the controls, and the negatives were moved around until a
thumbprint from the windowsill was directly beside a thumbprint from the
cigarette case. Then he focused the machine until the images were clear and
distinct and almost two feet tall.
"Look the same to me," Sandy said... For almost a minute Steiner studied the
two prints. He walked forward and looked at them closely, keeping to one side
to stay out of the beam of light. When he turned, he was shaking his head.
"No. They're not the same."
This, I remember thinking, is extraordinary. The two fingerprints are so
close to identical that only after considerable examination of images blown
up to TWO FEET in height -- can this expert finally declare that they came
from two different human beings.
I thought: What if the expert wanted to believe that they came from the same
person?
I thought: What if Sandy Allen, who thought they looked identical, was on a
jury deciding the fate of someone who was accused on this basis?
As years passed, I paid more attention to the subject of fingerprints. A
radio station I worked at for a long time regularly received a slick
newsletter called the "FBI Law Enforcement Bulletin," and every issue
contained a feature on hard-to-classify fingerprints. I pored over this
material.
As decades passed, I picked up here and there in used bookstores, several
volumes relating to criminology and fingerprinting. The more I studied this
material, the more convinced I became that while it was probably true that no
two persons anywhere in the world had ABSOLUTELY IDENTICAL prints, in
real-life situations where there were partial prints, or blurry prints, or
exceedingly faint prints, it might well be the case that prints from a crime
scene would appear to match those of some defendant. By this time, I may as
well say, I had lost every shred of confidence in the objectivity and even
the basic human decency of most police officers and prosecutors; and I could
now all too easily imagine how some utterly innocent person could be
convicted following arguments that his fingerprints matched some pictures in
a police crime file.
The article following this introduction is the result of this rather long
cogitation (I read the Ken Holt book in 1956 and a few times thereafter, and
wrote the article in 1997).
At the time of writing it, I believed I was the only person anyplace who had
doubts about this supposedly exact science. However, following publication of
my article, at least a few people started to pay attention, and my article
got reprinted in another magazine. Then, late last year, an article in
"Lingua Franca, a Review of Academic Life" covered this exact subject; it was
called "The Myth of Fingerprints: A Forensic Science Stands Trial," and was
written by Simon Cole, who hit most of the same points I did, omitting only
the reproducibility of fingerprints. Cole was backed up, amazingly enough, by
Harvard University, which has published his book: "Suspect Identities: A
History of Fingerprinting and Criminal Identification" (Harvard University
Press.)
On December 16, 2000, The Economist in London, in its section on Science and
Technology, published a similar indictment of the fingerprint. The article
notes that the subjectivity (Sandy Allen says the prints match; Capt. Steiner
says they don't) "...puts fingerprinting on shaky theoretical ground." The
publication goes on:
"And two other things make the situation worse in practice. The first is that
fingerprints found at crime scenes tend to be incomplete. What are being
compared are not whole prints, but mere fragments... The second difficulty is
that most fingerprint evidence found at the scene of a crime is 'latent'. In
other words, it requires treatment ... to make it visible enough to work with
-- and even then, it is often indistinct. How valid it is to compare such
'filtered' evidence with the clean crisp prints obtained from suspects in
controlled conditions is another unexplored question..."
Worse yet, from my point of view, is the fact brought out in another article
I've seen just now, that in cases where a print is found at a scene but is
incomplete, "computer enhancement" is often used to "RESTORE" the missing
sections! Good god! The computer invents a print and compares it to yours!
Further problems with fingerprints, in my estimation, lie with the elastic
deformation inherent in the fleshy pads of fingers -- or in the materials the
prints are deposited on. Suppose your signature, written on a stretched piece
of rubber, was compared to the same signature on a sheet of paper; how much
resemblance would there be? With prints, such comparisons are often not
science at all, but opinion. The Lingua Franca article relates how the FBI's
Stephen Meagher, a supervisor fingerprint specialist, in February of 1999
tried to discredit a challenge to fingerprint identification by sending two
prints to FBI labs in all 50 states. The fingerprint EXPERTS at the labs were
to identify the prints and, presumably, all come up with the same result.
They didn't. SEVEN of the laboratories couldn't agree that one of the prints
matched, and five of them couldn't agree that the other did.
(So then the FBI sent ENLARGEMENTS of the disputed prints back to the
recalcitrant labs, and "suggested" that the technicians take another look.
The FBI supervisor ordered them to "test your prior conclusions against these
enlarged photographs with the marked characteristics," and he'd conveniently
drawn arrows pointing to where people were supposed to find a resemblance.
Not surprisingly, under this sort of pressure, the labs all then decided they
agreed with their supervisor.)
Whether my article sparked this firestorm of criticism of fingerprints, or
whether these other writers independently had been thinking along the same
lines, is something I don't know. I do know that my own piece was first, and
if my criticisms of this "science" had something to do with focusing some
others' thought on the topic, I'm very, very glad indeed. And, appropriately
for this present issue devoted to Sam Epstein's writing, I therefore want to
signal the tip he gave me long ago that started at least my speculations in
this direction.
I imagine that Sam had pondered the possibility of prints' seeming identical
when they were not, and in writing the passage in the Ken Holt story he drew
slightly on that prior speculation to create a realistic scene, as opposed to
some Stratemeyer Syndicate cartoon in which prints are treated like words on
a sign, where any old person can look at them and know in a second if they
match some others. In short, he packed REAL THOUGHT into the book; and this
is only one of dozens, possibly hundreds of places, where a murmured word
gave rise to decades of pondering in my own mind. The article that follows is
just a single result.
And remember: a book just published by Harvard University now expresses these
same views. [Ed. note. The book is Suspect Identities: A History of
Fingerprint and Criminal Identification by Simon A. Cole, published by
Harvard University Press, 2001, 369 pages.]
* Mr. Woodworth is a writer and printer living in Tucson, Arizona. This
article originally appeared in a publication edited by Mr. Woodworth: The
Mystery and Adventure Series Review, No. 34, Summer 2001. The address is: The
Mystery and Adventure Series Review, PO Box 3012, Tucson, AZ 85702.