If you ever find yourself in court, as

If you ever find yourself in court, as
a member of the jury, lawyer or even as the defendant, you would expect and
assume the evidence to be infallible. 
However, this may not be the case. In this essay I hope to explore both
side of the debate, arguments to show these methods are reliable, what makes
them unreliable and also whether other types of evidence are reliable.

The application of scientific
procedures to solve crimes is known as forensic science (1). Forensic science
can be split into some of the following categories; toxicology, anthropology,
digital forensics and DNA analysis.  Over
the last several decades there have been dramatic changes to the world of
forensic science due to technological advances. The discovery of DNA
fingerprinting by British scientist Alec Jeffreys in 1984, laid the foundations
for science and its use in the legal system. It was first used in England to
verify a confession made by a 17 year old, and by using this technique, police
were able to confirm that he was not guilty. Police later used this same
technique to convict the real killer (2).


testing and analysis is considered to be the most reliable forensic field.
There has been a vast amount of studies and research undertaken to investigate
its reliability. The process has been
around for several decades and has been successfully used around the world to
convict criminals or exonerate if wrongly convicted. In just one human
cell, there are over 3 billion DNA base pairs present. It is a common idea to
believe that your DNA is completely unique, but surprising research has shown
that only 0.1% of your DNA is unique to you. This will still however be
sufficient to create a DNA profile unique to you (3). A technique known
as the polymerase chain reaction is a valuable tool used by DNA experts to
amplify samples they may receive, a minute flake of skin found at a crime scene
for example. This process creates a larger sample, making it easier for
examiners to work with. You can then use this larger sample to create a DNA
profile. This is also extremely useful when examining old samples or samples of
low quality. Harry T. Edwards, a U.S. federal judge
explained in an interview with FRONTLINE magazine that, “DNA is really the
only discipline among the forensic disciplines that consistently produces
results that you can rely on with a fair level of confidence, when you’re
seeking to determine whether or not a piece of evidence is connected with a
particular source.” (4).  As DNA is believed to be so reliable, other
fields within forensic science have began to adapt their techniques to make
their methods as consistent and valid as possible. Forensic toxicology is another discipline that
is very reliable. It is the analysis of samples to check for drugs or other
chemicals present in the blood, saliva or urine. It has a large quantity of scientific testing to prove it validity and
reliability, which can be backed up by The National Academy of Science (NAS).
The have stated that “there is adequate understanding of the uncertainties
and potential errors in the analysis of controlled substances due to rigorous
testing” which suggests that there is sufficient data to believe that drug
testing is reliable. Analytical chemistry has been used to develop the
technique, using methods to see whether the chemical present in the body is
linked to the cause of death. The Klimisch score is a way of evaluating the
accuracy of data but particularly data from toxicological studies. There are
four codes that can be applied to the data produced; “1-Reliable without
restriction, 2-Reliable with restrictions, 3-Not reliable and 4-Not assignable” (5). A code 1 would be
the best score and this means that the study was performed using generally
accepted techniques and following the guidelines. On the other hand, a code 4
would be allocated to a study if the techniques and experimental details had
not been set out clearly. This scoring technique was designed by scientists to
improve the consistency across all studies. This will increase the reliability
of the information as a higher scoring piece of evidence (code 1 or 2) will
carry more weight in court as the jury and judge can be confident that the
tests were carried out following guidelines and that the method was deemed
acceptable. However, others
argue that there are other methods that are not up to the same standard and
therefore more research needs to be conducted.


are one of the most common techniques used by forensic examiners to work out
who was at the scene. There is a basic assumption that everybody’s fingerprints
are different, including identical twins. Even criminologists and law
enforcement officials vowed that fingerprint identification was infallible.
However, after an important development in 2004, the FBI no longer testifies
that this is true. In March 2004, a terrorist attack in Madrid, killed 191
people and injured almost 2,000. After the bombings, a single fingerprint was
found by crime scene investigators on one of the detonating devices, it was
then sent to the FBI by the Spanish Police. At the same time, several thousand
miles away, Brandon Mayfield was a lawyer in America He had previously served
in the Army. Due to his former employment and a previous arrest 20 years prior,
his fingerprints had been stored in a national database. When an FBI forensic
examiner scrutinized the print, he concluded that it belonged to Mayfield. Although
there were 20 potential matches, the expert believed the differences in the
prints were insignificant as there was more than 15 points that matched
Mayfield and the print found at the scene. Four other experts also examined the
prints and they all came to the same conclusion. Mayfield was arrested on
suspicion of involvement in the attack. He was later released from custody,
after records revealed that he had never travelled to Spain before. The Spanish
National Police announced soon after that they had matched the prints to
another man who had a long police record and also was a resident in the Spanish
capital. But how did this go so wrong? The print found at the scene was only
partial which meant there was not enough evidence to be able to make such a
confident accusation. This reduces reliability of analysing fingerprints as the
full, intact print must be found at the scene for it to be reliable. Often at a
crime scene or place of interest, full fingerprints are not found. Like stated
previously, people believe that fingerprints are completely unique and will now
take extra caution not to leave a trace behind.

Partial prints
are more commonly found at a scene, this is because it is not often that
someone will lean with equal pressure onto a surface for example, and they are
often left accidentally. They are not a reliable, as shown in the case
involving Brandon Mayfield, which will make it harder for forensic examiners to
give a definitive answer. Examiners must make sure that the evidence and
analysis is correct because an incorrect statement can lead to a wrongful

Several other
forensic techniques have not been able to generate dependable results and
therefore their reliability has been questioned One of these reasons is that
there is unreliable or insufficient evidence to support the use of certain
techniques. Bite mark analysis, more formerly referred to as forensic
odontology, is one of the most controversial forensic techniques due to the
lack of research and the variability that occurs between samples. The most
variation between samples comes from the amount of swelling soon after the
bite. Other factors include the amount the skin deforms after a bite and how it
heals. The research is improving due to more modern technology. However, a
journal article published by the Californian Dental Association states that
using cadavers to analyse different bite marks is not an effective method
because studies “will produce totally different data results than research
conducted on living people.” (6). Although this
information is valuable, it is also extremely difficult to find volunteers for
the research which limits the quantity of research that be conducted. Fire is another discipline that, although
there has been extensive research into it, experts do not fully understand the
chemical processes that take place. More research needs to be carried out,
focusing on the variability of burn patterns and how these change if other
accelerants are present, for example petrol. However, the presence of materials
such as polyester, often found in cushions, can mimic the spread of a fire
fuelled by petrol (7).
This makes it more difficult for forensic fire investigators to make clear
decisions on key details about the fire, such as how it started and if there
was any accelerant used. Gerald Hurst was quoted to say, “I could take almost
any fire and if I were so inclined, convince a jury that it was arson. It’s
frighteningly simple, frighteningly easy.” Hurst was a leading fire expert who
worked in the industry for several decades as well as publishing reports and
investigations. He explained in these reports that due to advances in fire
investigation and an inconsistent study, a man convicted of arson, Cameron
Willingham, and facing a lethal injection was not guilty. Although this
research was not enough to get the Willingham released, Hurst was also part of
another team, which was able to prove that the same defective techniques that
had been used in the first case had also been used in a second. This report led
to the exoneration of Ernest Willis, who had also been on death row after being
wrongly imprisoned for the killing of two women. This quote also links to
another argument raised when questioning the reliability of forensic science.

It is a common procedure for expert
witnesses to give evidence in court within their proficiency, and give their
opinion on what this evidence may mean or suggest.  Experts are sworn into oath and are therefore
required to tell the truth. The American Board of Criminalistics’ (ABC) and the
American Academy of Forensic Sciences have both produced codes in which all
members have an responsibility to tell the truth and the findings should be offered
in such manner that is not confusing or misleading. Even though these codes are
in place, there are still circumstances where they are not followed. This reduces
the reliability of each practitioner involved as they are not following the
guidelines set out to make all procedures valid and consistent. A study by
Brandon L Garrett and Peter J Neufeld investigated connections between an
expert testimony and the convictions, which then led to exonerations, due to
post-imprisonment analysis of samples. Garrett and Neufeld found 156 cases and
studied the court transcripts. They found that 60 percent of those reviewed
included a testimony that was deemed invalid. They split the different types of
void testimony into six categories which included; “non-probative evidence
presented as probative (meaning tending to prove), discounting exculpatory
(meaning proving innocence, defending those convicted) evidence and the
inaccurate presentation of statistics” (8). Like all
experiments, analysis tests have an error rate that is not zero. However, a
report published in 2016, stated that expert witnesses often exaggerated
claims, using terms such as “100 percent certain”, “zero” or “vanishingly
small” and these statements are not scientifically valid (9). They are also
subjective; what one expert may believe to be insignificant could be deemed as
imperative to another, and such remarks could be confusing to the jury and
others present in the court.

Within laboratories, there are certain
standards that must be met in order to create valid and reliable results across
the board. The United Kingdom Accreditation Service certifies that a lab is
adhering to these regulations at the ISO/IEC 17025 level. They test the
“continuity of evidence, management of case files and storage of exhibits. In
addition, accreditation determines the competence of staff, the validity and
suitability of methods, the appropriateness of equipment and facilities, and
the ongoing assurance through internal quality control. (4)” If a laboratory
was to deviate from these strict criteria, a chain of problems will arise.
Randox Testing Services drug and alcohol testing service which has used by police
across England and Wales. In November 2017, the news broke that there may have
been a manipulation of data within the company. It was later revealed, by
several newspapers, that this could have potentially affected over 10,000 cases
dating back to 2013. Two employees of the firm were arrested on suspicion of
evidence manipulation.

Another example where laboratory
standards were not followed closely was the Castro vs. the People of New York.
In 1989, Joseph Castro, aged 38, was suspected of the murder of his neighbour
and her daughter, at their apartment in the Bronx. A forensic laboratory report
concluded that the blood found on Castro’s watch matched the blood of the
victims. (5) A 12-week pre-trial hearing took place. The pre-trial was in place
to conclude if the laboratories involved had carried out the procedures in line
with the scientific standards set by the court. It was essential that the
widely accepted techniques were used to ensure the results produced were valid
and reliable. An expert testimony revealed that this particular lab had failed
to use the recognised techniques, which could have proved that the blood found
on the watch was the blood of his neighbour. However, the court did allow the
DNA tests to prove that the blood was not Castro’s. This is because determining
a match for a DNA sample is much more complicated that ruling out a match.


Forensic evidence is not the only method
used in court. There will be a search for any weapons nearby or at the
location, CCTV will be used to follow any alibis the defendant may claim to
have and the defendant will also be investigated to see if there was any motive
behind the crime. Alibis are a source of testimonial evidence and can be
presented in court by defendants. They can provide crucial evidence to the
case, for example, saying that they were with or saw the defendant at a certain
time and this can also provide a time frame for investigators to focus on. However,
a study in 2015 by Ricardo Nieuwkamp highlighted that only 1.6% of alibis presented
in court, out of 191 studied, were deemed as believable. Alibis are believed to
be extremely biased. This is because usually the alibi is a friend of the
accused or someone who knows them personally and so this type of evidence is seen
as weak. There has been very little investigation into the reliability of
alibis and therefore another type of evidence would have to be used in
conjunction to increase the reliability of testimonial evidence. Additionally,
humans have a natural bias, known as cognitive bias, which can be broken down
into contextual bias, when someone is influenced by any background information
they may already know, and conformational bias, when someone us trying to
interpret their findings to support what they already believe. This is commonly
found in witness testimony, where a witness to the event will stand in court
and describe the events that took place. They should explain in as much detail
as possible to give the jury a clear idea of exactly what happened. However, it
is common for these statements to be slightly deviated from what happened due
to shock or fear. Witnesses may collaborate after the event and talk about what
they saw; this could cause their story of what happened to change, as it is
common for each person to remember the sequence of events differently. There is
also evidence that the way questions are staged in court can have an effect on
what the witness remembers. A study conducted by Elizabeth Loftus in 1974 discovered,
through the use of psychological techniques, that the different terms used when
the witness was asked questions had an effect on the response given (10). For example, the
word “crash” made the witnesses exaggerate their statements more than if the
word “collision” was used. This study highlighted the effect that a lawyer or
other legal executive could have on the statement provided and similarly to the
reason stated previously, another type of evidence should be used alongside testimonial
evidence to back up the statements given. It also shows how easily a witness
can be manipulated by lawyers and therefore reduces reliability as there is no
way to judge how much psychological tricks have had an effect on the witness.

CCTV is an example of documentary
evidence. It is can provide a definite answer of where the person accused was and
a specific time frame. This increases reliability as it is pure factual
evidence and therefore there can be no manipulation of the records. Even though
the science behind technology has dramatically improved over the last decade; the
number of mistaken identity cases has also increased. Identity is easy to keep
hidden, with attackers often using materials to cover their faces and baggy
clothing to cover their build. This makes it harder for investigators to be
fully certain they are following the correct person. So although there is a
definite answer as to whether someone was or was not present at a certain place,
the question is, is the person on the CCTV the person the investigators are looking
for. Additionally, CCTV also does not give the whole story. Cameras are only
positioned in certain areas, in more public areas or areas of high crime. In the
UK, as of 2007, there were over 4 million surveillance cameras installed, I would
expect this figure to have dramatically increased over the last decade in order
to keep up with the growing population (11). The rise of CCTV
and surveillance camera use has made it more probable that the suspect will be
caught. However, there has been opposition to the increase in the number of
cameras installed with people calling it a breach of their privacy so a moral
issue has arisen, should we increase dependency on the cameras to help increase
the reliability of other types of evidence, such as forensic? There is also the
problem that the surveillance cameras are not continuous. There may be breaks
in footage where there is a long distance between the closest cameras.


Ultimately, I believe that the
majority of forensic evidence is reliable. There are countless journals and
articles explaining why forensic techniques are unreliable and this in itself
shows that it is in fact reliable.
Although this seems contradicting, it may explain why the quantity of reports
from both sides differs so much. The evidence produced in these reports against
forensic science has so far not been substantial enough to change the way that
forensic evidence is used in court. I still believe more research in all
aspects of forensic evidence should be conducted especially in areas such as
fire, as this is still a grey area. However, the most common techniques (DNA
and toxicology testing) used to analyse evidence are two of the most
reliable.  They both have a substantial
amount of research behind them which has influenced other disciplines to increase
their standards to match. However, there are a few ways that could increase
reliability of these forensic techniques. Biases are difficult to remove, like
explained before all humans have a fraction of natural bias, and even if humans
were completely logical this would still be present. “Sequential unmasking” is
the act of analysing all samples, documenting evidence and findings and then
once this is complete, comparing the similarities. Although this is the
procedure for some laboratories, it is not yet mandatory and the majority of
labs will analyse as they go along. It is also suggested that there should be a
separation of laboratories that collect the samples, labs that test and analyse
the samples and finally the labs that interpret the results (12). By following this
code of conduct, the amount of bias would significantly decrease due to the
independence of each stage which therefore would decrease contextual bias. It would
prevent each examiner having an idea of what may have happened and would
possibly prevent the examiner from trying to interpret the results to fit their

I believe that when several types of
evidence are used, the reliability rate would increase as each piece of
evidence will support another. Therefore, I believe that although forensic
evidence is a fairly reliable source of evidence, it should be used alongside
other types, such as documentary, and this should increase reliability.