Forensic fundamentals
An introduction to forensic science
The beginning is the best place to start, so my first blog post is, naturally, a back to basics primer on the what forensic science is, including:
- Defining forensic science
- A (very) short history of forensic science
- ‘Founding principles’ of forensic science
- Opinion-based evidence
- Forensic feature-based comparison
Defining forensic science
Forensic (adj.)
“The debate or discussion of a matter relating to a court of law”
Science (n.)
“The pursuit and application of knowledge and understanding of the natural and social world following a systematic methodology based on evidence”
Based on the above definitions forensic science is the application of scientific methodologies to matters relating to a court of law.
Initially, the term forensic science was limited to evidence and testimony in court given by medical doctors and surgeons. This definition was then expanded to include all the myriad of techniques used to provide forensic evidence.
Often the questions we try to answer using forensic science techniques are determining the source of a physical or digital item, or attempting to understand an activity that has taken place.
Forensic science techniques are used to help answers questions about people, objects, locations and activities related to a criminal investigation.
A (very) short history of forensic science
Using science to better answer questions about a crime is not a new idea. The Yhuan Shuh Lih (or ‘Discourse on Cases of the Bringing to Light of Unjust Imputations’), written in China in the 6th century CE, is believed to be the earliest known text on forensic medicine (forensic medicine uses medical science to determine cause of death).
The text is now lost, but the ‘Washing away of wrongs’ written in 1247 by Song Ci, another forensic medicine text, still survives today, and is essentially a coroner’s handbook including case studies and methods for conducting autopsies.
The next 700 years saw many scientific innovations that were subsequently used in a forensic capacity, including the first recorded medico-legal autopsy in Europe (1302) the invention of the microscope (1590) and the creation of a chemical test for detecting arsenic (1773).
On the 25th January 1784 in Kent, England Edward Kelsall was shot and killed by John Toms during a robbery. Toms was identified as the killer by linking the newspaper found in his pocket to the paper wadding recovered from the back of Kelsall’s skull, fired from the pistol used to kill him. This case is the first recorded example of physical fitting, or linking two items together by similarities in surface features. This is a technique still widely used in forensic science today (I’ll do another blog post discussing the importance of similarity as well as typicality in forensic evidence).
The 1800s saw the introduction of fingerprint comparison as a means of identifying offenders in criminal investigations, and was first accepted as evidence in a UK court in 1901.
It was not until the late 19th and early 20th centuries that forensic science began to be practised in a more standardised and co-ordinated fashion, notably with the founding of the first police forensic laboratory in Europe by Dr Edmond Locard, considered by many a pioneer of modern forensic science.
‘Founding principles’ of modern forensic science
In 1910 Dr Edmond Locard, a French criminologist, established the first police forensic laboratory in Europe and is credited with establishing one of the basic principles of modern forensic science:
“Every contact leaves a trace”
Locard’s exchange principle theorises that any contact between two items, a person and an item or two people will result in the exchange of material. The recovery of the exchanged material or ‘trace’ becomes the evidence used to determine the source of the evidence or the activity that caused it using forensic science techniques.
From the shedding of fibres from the clothing of the assailant onto a victim, the deposition of a fingerprint mark at a crime scene to the digital capture of an offender’s face on a CCTV recording system, Locard’s exchange principle is applicable to almost all forensic science disciplines, even cutting edge digital forensic science methods.
In practice, of course, Locard’s exchange principle is not so straightforward. Even if a trace of material is transferred there must be enough quantity of material to be recoverable for examination. Hence, forensic scientists aim to develop evermore sensitive techniques to recover evidential material.
Even if it is possible to recover trace material, the recovered sample must be of sufficient quality to be interpretable.
Two other, now somewhat contentious, principles of forensic science were also developed later in the mid 20th century.
The principle of identification by Canadian handwriting expert R. A. Huber in 1959:
“When any two items contain a combination of corresponding or similar and specifically oriented characteristics of such number and degrees of significance as to preclude the possibility of their occurrence by mere coincidence, and there are no differences which cannot be accounted for, it may be concluded that they are the same, or their characteristics may be attributed to the same cause.”
And the principle of individuality by American biochemist Paul L. Kirk in 1963:
“Individuality implies that every entity, whether person or object, can only be identical to itself and so is unique.”
Taken together, these two principals have been considered justification for forensic comparison experts to conclude that two items of evidence share a common source with absolute certainty, sometimes referred as a match or an identification.
For many years the principles of individuality and identification were held almost as a truism in forensic science disciplines, notably in fingerprint analysis, and experts would routinely conclude, with 100% certainty that a trace sample and reference sample came from the same source (i.e. the trace and reference sample match).
However, in the last 30 years the principles of individuality and identification have been challenged repeatedly as being logically unfounded, sometimes being referred to as the “uniqueness fallacy”, on the basis that with the limited data available forensic scientists cannot meaningfully determine whether an observed characteristic is truly unique.
Forensic scientists are now taking more scientific approaches to ensure methods are reliable, including:
a. Empirical validation studies
b. Determining the limitations of techniques
c. Estimating the uncertainty associated with a result
Opinion-based evidence
Forensic scientists report the outcome of their examination as an evidential conclusion, which the courts can then use to help them determine whether the defendant is innocent or guilty.
Most forensic scientist’s conclusions are classed as opinion-based evidence.
Although forensic examinations often make use of complex technology and analytical techniques, the final conclusion is typically presented as an opinion of the forensic scientist, but not just anyone can give an evidential opinion.
In order to be admissible (accepted) as evidence in court an opinion must be made by someone the court considers to be an expert.
“There are, and probably always will be, charlatans eager to offer themselves as witnesses before the courts for monetary or selfish motives. They claim proficiency in all branches of scientific inquiry: handwriting experts, firearms experts, chemists, polygraph operators, etc. The problem is not a new one to the professional ranks”
R. A. Huber, 1959
There is no single definition of what makes someone an expert in the eyes of the court, but generally it is an individual who has specialist knowledge and abilities beyond that of the jury.
Forensic scientist must be able demonstrate their expertise and specify within what field their expertise lays. It would not be acceptable, for example, for an expert facial examiner to give an opinion on fingerprint comparison, unless they could also demonstrate expertise in fingerprints.
The case study below provides an example of need for relevant expertise when presenting opinion-based evidence.
R v O’Neill [2019] NICC 12
On 2nd of August 2015 the body of Jennifer Doran was found in the burnt out kitchen of her home with three stab wounds to the chest.
Raymond O’Neill was charged and convicted of murder and arson with intent.
During the trial, Mr Kinnen, on behalf of the prosecution, presented an expert report detailing similarities in clothing, build, face shape and gait between known CCTV footage of the defendant and contested footage of the offender.
At trial, the judge deemed Mr Kinnen’s opinion on whether these similarities were sufficient to identify the defendant as inadmissible.
Mr Kinnen was determined by the court to be an expert in the preparation of enhanced images but not the comparison of people and objects in images.
Conversely, Professor Birch, an expert in forensic gait analysis, examined the same CCTV. Professor Birch’s opinion, based on his examination, was that the similarities in walking pattern (gait) lent limited support for the proposition that the figures in the CCTV were the same individual.
Despite the limited value of Professor Birch’s evidence the judge deemed it admissible, as Professor Birch could demonstrate his expert in forensic gait analysis.
The hierarchy of propositions
Forensic science providers offer a wide-ranging and ever growing list of services, from commonly known techniques like fingerprint comparison and DNA analysis, to more niche disciplines like forensic tattoo and jewellery analysis. Basically, if you can think of it there is probably a relevant forensic expert somewhere in the world.
The more common forensic disciplines fall into two broad camps, traditional or physical forensic science and digital forensic science.
Physical forensic science covers a huge range of disciplines that analyse and compare traces recovered from physical objects or people, some examples include:
· Comparing a footwear impression from a crime scene to a suspect’s shoe.
· Analysing a white powder to detect the presence of illegal substances.
· Examining the distribution of blood spatter patterns to determine their cause.
Digital forensic science concerns the extraction and analysis of data from digital devices. Once a relatively niche area of forensic science, recently there has been an exponential growth in the demand for digital forensic science, linked to continual growth in use of digital devices in our everyday lives. Digital forensic science disciplines are also highly diverse and rapidly growing, some common examples include:
· Analysing data extracted from a computer to determine if illegal content has been viewed.
· Examining a digital image to determine whether it was taken by a particular camera.
· Reviewing data from a mobile phone to establish the location of the phone at a particular time.
Despite the diverse nature of forensic science, forensic scientists are typically an opinion on one of two questions:
1. Determining the source of the trace evidence.
2. Determining the activity that lead to the creation of the trace evidence.
In 1998 scientists at the UK Forensic Science Service framed these general questions into a hierarchy of propositions to help forensic scientist across all disciplines to express their expert opinions in a more consistent way.
Forensic feature-based comparison
Forensic feature-based comparison methods can be defined as:
“Methods that attempt to determine whether an evidentiary sample is or is not associated with a potential source sample, based on the presence of similar patterns, impressions, or other features in the sample and the source”
Or more succinctly:
“The visual comparison of individual features observed in two examined samples”
DoJ Statement on the PCAST Report
So, feature-based comparison methods compare the characteristics of two or more sources of evidence to help to determine whether they come from the same source.
Comparison methods report at the source level of the hierarchy of propositions and typically compare an item of trace evidence (sometimes called questioned evidence) to reference evidence from a known source.
There are many feature-based comparison methods in forensic science, predominantly in the physical forensic sciences, such as comparing striations on a bullet casing to the barrel of a firearm, comparing a DNA profile from a stain at a crime scene to a reference database to more contentious methods like bitemark comparison and hair and fibre comparison.
Another common feature-based comparison method, which like facial comparison is also a type of biometric, is fingerprint comparison. Although faces and fingerprints are very different types of stimuli, requiring different knowledge and skills to reliably compare, there are some similarities in the way face examiners and fingerprint examiners work.
Forensic practitioners across a range of feature-based comparison methods carry out their examinations using the ACE-V framework (Analyse, Compare, Evaluate, Verify), first described by R. A. Huber of the Royal Canadian Mounted Police in 1959 and referred to as the “Identification process”:
“There are three distinct stages in establishing the identity of any person or thing through which the forensic scientist must pass---consciously or otherwise, in the course of his examination.”
1. Analysis. The "unknown" item must be classified according to its properties or characteristics. These properties may be directly observable, measurable, or implied, but they are the parts which make up the whole.
2. Comparison. Stripped of its pictorial effects, its subjective disillusions, and now seen for what it really is, a comparison is made of the properties of the item found through analysis with the known or recorded properties of others whose identity is unquestioned.
3. Evaluation. It is not sufficient that the comparison disclose similarities or dissimilarities in any of the characteristic properties of knowns and unknowns. Each property will have a certain value for identification purposes, determined chiefly by its relative frequency of occurrence. The weight or significance of each must be considered.”
Huber further alludes to what would later be the verification stage as
“...the forensic scientist "must restrict his conclusions to what he can be confident would be the conclusions of any other competent forensic scientist".”
Huber referred to what would be known as the ACE-V framework as a “scientific method”. The adoption of ACE-V is now widespread across forensic feature-comparison methods.
I hope this has been a useful introduction to forensic science. I’ve covered a lot of topics but really only scratched the surface. Look out for future blog posts for updates on new innovations and critical analysis in forensic science and bometrics.