Offender “profiling generally refers to the process of using all the available information about a crime, a crime scene, and a victim to compose a profile of the unknown perpetrator” (Ainsworth 2001:7). It’s based on the belief that the offender’s behavioural, socio-demographic and personality characteristics can be predicted based on the evidence from the crime scene. All criminals have a MODUS OPERANDI (typical way of acting) consisting of techniques, habits, and peculiarities in their criminal behaviour (Turvey 2002:340). This enables the profiler to infer the characteristics of an unknown offender from their crime scene (Holmes 1988:72). The popularity of offender profiling has increased in recent times due to media attention, which has misinterpreted and distorted the empirically recognised capabilities of the technique (Dowden 2007:2). It’s important to separate facts from reality, and from an objective assessment of whether offender profiling helps the police. Two approaches will be discussed in this report: the British ‘bottom-up’ approach (Statistical), and the FBI ‘top-down’ approach (clinical). The FBI approach attempts to get into the mind of the offender. Whereas the British approach focuses on behaviour. The key principles of both approaches will be discussed. Both approaches will be critically evaluated, by analysing the accuracy, methodological issues and effectiveness taking into consideration their strengths and weaknesses. This report will argue that the statistical approach is better than the FBI, as its more scientific and objective. Lastly, the report will conclude that neither methods should be used in criminal investigations, and will offer policy recommendation to improve them.
Description of the two approaches
The FBI approach was developed in the late 1970s. It was originally based on interviews with 36 convicted serial killers, combined with insights from many crimes investigated and solved by the FBI (Douglas 1986:405). Typologies of those who committed the crimes were made, suggesting that there are two different types of offenders, organised and disorganised (Hazelwood 1980:20). The premise of the dichotomy is the differentiation of criminal behaviour by its inherent level of sophistication (Kocsis 2008: 397). The organised typology is reflective of methodical and planned behaviour that is concord with offender characteristics such as high intelligence and social skills. It’s also reflective of an individual with a comparatively organized lifestyle. Whereas the “disorganized” typology is reflective of behaviours that are spontaneous, unplanned in nature, and that is said to be similarly demonstrative of the offender’s personal characteristics such as being socially inadequate and below average intelligence (Kocsis 2008:398). Ressler et al. (1986) claimed that “. . . facets of the criminal’s personality are evident in his offence. Like a fingerprint, the crime scene can be used to aid in identifying the murderer” (p. 291). The crime scene for an organised killer will reflect a methodical and ordered approach, in contrast, the disorganised killer’s crime scene will be chaotic, due to a lack of planning (Canter 2004: 500). The main objective of the clinical approach is to get into the mind of an offender, by establishing their thinking patterns. Identifying their motives and personality increases the chance of catching them.
The first difference between the clinical and statistical approach is how data is gathered and interpreted. The clinical approach’s knowledge is based on intensive, prolonged investigations and interpretation of individual cases (Alison 2004:74). It views each case subjectively. Canter (1989) who developed the statistical approach, acknowledged the undoubted accomplishments of the clinical approach. However, he argued that the principles of science should be applied to offender profiling (Canter 1989:12). Inferences should come from empirical, peer-reviewed research, as opposed to investigative experience like the clinical approach (Canter 1989:13). It’s a bottom-up approach to offender profiling as it doesn’t attempt to fit crime into an existing theory, unlike the FBI approach which is a top-down approach to offender profiling as it attempts to fit crimes into predetermined categories. The statistical approach is more scientific and objective as it analyses all the details of a crime scene and using this to develop a profile. In this approach data is imputed into a database, this shows the statistical prevalence of behaviours over a period. This allows behavioural patterns to be statistically compared (Gavin 2014). Data from previous instances are used to indicate a particular offender. (Ainsworth 2001). Base rates are used to determine how criminals were characterised and behaved in the past. This provides an incidence of crime among the general population. If you develop a profile of a violent crime, the statistical pool of data has shown its most likely to have been committed by a man. This allows for the detection of distinct characteristics of the offender and allows the profiler to separate the perpetrator from the large statistical pool, and focus on the individual. The offender may have left psychological signatures at the crime scene, these are behavioural clues. The aim is to look at the uniqueness of the crime scene, such as the treatment of the victim or the location in which the crime took place. These distinct behaviours can be used to link cases. Linkage analysis is used by both the statistical and clinical approach, to focus on abnormal similarities between crimes, likely linking them to a possible offender.
The clinical approach is subjective as the method of analysis and data collection relies on personal interpretation and intuition. The statistical approach avoids this by using content analysis. Therefore, there is a degree of objectivity as the crime scene can be operationalised. The clinical approach lacks this objectivity as there is no operationalism in place, therefore bias can occur. With the clinical approach, Individuals will interpret a crime scene differently. What one profiler perceives as organised may be disorganised to another. Subjectivity is a huge problem with the clinical approach. The FBI approach relies on the personal perspectives and expertise of a profiler (Ressler 1998:140). Therefore, it relies on individual interpretation (Alison 2005:125). The clinical approaches interviews on sexual murders as a method of data collection is problematic (Alison & Canter 1999:111). As the interviews relied on retrospective self-report data, which can be inaccurate as they rely on the offender’s memory (Canter 2004:7). They trusted that the offender did not lie about their experience and offences (Canter 2004:8). This is a mistake, as the majority of the 36 serial killers interviewed were psychopaths. They could have been selective in what they revealed, and due to their manipulative nature, they may have been untruthful during the interview. It’s also androcentric as the interviews were done on all male sexual killers. This means it cannot be generalised and applied to women. As the motives and reasoning for why women murder could be different to men. This makes any profile where the interview data are used, less reliable and valid. However, the statistical approach does rely on crime data, this is problematic, as the approach is not useful for crimes that are rarely reported such as a white-collar crime.
The objectivity of the statistical approach is good, as it allows for a greater scope. More types of crimes can be analysed, as opposed to the clinical approach which can only work for murders because of the typologies. However, the clinical approaches belief in criminal consistency is problematic. Criminal consistency is the belief that the offender will show some consistency between the nature of their crimes and other characteristics they exhibit in other situations (Canter 2010:239). The way an offender carries out a crime on one occasion will have some characteristic similarities to the way that he or she carries out crimes on other occasions (Canter 1995: 347). The foundation of this assumption is problematic, as it relies on a rather simplistic theory of personality traits and their control over behaviour (Alison 2002:125). It is unlikely that an offender will demonstrate perfect consistency; offenders will adopt some behaviour to new environments, or other factors may prevent offenders from performing their chosen actions (Mazur 2005:121). There is also likely to be a great deal of variation in offenders: some may vary their behaviour widely while others may remain remarkably consistent (Canter 2004:295). It’s deterministic to assume offenders will act in the same manner for every crime they commit. Human behaviour is erratic, there is no logical pattern. The clinical approach is also not useful when the motives of an individual are clear. The approach places great importance on identifying the motives and personality of an offender from the crime scene. Whilst there is a lot of information to gather about an individual if they are a sexual murder. However, when the motive is clear as with most utilitarian crimes, the approach has little use. Organising a drug crime into an organised or disorganised typology is useless, and wouldn’t really help with an investigation. As there would be no evidence left to indicate a typology.
No direct link between offenders being captured and clinical approach has been measured. Therefore, there isn’t enough evidence to support and validate this approach (Ainsworth 2001). Therefore, critics have suggested that the typology of the clinical should only be used as a tool during an investigation, as opposed to leading it (Ainsworth 2001). However, using the typology can be beneficial as they can provide a profile with a rough understanding of the characteristics of an offender (Ainsworth 2001), making them more prepared for the interviews with the suspect. However, it is reductionist as it reduces criminals into just two categories of organised or disorganised. Human behaviour is far more complex than organised or disorganised (Gavin 2014). This does not mean that typologies are inaccurate: Ressler (1998:294) found that the FBI data on the 36 sexual murders and their classification were mostly accurate. However, the original sample of men identified as either organised or disorganized were not based on any scientific research. Rather they were based on a combination of experience and intuition of the officers involved in conducting the study (Muller 2000:240). The separation of the men into organised or disorganised were done before any statistical tests were used to analyse the differences between the two groups. Which some argue led to a self-fulfilling prophecy, rather than a valid behavioural dichotomy (Kocsis 1998:127). Also, the two distinctive typologies of serial murders are often blurred, offenders can be a hybrid of more than one type (Canter et al 2004:303). A crime scene can appear both organised and disorganised, suggesting the possibility that individuals can be a mixture of both. The typologies also don’t take into consideration that criminals enhance their skills, they learn and improve. Therefore, an offender may have been disorganised when they started their criminality. However, over time they will get better and became an organised criminal. Overall the typologies try to reduce complex human behaviour into just two categories, which reduces the validly of the typologies.
The clinical approach reflects positivist criminology, which is the idea that an individual’s biology makes them predisposed to criminal behaviour. However, there is a lack of empirical scientific support for the FBI’s organised/ disorganised typology, and profilers lack reliability and validly due to personal bias (Alison 2010:118). In contrast, the statistical approaches objectivity allows for connections to be made. Using statistics to plot the likelihood of certain offender trails enables a more accurate profile. However, it’s important to note that science is not the be end or end-all. A clinically based experience can be useful, as it’s difficult to infer human behaviours and emotions from numbers. They don’t really allow us to fully understand the offender, as well as the clinical approach, which attempts to get into the mind of an offender. However, a big problem with a lack of control and measures in place with the clinical approach is that it’s more prone to individual mistakes and errors. If an offender is identified as an incorrect typology, this will take over an investigation, and the real offender may get away with their crime as a result. Many of the profiles provided by this approach are ambiguous and appear to describe any suspect (Alison 2003:189). Police officers found profilers to be useful, regardless if the profilers were fabricated or genuine (Allison 2010:190). Once a profiler is made officers tend to believe that the offer must fit those characteristics. This is problematic if the profiler itself is incorrect, to begin with, as offices will be wasting their time trying to find a person who most likely isn’t the criminal. A big problem with the clinical approach is that it allows assumptions to be made and for bias to take over an individual due to the lack of operationalisation in place. This can be seen from the Beltway sniper attacks, where investigators were sure that the suspect was white (Koerner 2002). This shows how stereotypes produced an inaccurate racial profiler which ultimately meant that profiling played a little role in their final capture (Koerner 2002).
However, even with these issues, both the clinical and statistical approaches have been used with the capture and prosecution of murders. The FBI approach was used to capture Ted Bundy, his victim selection and killing patterns were identified, the Bureau discovered Bundy favoured killing young women and was most likely an organised killer, which assisted the police with his capture. Canter used behavioural and geographical consistency to capture the railway killer (John Duffy). Canter’s scientific analysis of the crimes and their locations led him to hypothesise, where the culprit might live as well as personal information about the culprit (Crace, 2004). The degree of domination used by the killer, as he tied up all his victims indicated that he was a small man and was not strong enough to restrain them without help. All the victims were killed near each other, which suggested that the killer lived near the crime scene, as he would use the railway to escape. The victims were all women, which showed he had problems with his marriage and might have been sexually frustrated. Lastly, the crime scenes were all very organised, very little evidence was left behind, this shows that the killer was an intelligent man. Canter used both the geographical and behavioural to form a personality profile which eventually helped catch the killer, this shows how elements of both the statistical and clinical approaches can be used together (Ainsworth 2001). However single cases are not useful to demonstrate the effectiveness of offender profiling. Pinizzotto and Finkel (1990) found out that in 192 profiling cases, the profile contributed to the suspect being identified in only 15 of them. Despite this Pinizzotto and Finkel (1990) also found out that in 77.2% of profiling reports provided by the FBI, the information was useful in providing an external perspective on cases and helped narrow down investigations. Suggesting offender profiling should be used to assist cases not lead them.
To conclude, offender profiling has helped law enforcement. However, neither approaches should be used solely in an investigation. The clinical approach relies on evidence-based expert knowledge, this is a mistake as humans are susceptible to cognitive biases and faulty decision making (Tversky 1982:1128). The application of the organised and disorganised dichotomy is problematic, as human behaviour is far more complex than that. A recommendation for this approach is to operationalise the typologies. Before any typologies are made, rigorous and extensive research should be carried out which is backed by academic works. This will make the typologies more specific and more importantly accurate. Therefore, the profiler will have a more objective measure of human behaviour. Whilst this will require a lot of time and effort it will be hugely beneficial as it will reduce the subjectivity of the FBI approach.
The statistical approach is the most useful of the two approaches as its more scientific and objective. Its range is far bigger than the FBI approach as it can be used on different types of crime, as opposed to the FBI approach which can only be used in murder cases. It’s also useful for crime prevention as the focus on locality enables law enforcement to target these areas that have high crime rates. A recommendation for this approach is to use content analysis for a larger variety of crimes. Therefore, different types of data can be check listed at various crime scenes. This will expand the ability to use linkage analysis to assist with more types of crimes. Overall offender profiling should only be used as an investigative tool, rather than a prosecutorial tool (Chelsea 2015:142). Both approaches should be used as a tool to help law enforcement, rather than to solve cases, as they both have their flaws (Chelsea 2015:143).