20 March 2012
Profiling a Killer
Professor Glenn D Wilson
Psychological profiling is a useful adjunct to police investigations. It is not a substitute for forensic evidence but may help to focus the search by suggesting what sort of person is likely to be responsible for a crime, where they might live and work and how they might travel.
The type of psychologist popularised by the TV series Cracker excites the public imagination but real profiling is both more prosaic and more scientific. Whereas Cracker brings to bear uncanny intuition into what is going on in the mind of the killer and uses artful interrogation techniques to extract confessions from suspects, modern profilers use cold logic and look to what research experience tells about a crime scene and the likely perpetrator. There is nothing mystical or magical about psychological profiling and certainly nothing “psychic”.
Take the most famous of all serial killers, Jack the Ripper. This is salient in folklore because it occurred around the beginnings of “tabloid” journalism and because the case was never solved. While it is interesting to speculate about the perverted sexual motives of Jack, this would not necessarily have narrowed the search for him. However, details such as his choice of victim, the times and places at which the murders occurred and the mode of killing may have helped. The police surgeon who examined the injuries to the Whitechapel victims (Dr George Phillips) reckoned that the way in which their organs were removed so expertly suggested professional skill, pointing perhaps to a surgeon, butcher or mortuary worker. Phillips’ opinion has been disputed, and we may never know whether he was right, but this is exactly the sort of deduction used by today’s profilers.
Serial killers are distinguished from spree killers by killing 3 or more people with down-time (intervals) in between. They show a number of common traits. They are usually men of around 20-40 years old. They may be loners or have a history of unstable relationships. They move house frequently (perhaps distancing themselves from their crimes). Their employment history is also unstable – usually low-level, itinerant, male-typical jobs that provide no career development (e.g., security man, fairground worker, driver, barman, bouncer or wheel-clamper). They are not usually mad; just psychopaths with a background of child abuse, irresponsibility and spotty criminality such as burglary, trespass and assault. It goes without saying that they are self-serving, skilled liars and lacking in empathy for their victims.
Serial killers are sometimes said to be charming, plausible and intelligent. This may be because the successful ones become best known; those less bright and persuasive get caught on an early outing, so are “nipped in the bud”. In addition, interviews with those most eloquent are more frequently aired. They are also said to be mostly white, but this may be down to selective reporting of cases that occur in “respectable”, middle-class neighbourhoods. Based on percentages of the population, no particular race is under-represented (Morton, 2011).
The motives of serial killers are frequently sexual and sadistic, and may include revenge, impotence and hatred for women. They tend to lack self-esteem and have a need for empowerment. They kill strangers (of a particular category) using intimate, contact forms of killing like as beating, stabbing and strangulation. They may take pleasure in the cruelty inflicted on their victim as evidenced by overkill (e.g., stabbing the victim many more times than necessary to kill them or disembowelling them). Some take souvenirs (trophies), such as clothing, jewellery or body parts to help relive the crimes. Some pride themselves in their ability to terrify the public and baffle the police. Ultimately, they may need to get caught in order to take the credit. Some killers claim to be mission-oriented (e.g., ridding the world of undesirables such as prostitutes); these may be psychotic or rationalising behaviour they enjoy.
Profilers distinguish the modus operandi, which may vary because it depends upon non-essential matters like availability of weapons and access, fromsignatures, which reveal psychological needs (e.g., ritual aspects of the crime, notes with a nickname, taking of trophies, posing of the corpse and inserting objects into orifices). Signatures are unique and stable (Keppel & Birnes, 2003) hence better for linkage analysis (determining whether the same individual is responsible for a series of crimes).
The signature of the Yorkshire Ripper was a deep wound to the stomach of his victims inflicted with a screw driver after hitting them in the head with a hammer. Peter Sutcliffe was a devout Catholic who claimed to be inspired by the voice of God. However, his detestation for women, especially prostitutes, and the sexual excitement derived from inflicting these deep wounds was probably the true driving force (Burn, 1984). His fetish can be traced to his fascination with a Victorian “educational” exhibit in a Morecambe waxwork museum showing a series of female torsos with windows in their belly illustrating “the nine stages of pregnancy”. According to Sutcliffe’s brother, Peter would linger at this gruesome exhibit with a strange grin on his face. It seems it was this hole in the belly that he sought to replicate with his screwdriver when he attacked a woman. He was caught because, when stopped by police in the Leeds red-light district of Leeds, he tried to hide a tool-kit containing a hammer and screwdriver.
Various types of serial killer have been suggested. The FBI distinguish organised and disorganised types (Turvey, 1999; Canter et al, 2004).Organised killers are said to be above average in IQ. They plan their crimes carefully, usually abducting victims, killing them in one place and disposing of them in another. They operate more by daytime, engage victims in conversation and lure them with cunning ploys. (Ted Bundy put his arm in a fake plaster cast and asked women to help carry something to his car.) They control the crime scene carefully so as to leave few clues, have a knowledge of forensics and follow their case in the media. They are more likely to be socially adequate, with friends, lovers, wife and children.
Disorganised killers have below average IQ, are socially inadequate and likely to live alone, near the location of their crimes. They are impulsive; using whatever weapon is to hand and make no attempt to hide the body. They operate at night, depersonalise the victim and leave a chaotic crime scene. They have minimal interest in the media or police work and often have a history of mental illness.
Holmes and Holmes (1998) propose five subdivisions:
1. Visionary – psychotics acting on the voice of God or the Devil. Killing is a “job to be done”.
2. Mission – compelled to rid world of some undesirable group (e.g. prostitutes).
3. Hedonistic (Lust) - seeking sexual gratification based on a fantasy that involves torture and mutilation (may include necrophilia and cannibalism).
4. Thrill-seekers – excited by the pain or terror of their victims. Seek the adrenalin rush of hunting and killing. Lose interest after victim is dead.
5. Power/Control – need to exert total domination over victim. Like to extend the process for a long time.
Both classifications are arbitrary, with little empirical support and considerable overlap. An alternative approach is that of Canter (2011) who notes that comparisons in the form of “people who bought this also bought ….” may be applied in the field of crime. Canter and Young’s (2009) produced a Smallest Space Analysis of the actions of serial killers based on their co-occurrence. This revealed four major facets labelled mutilation, execution, plunder and sexual control. This Radex Model, which may assist in linkage analysis, features concentric circles of depravity, with common features at the centre and the most extreme activities (e.g. disembowelling and dismemberment) furthest out.
A typical serial killer is Steve Wright, who was convicted of killing 5 prostitutes in Ipswich in 2006. He had no hatred of prostitutes (he used them all his life) but killed them probably because he found strangling women empowering and sexually exciting. Prostitutes were targeted simply because they were vulnerable and already sexualised. Some of his corpses seemed to be posed on their back in a crucifix position, but this was not a consistent feature (Harrison & Wilson, 2008).
Wright’s parents split when he was young but his childhood was otherwise unremarkable. He grew up as a shy young man with difficulty holding down relationships and jobs and a proneness to gambling, drinking, petty theft, debt and suicide attempts. He never settled for long in any one place, moving around various locations in East Anglia and Kent and working variously as a dock worker, lorry driver, barman, pub landlord, fork-lift truck driver and ship’s steward.
Wright was 48 by the time he was arrested but could well have killed earlier. A number of unsolved murders around Norwich, where Wright had lived at the time, bear his trademark (prostitutes abducted, strangled and left in water-filled ditches). These go back to at least 1992, when Wright was 34 - a more usual time for a serial killer to commence a career. The sudden flurry of murders that led to his capture smacks of an “endgame”.
Although Wright was a typical serial killer, many don’t conform to the template. Some target gay men and are usually homosexual themselves (e.g., Dennis Nilsen, Jeffrey Dahmer). Some (like Delroy Grant, The Night Stalker) target older women. This seems strange unless one supposes they are gerontophilic, or resorts to a psychoanalytic interpretation such as “revenge on the mother”. Harold Shipman, the Yorkshire doctor who “euthanised” about 218 elderly women in West Yorkshire apparently witnessed his mother being “put to sleep” when he was 17, and seemed to be revisiting this scenario over and over. Had it become eroticised or was he striving to come to terms with it? Profit was also involved; he was caught because he adjusted the will of one of his victims in his favour.
Female serial killers are not unknown (Vronsky, 2007). One of Britain’s earliest cases was Mary Ann Cotton, a Durham woman who poisoned 21 people around the 1860s/70s including her mother, three husbands, ten of her own children, five stepchildren, a lover and her best friend (Wilson, 2012). This typifies the female serial killer in that they tend to use poison, kill those close to them and profit is the motive. Categories include: black widows (killing husbands for inheritance/insurance) and angels of death (killing helpless people in their care). When women engage in lust killing it usually in tandem with a male partner as a kind of folie a deux (Myra Hindley; Rose West).
Can profiling help to locate a killer? Geoprofiling was first used to locate the wells responsible for a historic (1854) outbreak of cholera in Soho. Today, it gives clues as to the likely place in which a serial criminal lives, with computer generated “hotspots” (e.g., David Canter’s Dragnet). Although approaches differ, two key principles are: (1) Distance Decay (most crimes occur close to where the perpetrator lives), and (2) the Buffer Zone(there may be an area very close to home that is avoided to reduce detection risk). Because the buffer zone may not be adopted initially, early offenses may be more diagnostic than later ones. Canter (2005) believes the killer can usually be found living within the triangle formed by his first three strikes.
Certain geographic features induce Psychological Barriersthat might need to be taken account of (e.g., the Whitechapel road may have been important divider for Jack the Ripper, separating his doorstep from his hunting ground). Two prime suspects (Kosminski and Tumblety) lived at a fulcrum equidistant from most of the murder locations but on the “safe” side of Whitechapel Road. For Levi Bellfield, the M4 might have been a similar divider.
A distinction has been made between Marauders (who operate within the zone where they live) and Commuters (who travel away from home to commit their crimes). Geoprofiling works better with marauder types but it may be difficult to tell in advance which type a criminal belongs to. Paulsen (2007) reported that marauders operated over a larger area, with an average dispersion of 25 miles against 10 for commuters. There was also a longer interval between successive crimes for marauders (61 days against 18 for commuters). Some, like the Night Stalker,seem to live on the edge of their crime scenes but favour a particular direction of travel.
Geoprofiling also works with body disposal sites. A serial killer may betray his home location by depositing corpses a standard distance from his own doorstep in a radial pattern (Lundrigan and Canter, 2001). In an attempt to deflect attention from his home location the killer is actually giving it away. However, in other cases (e.g. Fred and Rosemary West) the bodies are disposed of at home, so there is no danger of being caught in transportation.
A celebrated case in which offender profiling was helpful to police was that of the Railway Rapist. In the 1980s a series of 24 sexual assaults and three murders occurred near to railways in North London. David Canter drew up a profile based on all the information available to him, such as what had been said at the time, and the developing location and nature of the offences. This included:
1. Lives in the area circumscribed by the first three cases.
2. Probably lives with a woman, perhaps without children.
3. Mid to late 20s, right-handed, blood type A.
4. Semi-skilled job with weekend work.
5. Knowledge of the railway system.
6. Criminal record involving violence.
When John Duffy was arrested in 1988 he turned out to live in Kilburn, was separated from his wife, in his late 20s, right-handed, blood type A, was a travelling carpenter for British Rail and was known to the police for having raped his wife at gunpoint. He had been 1505th on a list of suspects but was immediately promoted on the basis of this profile and soon apprehended (Canter, 2005). Duffy liked to tie up his wife before sex, had Kung-Fu weapons at home and had collected 33 door keys as souvenirs from his victims (all the sorts of things Cracker might have guessed). However, none of these fascinating tidbits were discovered until after his arrest; it was the more practical details that led to his capture.
It may be useful to ask how the criminal travels. The self-styled Mardi Gra Bomber targeted Barclays Banks around London in the 1990s in an attempt at extortion. The pattern of his attacks seemed to be in the vicinity of underground stations on lines running north-west from Chiswick (e.g., Ealing, Ruislip), so it was supposed that he used the tube and lived near Chiswick. After this theory was publicised, he switched attacks to Sainsbury’s stores in the exact opposite direction (e.g., Sydenham and Chislehurst). This may have been intended to obfuscate his home base; instead it suggested they were wrong about his mode of transport, since there was no tube coverage in the South-East. Edgar Pearce (aged 61) was eventually caught being driven by his brother and he indeed lived in Chiswick.
Profiling has its limitations. Certain background details are said to be common in psychopaths (e.g., bed-wetting, fire-setting and animal cruelty) but these are widespread in the community, whereas serial killers are rare. Childhood abuse and neglect may lead to heinous crimes but equally motivate others to rise above their difficulties and develop a brilliant career (c.f., Dickens & Chaplin). The majority of murders are uncomplicated, usually domestic and committed by people known to their victims, with straightforward motives like jealousy or greed. The Sutcliffes, Wests and Bellfields get much more press attention and we can easily gain the impression that murders are the work of “evil monsters”. Insanity in murderers is exaggerated because psychiatrists are more often called upon to examine extreme and bizarre cases (Dietz, 1986; Fox & Levin, 2005).
The fashionability of psychological profiling has led to certain miscarriages of justice where individuals seem to have been targeted by police simply because they fitted a preconceived profile of the likely killer. Colin Stagg was charged with the Wimbledon Common murder mainly because he lived nearby, was on the common that morning and seemed a bit of an oddball. After a “profile” of the killer was publicised on Crimewatch, several locals phoned in to suggest Stagg and from that moment he was firmly in the frame. He had no previous convictions but the police charged him with obscenity after he admitted having sunbathed nude (thus enabling the tabloids to describe him as a sex offender). After spending a year in prison awaiting trial, he was quickly cleared on the direction of the judge because of insufficient real evidence. The actual killer was later established through DNA evidence.
Another case is that of Barry George, the fantasist loner who was convicted of killing TV presenter Jill Dando, whom he lived close to in Fulham. It was said that he stalked women, was obsessed with celebrity and played with guns. According to the tabloids, he kept news-clippings about Jill Dando but the truth was he had a stack of old newspapers, some of which referred to her (the articles had not been clipped out or highlighted in any way). Forensic evidence was thin and he was convicted largely on the basis of his “profile”, even though no psychologist had formally tendered one.
After several years in prison, George’s conviction was overturned on appeal. The case was never solved but many now believe she was killed by a Serbian hit man in revenge for the NATO bombing of a State TV station in Belgrade three days previously. Miss Dando had made an appeal on behalf of Kosovan refugees some weeks earlier which would have angered the Milosevic regime and the method of her assassination was strikingly similar to that used to kill other Serbian dissidents (intercepted on arrival home, forced to the ground and “executed” with a single bullet to the head).
The lesson is that while psychological profiling may reduce the size of the haystack in which the needle is sought (the Yorkshire Ripper enquiry had to process 268,000 named suspects) it should not be taken as evidence in itself. DNA profiling is proper evidence on which to base a conviction; psychological profiling is not. Nor can profiling identify a killer before he kills (c.f., the “precrime unit” in the film Minority Report). It is rather in the position of seismology - very good at explaining earthquakes after they have occurred, much less impressive at predicting them.
Acknowledgement: Thanks are due to Professor David Canter for his helpful comments on an earlier draft of this talk and his permission to use his Radex diagram.
© Professor Glenn D Wilson 2012
Burn, G. (1984) Somebody’s Husband, Somebody’s Son. Heinemann, London.
Canter, D.V. (2005) Mapping Murder. Virgin Books, London.
Canter, D.V. (2011) Resolving the offender “profiling equations” and the emergence of investigative psychology. Current Directions in Psychological Science, 20, 5-10.
Canter, D.V. et al (2004) The organised/disorganised typologies of serial murder: Myth or model? Psychology, Public Policy and Law, 10, 293-320.
Canter, D.V. & Youngs, D. (2009) Investigative Psychology: Offender Profiling and the Analysis of Criminal Action. Wiley, Chichester.
Dietz, P.E. (1986) Mass, serial and sensational homicides. Bulletin of NY Academy of Medicine, 62, 477-491.
Harrison, P. & Wilson, D. (2008) Hunting Evil: Inside the Ipswich Serial Murders. Sphere, London.
Holmes, R.M. & Holmes, S.T. (1998) Serial Murder (2nd Edition). Sage: Thousand Oaks, Ca.
Keppel, R.D. & Birnes, W.L. (2003) The Psychology of Serial Killer Investigations: The Grisly Business. Academic Press, New York.
Levin & Fox, J.A. (2005) Extreme Killing: Understanding Serial and Mass Murder. Sage, New York.
Lundrigan, S. & Canter, D. (2001) A multivariate analysis of serial murderers’ disposal site location choice. Journal of Environmental Psychology, 21, 423-432.
Morton, R.J. (2011) Serial murder. FBI Statistics.
Paulsen. D. (2007) Improving geographic profiling through commuter/marauder prediction. Police Practice and Research, 8, 347-357.
Turvey, B. (1999) Criminal Profiling: An Introduction to Behavioural Evidence Analysis. Academic Press, New York.
Vronsky, P. (2007) Female Serial Killers. Berkeley, Penguin.
Wilson, D. (2012) Murder Grew on Her: On the Trail of Mary Ann Cotton. (Excerpted Mail on Sunday, 5/2/12).