Articles

Classification of Firesetters: Who Dominates?

A paper by Trevor Pillinger - Salamander Consultants

Abstract

The number of arson attacks throughout the world is on the increase with the greatest rise being in motor vehicle fires. A great deal of work has been carried out by researchers in attempting to establish a classification system for firesetters in a bid to identify firesetting behaviour and to create a "profile" of an arsonist. This research has been dominated by the two strongest criminological perspectives - Positivism and Classicism with little input from any of the other criminological schools. This paper will examine the influence of these two mainstream perspectives in an attempt to gain an insight of who dominates and in what areas.

Introduction

Fire seems to symbolise life and death. It transforms matter in a way that no other form of energy can. Fire has always been associated with the sun and so peoples from as diverse areas as Egypt, Japan and the Indonesian Islands have kept it alive and used it sacrificially. Fire has always been an intrinsic part of myth, religion, folklore, magic and mysticism. No wonder it is such a part of our psyche. Staring into the flames of an open fire is often hypnotic. It holds for mankind a fascination and fear like no other element. Those who find delight in flickering flames or the glow of hot coals are by no means abnormal but the difference between those that are mesmerised by "good" flames and the fire setter (arsonist) is only a matter of degree, not of kind. It must also be borne in mind that "arson" is the specific crime (criminal offence) whereas "firesetting" is the behaviour of the individual and the motivation behind lighting those fires.

Arson, both in Australia and throughout the western world is a growing concern. Depending on what figures are quoted, arson in Australian alone costs many millions of dollars each year. Typologies, classification systems, matrixes and profiles have been devised by researchers in attempts to fit arsonists into "boxes" - all to make the identification of firesetters and arson crimes easier.

The hypothesis for this paper is that the classification of firesetting behaviour is dominated by two criminological schools of thought - Positivism and Classicism. It will be further argued that practitioners in the investigation (not research) process are strongly influenced by classical theory. In support of this hypothesis, the author will analyse and critique the literature on classifications of firesetting behaviour which appears to be dominated by these two approaches.

The author will first examine the statistical data regarding the prevalence of arson both in Australia and overseas, then distinguish between fireplay (juveniles) and firesetting, analyse and critique the differences between the two predominant criminological approaches in place and look at how these concepts influence the differences between theoretical research and the pragmatic investigation process. The paper will also examine the interesting classification of the serial firesetter which appears to be linked in motive to other serial offences such as rape and murder.

Most research into arson and firesetters has been carried out around either a positivist or classical theory framework with very little work done in the socio-economic or sociological areas. Fire setters (for the most part) are divided into either mad or bad with little regard for those other factors that influence society and individuals.

In Victoria, under Section 197.1 (a) of the Crimes Act, the charge for arson results when a person destroys or damages property fire. The charge would be that:

"the defendant at [..place..] on [..date..] intentionally and without lawful excuse did (destroy/damage) property namely [..property..] belonging to )*[..owner..]/{himself/herself} and [..joint owner..] and valued at [..$..] by setting fire to the said [..property ..] and did thereby commit arson.

Where a death occurs, Section 197 (b) also adds "and thereby causes the death of another".

This paper will use the term firesetting interchangeably with arsonist. Both are terms used for those who set fires with a specific purpose.

Prevalence

Throughout the western world there has been a marked increase in the incidence of reported arson. Regardless of where statistics are cited, the problem of arson as a crime is rapidly becoming a major issue for law enforcement bodies and fire investigators. If we look at just motor vehicle fires as an example, Potter's research shows that in England motor vehicle arson had increased by 8.3% between 1986 and 1996. Here in South Australia, the rate has more than doubled in the same period with a corresponding increase of estimated value of vehicles stolen and burnt from $1.317, 932 (1995) to $3,385,820 (1999). He estimates that if the trend continues, this will increase from 3.8% (1999) to 8% of stolen vehicles recovered by the year 2006.

Arson costs the community millions of dollars, costs lives and injures indiscriminately and causes incalculable misery and disruption. In England, arsonists light approximately 3,600 fires at a cost of around $AUD101 million, two deaths and 60 casualties each week, an increase of some 80% between 1991 and 2001. In Australia, Doley states that "every hour of every day at least one arson fire is lit" with a cost of around $157 million every year just in direct property loss . Other figures suggest that the cost is greater, with the cost of fire in general being in the region of $600 million per year with arson accounting for approximately 30% of that figure and, with the total costs for maintaining fire services and associated structures the figures may grow to as much as four times that.

One of the major difficulties with the statistical data for arson reporting is that at most, it is "estimated". Whilst quantitative research allows us to gauge trends (and all indications are that it is increasing) it tells us nothing of the individual. From a pragmatic view point, a fire is not "suspicious" or arson until declared so by an investigator who has examined the scene and looked at all the evidence. Except in the most obvious cases, it is not until all accidental causes have been looked at and discarded can the investigator be satisfied that there is cause for further exploration and analysis. Where the investigator misjudges the fire cause and origin or where the firesetter is sufficiently skilled to "get away with it", the fire will not be recorded as an arson. Whilst this percentage may be small, it could skew the actual reported figures and thereby lead to underreporting as with any other crime. We have no idea of how many small fires are not investigated and/or reported to insurance companies that may be as a result of arson. This is particularly the case with juvenile firesetters where parents are afraid that they may lose their insurance claim or have a child prosecuted.

Fireplay

There needs to be established that there is a difference between "fire setting" and "fire play"- two very different terms. Nearly all children have a fascination with fire. For most children, this fascination is outgrown at a fairly early age but where the fascination and experimentation extends into adolescence, the "fireplay" is indicative of a more deep rooted problem. Fireplay is a term that is used (generally) for children who are curious about fire whereas, firesetting is the deliberate and willful setting of fires. Most juvenile fire awareness/intervention programs are based around the concept that children play with fire up to (around) age eight years, but that after this age, fires are deliberately set with more than just experimentation in mind.

Kafry's et al research reveals that it is not unusual for children to have an interest in fire and that this can start at around age three and whilst often there is no long lasting effects, many children and adults die or are injured each year due to children playing with matches or lighters. In Dallas, Texas, from 1991-1998, 76 children were injured in residential fires (39 deaths, 37 non-fatal) and in the whole of America, the National Fire Protection Association (NFPA) estimated that in 1998, fires set by children and juveniles resulted in 6,215 American deaths, another 30,800 injuries, and two billion dollars in property damage.

As indicated previously, the incidence of juvenile firesetting is a much underreported area. Parents are naturally, afraid their children will get "labelled" as arsonists. As Slavkin identified "It would appear that African-American youth are at a heightened risk of being labelled deviant or delinquent firesetters than to be labelled in need of emotional support or assistance". In the authors experience it was not unusual for parents to deny a child's involvement in fires because they were either afraid of them being labelled, prosecuted or a denial of an insurance claim. Even though juvenile fire awareness programs are designed around a non-interventionist approach, once a child has "gone on record" it is difficult to minimise the effect of involvement of counsellors and mentors. Both the fire services and insurance companies regard juvenile fireplay, as being "accidental" but this in turn exacerbates the inherent lack of reporting and inclusion in statistics, as it is still a deliberately lit fire.

Classification of firesetters

Positivist typologies

When studying the positivist typologies of firesetters it is appropriate to look initially at perhaps the most publicised type of firesetter - the pathological firesetter or pyromaniac. The concept of a person who lights fires and then gains sexual gratification from watching the flames makes for good media and fascinating reading for those who have a
voyeuristic interest in deviant behaviour. From a research viewpoint, it also makes for an interesting area of study particularly for those researchers who approach the issue from a psychodynamic perspective.

Pyromania is classified as a psychiatric disorder in DSM-IV (1998) - not as a specific classification, but under the category of "Impulse control disorders not elsewhere classified" in other words, a sub-branch of those disorders that do not quite fit neatly into any other category.

Freud connected firesetting with bedwetting (enuresis) and suggested that to light fires was a form of symbolic masturbation and later wrote "the warmth radiated by fire evokes the same sort of glow as accompanies the state of sexual excitation, and the form and motion of the flame suggests the phallus in action. Yarnell's 1940 study on "Firesetting in Children " followed Freud's line, in that fire for youth, represented an attempt to control adult authority and according to Slavkin, little research has been done to research the area to look at the problem from anything other than a psychodynamic perspective.

The 1951 study by Lewis and Yarnell has in the past, been considered the definitive study in firesetters. Their analyse of 1500 firesetters was the most comprehensive study undertaken at the time, with their research examining the records of arsonists collected from the National Board of Fire Underwriters, psychiatric institutions, and clinics. Their research was designed initially to investigate the incidence of pyromania and as part of this they classified firesetters into five main groups. They classified all these as people who set fires for "mental reasons" and that no "normal person" lights fires habitually16 Geller studied psychiatric patients in 1992 and divided the arsonists into four major categories: arson unassimilated with psychobiologic disorders, arson associated with mental disorders, arson associated with medical or neurological disorders and a specific group of juvenile firesetters.

A more recent study by Barker (1994) classified arsonists into four separate types of psychological motivated groups; acquisitive; instrumental; vindictive; cathartic and "no obvious motive'. This is an interesting classification in that the acquisitive group also includes those who set fires for some form of profit (not necessarily financial) motive.

A number of other studies have been conducted by those coming from a psychodynamic perspective in an attempt to define the causes and enable some prediction of firesetting behaviour including 17 in the United States, England (13), Finland (10), Germany (4), Japan (2) France (1) and Australian (1) between 1992 and 2000 and according to Doley, many of these were as result of Lewis and Yarnell's research being published and, to further entrench the research into a positivist framework, each study was conducted using subjects within the forensic psychiatric domain. It would appear that once an initial question was raised regarding firesetting as being a "mental problem" there was a rush to find psychodynamic explanations for motivation - mostly using data those within mental institutions or from those requiring psychiatric assessment prior to punishment or treatment.

Doley cites a number of examples of overseas researchers who all carried out research using data obtained from information forensic areas including hospitals and prisons using the DSM-III-R diagnostic criteria to establish and diagnose psychiatric and behavioural problems associated with firesetting behaviour.

Leong and Silva's research samples came from "court-ordered psychiatric evaluations of criminal defendants charged with either arson or aggravated arson from the central Ohio (US) area over a five year period".

Ritchie and Huff studied records of case files of arsonists from the FBI Behavioural Science Unit and conducted prisoner interviews, examined the records of Maryland's forensic hospital and the Disciplinary Barracks at Fort Leavenworth and, whilst acknowledging that their findings had some limitations, they found that "the demographics [of arsonists] would include being male, unmarried, unemployed, and with a history of criminal behaviour, psychiatric disease and/or substance abuse". They also summarised their findings in an atypical positivist fashion in that in nearly all cases, the psychodynamic research identified a number of behavioural, psychiatric or medical problems in those who were studied. This would be no surprise given the cohort being examined.

All the above research indicates the varied and to some degree confusing results obtained in studying those with diagnosed mental disorders.

Classifying firesetting as a mental disorder relieves the offender of any responsibility for their actions and places treatment clearly in the hands of the medical profession. As Doley states, "it alleviates the need to accept responsibility for our behaviour and provides a ready-made excuse for any damage we may cause. One of the problems with the positivist approach to arson being a mental disorder, is that it tends to overlook motives such as profit - purely and simply a crime committed to make money and to defraud an insurance company with no underlying pathology, apart from greed or the desire to gain financially.

Classical taxonomies

Whilst the positivist approach appears to be quite confusing with so many different models, the classical approach seems to be quite clear cut. In fact, so clear cut as to be simplistic. It does not appear to recognise any underlying pathology (apart from that of the pyromaniac) and neatly fits firesetters into clear taxonomies.

The Federal Bureau of Intelligence and Department of Alcohol Tobacco and Firearms (herein after referred to as FBI and ATF) and the National Centre for the Analysis of Violent Crime (NCAVC) carried out research and produced a classification system which distinguished six different motives 1) vandalism; 2) excitement; 3) revenge; 4) crime-concealment; 5) profit and 6) extremist. All six had sub -categories and all can apply either singly or in combination. The FBI and John DeHaan also break these into the multiple offences of 1) serial arson - where several fires (more than three) are set with a cooling off period in between, 2) spree fires - where a number of fires are lit without a cooling off period and 3) mass arsons where a number of fires are lit at the one place at the one time. DeHaan classifies the six FBI motives as "rational motives", that is, where the firesetter gains some specific benefit from the fire and therefore chooses to carry out the act for gain.

This type of taxonomy was also found in the work done by Canter and Almorid for the United Kingdom's Arson Control forum although they did concede that typologies do have inherent problems in that "assigning individuals to a few "types" is likely to be crude and that any classification can only be approximate.

Wood in his master's thesis research confirmed the FBI findings and derived several "new" motives including one of "despondency" and also included a specific category of "mental illness". One advantage of this approach was that for the first time, links between the crime scene and the offender were being considered rather than just the pathology of the offender. Cantor and Fritzon showed a relationship between the offender and the crime scene and derived a four-behavioural themed approach by including the "choice of target" model, in that fires were lit according to the target being either "object" or "person" and motive as "instrumental" or "expressive".

The classical approach has been adopted by most law enforcement and government bodies and in non-academic literature is by far the most cited taxonomy. This in itself may be a problem. There is a natural assumption by the public that government and quasi-government bodies such as the National Fire Protection Association (USA), Federal Emergency Management Agency (USA) Fire Protection Association (Australia), Arson Control Forum (UK) and, as a consequence, most of the fire services throughout the world "know best". Where one perspective is taken as the only "truth" naturally it will dominate.

Serial firesetters/profiling

Serial offences have gained a good deal of publicity over the last few years. The media is to a large degree, responsible many of the misconceptions that are promoted both in vision media and in print. Considering that the largest percentage of arson fires are lit by non-serial (by definition) offenders, it is surprising how influential their role has been not only in the media but also academic literature.

By developing an empirical model specifically for the study of serial arson, Kocsis and Cooksey have conducted (what they state as) Australia's only research into serial firesetters. They highlighted that previous studies into this area were flawed with a number of anomalies including categories that "did not allow for the discrimination of a specific pattern of offence from behaviours that may simply be typical of the crime or for linking of motives and categories of behaviour between events. As a result of their study, another motive classification system has been developed. Whilst there appeared to be a grouping of common behaviours, four other motives were identified. Their research also highlighted some flaws with the FBI's organised/disorganised dichotomy for serial offences, in that it may be theoretically limited and that a better construct could be a conceptual continuum - that planning (organisation) may well depend on the motive and in turn the type of target. For example, a series of arsons that may on first examination to be erratic but when examined in the light of all targets, may be a well organised thrill firesetter with clearly defined behaviours. Their researchers identified a number of interesting physiognomical factors including for the thrill seeker firesetter, "a tendency to have poor dental work and some outstanding physical feature such a s scarring"; anger firesetters (in their sample) "foreign nationals with noticeable accents" and with sexual firesetters a "tendency to have dark hair and eyes".

It is evident that the research findings lean towards a rational choice theory approach in that they found that with the thrill seekers, they "are older individuals who are socially competent in their lives and cognitively aware of their actions yet engage in a high degree of risk to commit the offence; with the anger group, once a target is decided upon, the only motivation is harm and they leave once they have lit the fires. The sexual classification differs from the previous research on serial offenders in that, according to Kocsis and Cooksey, while these firesetters gain emotional relief from lighting fires, "it is not typically explicit and consequently may not equate to sexual perversion" and they tend to only light small fires in easily accessible areas.

Whilst the FBI profile essentially follows the classical school perspective, it is most interesting to note Kocsis and Cooksey's attention to the offender's physical features and nationality. One would suggest that Lombrosian concepts would have difficulty finding credence in today's research literature. No doubt there was some logic behind using physical features as a variable but to suggest that poor dental work may be an indication of a proclivity to light fires is perhaps questionable. While the FBI profiling has provided a link between the violent crime scene and the offender in some instances, there have been occasions when a focus on a "criminal profile" has been counterproductive to investigations37 and when the profiles have been incorrect.

Conclusion

With all the research cited in this paper, there lies the inherent problem of studying firesetting behaviour from any perspective - sample bias. By virtue of the fact that each of the subjects had been "caught" for their firesetting behaviour, the data precludes those who have, for what ever reason, avoided detection. Self-report is a poor indication of actual behaviour, statistical data on previous conviction will only bring to light those offences where the criminal has been convicted for even though there may be strong suspicion; the western world still relies upon the presumption of innocence. The information provided can only be relevant to the study population. Whilst statistical data on the prevalence of arson can give an indication of trends and costs, it has been shown that the hidden cost of arson must be good deal higher than is reported.

From a pragmatic viewpoint the FBI classification matrix is used in most fire/arson investigation practitioner's manuals as the guide to arsonists and arson motives. The National Fire Protection Association sets the standard for fire, explosion and arson investigators throughout the world and uses this model as the template for investigations as does John DeHaan's investigation text, Kirk's Fire Investigation Manual. The end result of this is that whilst the positivist school holds sway in academic literature, at the coal face, cause and origin investigators are very much influenced by and are, to some degree compelled to utilise the classical approach.

The two perspectives by their very nature focus upon very narrow views of the firesetter behaviour problem. Whilst classicism focuses upon a rational choice approach without taking into account (in most cases) any underlying issues the positivist school appears to over-focus on behavioural issues from a defective individual with a disturbed mind and psychological/psychiatric conditions. There still lays the difficulty of how to gain relevant information about a crime when in practice, you can only research those that have actually been caught. It is recommended that further research be conducted on socio-economic and socio-geographic factors that may lead to arson (particularly juvenile arson) being committed. Another area that requires further research is how academic research can be made more accessible to practioners in the investigative area. Currently, it is highly unlikely that most fire cause and origin investigators have ether the inclination or the ability to access and interpret academic literature.

Four Questions

1. Is there a prevalence of arson particularly amongst socially disadvantaged groups or in lower socio-economic areas?

2. What is the best method to prevent recidivist arson?

3. How do we prevent injury and death amongst young children who engage in fireplay?

4. How do we provide a cross-over between the academic theorists and the investigation practioners?

Trevor Pillinger http://salamanderconsultants.com
Managing Director, Salamander Consultants