Serial Crimes in Australia - UWS ResearchDirect
Transcription
Serial Crimes in Australia - UWS ResearchDirect
Serial Crimes in Australia Investigative Issues and Practice Natalie Scerra Submitted in complete fulfilment of the requirements of the degree of Doctor of Philosophy Social Justice Social Change Research Centre College of Arts University of Western Sydney 2009 © Natalie Scerra 2009 Acknowledgements I must start with a warm thanks to my supervisor, Associate Professor Meg Smith. Thank you a lso t o Dr M arie S egrave and Dr M ichael K ennedy. Michael, your a ssistance w ith NSW Police and other contacts have been worth its weight in gold. I must also thank the numerous academics who have been a source of knowledge and advice to me throughout my candidature. Their wisdom and knowledge has been invaluable. A special thank you to Dr James Arvanitakis whose advice has been invaluable and provided the light when in very dark places. Thank you t o N SW P olice f or e nabling m e t o do t his r esearch a nd giving m e access to their knowledge. A special thank you to Dr Chris Devery of NSW Police, who worked ha rd a t ope ning door s f or m e w ithin t he organisation. A lso, a bi g t hank you t o Georgia Rennie from NSW Police Archives, who gave every assistance possible. I must a lso t hank t hose that pa rticipated i n t his r esearch. T heir contribution ha s been invaluable. Thank you for sharing your knowledge and your experiences with me. I have valued your passion for your work and the victims you have worked so hard for. To my family and friends, and fellow research candidates, a big, warm, thank you. Your s upport a nd pa tience ha s be en v alued a nd a ppreciated. I would not be t his s ane without i t. T he e ncouragement t o he lp m e t hrough a ll t he obs tacles a long t he w ay ha s been wonderful and was i nvaluable t o m e. A s pecial t hank you t o m y n ephew, J ames, whose concern and support has been felt throughout. Thank you for the special hugs and all the little things you did to make me smile. Statement of Authentication The w ork pr esented i n t his t hesis i s, t o t he be st of m y know ledge a nd belief, or iginal except as acknow ledged i n the t ext. I he reby d eclare t hat I ha ve not s ubmitted this material, either in full or in part, for a degree at this or any other institution. ……………………………………………….. Natalie Scerra TABLE OF CONTENTS TABLE OF CONTENTS I LIST OF TABLES VI ABSTRACT 1 INTRODUCTION: SERIAL CRIME- AN OVERVIEW 3 Pieces of the Puzzle- Identifying Serial Crimes 6 What Constitutes Serial Crime? 6 The Serial Murderer 6 The Female Serial Murderer 16 The Serial Rapist 18 The Incidence of Serial Crime in Australia 20 Serial Murder 20 Serial Rape 24 Rarity and the Investigation: Implications 26 What Makes A Serial Killer? 27 Why is this research into the investigation of serial murder and serial rape important? 30 Research Aims and Questions 31 Chapter Overviews 34 CHAPTER 1: CULTURAL CONSTRUCTIONS OF POLICING Policing the Serial: Cultural Influences on the Investigation Dictionary Knowledge Directory Knowledge Recipe Knowledge Axiomatic Knowledge Habitus of Policing Work One Part of the Whole - Police Subcultures What’s Mine is Mine - Ownership of Information Case Theories and the ‘Art’ and ‘Craft’ of Policing Investigating Stereotypes A Language of Their Own Who Could Be Bothered? Results of Accountability in Practice Discretionary powers (Over) Accountability - i- 39 40 44 45 46 47 47 50 52 54 58 60 62 63 65 Victims of Violent Serial Crime A Reflection of the Offender Rape Myths The Invisible Without a Trace- The Missing The Worthless Chapter Summary 66 67 68 73 76 81 84 CHAPTER 2: INVESTIGATING SERIAL CRIME Traditional Investigative Techniques and the Serial Problem Scientific and Technological Advances Media as an Investigative Tool Resources Intelligence-led Policing in the Context of Serial Investigations Using Intelligence-led Policing in the Investigation of Serial Offences Police Subcultures and Intelligence Gathering Joining the Dots: Crime Linkage The Picture of Evil - Serial Offender Typologies and Criminal Profiling Is There a Typical Serial Offender? Organised versus Disorganised Criminal Profiling Geographic Profiling Chapter Summary CHAPTER 3: METHODOLOGY 86 88 89 95 101 103 106 108 112 116 116 117 118 122 124 126 Research Question Research Aims Methodology Selection of Cases Stage 1: Case Studies Sample Sources of Information Documents Records Availability of Sources Case Study Analysis Methodological Issues: Case Studies Stage 2: Interviews with NSW Police Detectives and Professionals Interview Method Construction of Interview Schedule Participants Recruitment Sample - ii- 126 127 128 130 131 133 138 138 141 144 145 146 149 151 152 153 153 155 Confidentiality and Consent Interview Transcription Analysis of Case Study and Interview Data Methodological Issues: Interviews Ethical Considerations Research Limitations CHAPTER 4: CASE STUDIES – OVERVIEW AND TIMELINE OF EVENTS 156 157 158 159 159 160 164 165 175 184 Introductions…Serial Murderers Introductions…Serial Rapists Victimological Representations CHAPTER 5: RESULTS- EMERGENT THEMES Thematic Analysis of Case Studies and Interview Data Construction of an Investigation Cultural Constructions of Policing Intelligent Investigations and Culture CHAPTER 6: ISSUES IN THE INVESTIGATION OF SERIAL CRIME Pictorial of the Investigation A Resourceful Investigation: The Task Force Task Force Air- Milat Task Force Pulsar- Denyer North Shore Murders Task Force- Glover Task Force Chart- Snowtown Murders Task Force Ardmore- Johnson Strike Force Grahame- Trindall Strike Force Sapelo- Daley Critique of Task Forces Investigative Personnel Information Management Using the Media in Serial Investigations Investigative Techniques and Technologies Utilised Other investigative techniques Pressures on Police Chapter Summary CHAPTER 7: THE POLICING PERSPECTIVE- CULTURE & PRACTICE Reflecting Subcultures Competing Subcultures 185 185 186 188 191 194 195 195 196 199 200 201 201 202 203 204 207 213 216 230 235 239 244 245 246 246 - iii- Detective Experience Playing the Hunch Dictionary Knowledge and the Conceptualisation of Victims The Missing Dictionary Knowledge and Serial Crime Use of Language to Understand Investigations Chapter Summary 248 250 256 263 266 267 269 CHAPTER 8: THE CULTURE OF INTELLIGENT INVESTIGATIONS 270 Crime Linkage Timeframes of Recognition Time from Second Offence to Recognition of Crimes as Serial Time Until Arrest from Recognition of Series of Crimes Recognition of Crimes as Serial The Paper Trail Criminal Profiling Use of Typologies Modus Operandi & Signature Chapter Summary 270 270 271 274 276 286 288 298 299 302 CHAPTER 9: A NEW TYPOLOGY- VIOLENT SERIAL CRIME INVESTIGATIONS 304 The Serendipitous Investigation The Conspicuous Investigation The Delayed-Recognition Investigation RECOMMENDATIONS 305 307 310 316 Recommendations for the Investigation of Serial Crimes Recommendations for Further Research 316 318 CONCLUSION 320 REFERENCES 324 APPENDIX A 345 UWS HUMAN RESEARCH ETHICS COMMITTEE PROTOCOL APPLICATION FORM Interview Schedule- Police Interview Schedule- Forensic Technicians/Consultants Information Sheet for Participants - iv- 346 346 357 359 361 Ethics Approval- University of Western Sydney APPENDIX B 364 365 Example of Preliminary Thematic Cluster APPENDIX C 366 367 Letter to NSW Police Requesting Access to Closed Case Files Research Proposal to NSW Police Research Approval- New South Wales Police - v- 387 388 364 LIST OF TABLES Table 1: Incidence of Serial Murder in Australia from 1900......................................... 22 Table 2: Convicted Serial Rapists in Australia .............................................................. 25 Table 3: Selected Incidents of Serial Murder for Case Study Analysis ....................... 135 Table 4: Selected Incidents of Serial Rape for Case Study Analysis .......................... 137 Table 5: Sources of Information Used in Case Studies ............................................... 144 Table 6: Denyer- Timeline of Victimological Events.................................................. 166 Table 7: Folbigg- Timeline of Victimological Events ................................................. 168 Table 8: Glover- Timeline of Victimological Events .................................................. 170 Table 9: Milat- Timeline of Victimological Events ..................................................... 172 Table 10: Snowtown- Timeline of Victimological Events .......................................... 174 Table 11: Daley- Timeline of Victimological Events .................................................. 176 Table 12: Johnson- Timeline of Victimological Events .............................................. 178 Table 13: Presta- Timeline of Victimological Events.................................................. 181 Table 14: Trindall- Timeline of Victimological Events .............................................. 183 - vi- Abstract Violent serial crimes, in particular those of murder and rape, raise a number of issues for i nvestigators. This r esearch c ontributes t o t he bod y o f knowledge regarding t he practices of investigating violent serial crime by considering the contribution of a range of t echniques and t echnologies t o t he i nvestigative out come. This r esearch also examines how predominant pol ice c ultural und erstandings a nd know ledge of t hese crimes and the vi ctims of s uch crimes that i nform s uch i nvestigations will be considered. Additionally, the roles of various investigative techniques and technologies in a chieving pos itive a nd a cceptable i nvestigative out comes, i n t he form of an a rrest, are ex amined. Specifically, this r esearch focuses on the incorporation of i ntelligenceled policing ini tiatives such as c rime linkag e s ystems into t he i nvestigation of s erial crimes was also a focus of this current research. To address t hese ai ms and associated issues in the A ustralian context, two methodologies are employed: case studies and semi-structured interviews. Case studies of s erial crimes in Australia are e xamined i n c onjunction w ith i nterviews of N SW Police detectives and consultants that have investigated these crimes. Several factors were identified as impacting on the investigation of violent serial crimes. One ke y factor w as the ina bility o f Australian police or ganisations to effectively incorporate intelligence-led policing initiatives which enable the linkage of related crimes into the investigation of serial crimes, leading to delayed outcomes in the - 1- recognition a nd l inkage of r elated c rimes. T he w ay i n w hich pol ice ‘dictionary knowledge’ can be s een t o i nfluence t he w ay i n w hich i nvestigations a re c onstructed was al so a s ignificant factor ide ntified by the t hematic ana lysis. Another s ignificant factor that emerged from the research was the role of the media in police investigations. Based on the research findings, a typology of the investigation of violent serial crimes ha s be en c onstructed. T his t ypology reflects t he a bility of i nvestigators t o recognise and link related offences, and the influence that such recognition has on t he investigative p ractices e mployed and on achieving a pos itive i nvestigative out come. There a re thr ee c ategories in this t ypology: the s erendipitous inve stigation; the conspicuous investigation; and the delayed-recognition investigation. The factors identifies by this research are demonstrated to have an impact on the investigation of vi olent serial cr imes. Specifically they contributed t o t he delay i n t he recognition of these offences as serial in nature and in achieving a positive investigative outcome. A s i ndicated by t he pr oposed typology a nd recommendations, these f actors can be addressed by police organisations to improve the investigative capabilities of the police investigating these violent crimes. - 2- INTRODUCTION: SERIAL CRIME- AN OVERVIEW Serial crimes, particularly serial murder, have been the subject of academic research for several decades. M uch o f t his r esearch ha s focused on the of fenders o f t hese cr imes. This has been precipitated by the endeavour to understand the aetiology of such crimes. Studies have attempted to explain these crimes by proposing psychological, social and biological models f or u nderstanding of fenders’ be haviour (for ex ample Douglas & Olshaker, 1997; Leyton, 1986; P into & W ilson, 1990; S tone, 2001) . These as pects however, have minimal impact on the ability of police to investigate these serial crimes. This c hapter will di scuss s ome of t he c ommon unde rstandings of s erial c rime a s developed in previous research. The contributions and limitations of this knowledge to the inve stigation of s erial c rime will be a ddressed. Preceding thi s, the i ncidence of serial crimes in Australia will be presented. When di scussing s erial c rimes t here a re s ome m isconceptions a bout w hat crimes can be considered as serial. Academic researchers and the general public’s broad familiarity with serial m urder and serial r ape m ean that ot her crimes which can be considered a s s erial i n na ture a re of ten ove rlooked: f or ex ample, serial ar son which refers to acts of r epetitive f ire s etting (Mavromatis, 2000) , serial s talking (LloydGoldstein, 2000), and serial burglary (Schlesinger, 2000a). Forms of non-violent serial crimes inc lude e xhibitionism, frotteurism 1, and t elephone s catologia 2 (Krueger & 1 2 The sexualised bodily rubbing against another Obscene phone calls for gratification - 3- Kaplan, 2000 ). M any of t hese forms o f s erial c rime a re not widely acknowledged o r brought t o t he a ttention of t he g eneral publ ic. One r eason for t his i s t hat t he publ ic experience less heightened feelings of fear about these forms of serial crime than serial murder and serial r ape (Mavromatis, 2000) . A nother pos sible r eason f or t he l ack of general knowledge about these crimes is that they rarely receive the heightened media attention that the crimes of serial murder and serial rape would. Additionally, some of these crimes are difficult to identify and detect. Thus recognition of a s eries of related offences is minimised. This research focuses on the crimes of serial murder and serial rape because of the strong public responses that these crimes elicit and the widespread media attention they g ain. This at tention comes de spite, and because of , the r arity of t hese of fences, with few incidents of serial murder and serial rape occurring in Australian and abroad in comparison to other crimes (Pinto & Wilson, 1990). Carney (2004) contends that the repetitive na ture o f crimes s uch as s erial r ape c an produce an atmosphere of fear t hat affects a c ommunity i n a w ay t hat i s uni que c ompared t o t he i mpact of ot her c rimes. This fear can be seen to stem from the popular knowledge of these predatory offenders: these fo rms of s erial cr ime ha ve be en the focus of a num ber of popul ar crime genres and the general public has generally gained such knowledge about the characteristics of such crimes in films and literature (Fox, Levin, & Q uinet, 2005; J enkins, 1994; Simpson, 2000) . In a ddition, t he r eportage of s uch c rimes i n t he m edia i s of ten extensive a nd s ensationalist. T he general familiarity with these f orms o f s erial c rime - 4- may le ad to extra pr essure be ing a pplied on t he investigators by t he general publ ic when attempting to apprehend a violent serial offender. Once a series of these crimes has been identified, the nature of the investigation into those crimes shifts. That is, there is a sharing of resources in the investigation of each related offence w hereby i nformation relating t o each individual of fence i s considered i n t he c ontext of t he s eries of c rimes. T his i s i n c ontrast t o t raditional investigations into single-incident offences where information gathered is considered in isolation. T he r elatively uni que na ture of s uch i nvestigations i s a f actor dr iving t he focus of the current research. This cha pter will ex amine t he i ncidence of s erial m urder and serial r ape t hat have occurred in Australia and demonstrate that these crimes’ rarity has contributed to a gap in knowledge about how such crimes are investigated as well as how they differ to incidents of s ingle-victim of fences. This c hapter w ill pr ovide a context w ithin the research literature that exists to demonstrate that much of the previous research into this field provides an i nadequate c ontribution t o our unde rstanding of how violent s erial crimes are policed. - 5- Pieces of the Puzzle- Identifying Serial Crimes What Constitutes Serial Crime? Currently in Australia t here is no criminal of fence o f s erial mur der o r serial rape. T he t erms ‘ serial murder’ and ‘serial r ape’ ar e not l egal d efinitions and thus do not carry offence categories that can identify serial offenders as such. Instead, offenders are charged with each individual offence committed against each victim. For example, in NSW Ivan Milat was charged and convicted of seven counts of murder and one of attempted murder. Such terms however, are utilised by academics and practitioners in the field, including police and detectives, to describe a certain type of offender or how a number of crimes are related. The Serial Murderer The t erm ‘ serial m urderer’ or ‘ serial ki ller’ w as c oined by FBI a gent and c riminal profiler, Robert R essler, t o de scribe repeated ac ts of hom icide t hat occur i n a series (Ressler & S chactman, 1992). This w as done t o di fferentiate t he s erial of fender f rom the m ass m urderer w hich t he f ormer ha d pr eviously be en c lassified a s; this nom inal difference also highlighted key differences between the crimes such as the timeline of offences. Mass murders usually occur as a single event whereas serial murder denotes a - 6- number of separate criminal events (Fox, et al., 2005), and thus is repetitive in nature. The repetition, as described by Ressler, “leaves the murderer hanging”, in anticipation, presumably a waiting t he ne xt m urder i n t he s eries (Ressler & S chactman, 1992) . In Ressler’s c onception the te rm ‘serial m urder’ discounts the ne ed f or a minimum number of related offences, provided that they appear repetitive in nature. There i s no one d efinition of s erial m urder t hat has be en a greed upon b y t he majority of researchers in the criminal justice field. This is due to the variety of factors that help de fine s erial murder. T he m ain factors t o be c onsidered a re: t he num ber of victims, the victim-offender relationship, t he t ime s equence of events, t he m otivational forces, and the geographical nature of these events. These factors also help to distinguish incidents of serial murder from single-victim crimes. There also appears to be a conf lict in defining female s erial m urderers i n contrast t o male s erial m urderers. Currently female s erial m urderers can only b e de fined as s erial of fenders, unde r t he br oadest of definitions. In serial murder definitions, there are a minimum number of victims that must be achieved before such murders can be classified as serial in nature. According to Egger (1998), Goldsworthy (2002), Pinto and Wilson (1990) and Turvey (2002a) a minimum of t wo s imilar m urders m ust oc cur be fore t he m urders c an be c onsidered a s s erial murders. This s imilarity ma y b e in any f orm, for i nstance comparable vi ctims or ke y features of how the actual crimes were committed. However Hagan (1992) states that a minimum of three s eparate i ncidents m ust oc cur be fore i t can be cons idered as s erial - 7- murder, a de finition s upported b y H olmes a nd H olmes ( 1998). C onversely, F ox a nd Levin ( 1998) s tate t hat f our or m ore m urders m ust have oc curred before s uch a definition is employed. Egger ( 1998) r efutes t he ne ed f or s uch l arge n umbers, s tating t hat any t wo similar murders should be at least investigated as serial murder until proven otherwise. A minimum nu mber of tw o victims a lso allows f or le niency in the investigator’s imprecise knowledge of, and ability to link, similar murders, and the stage of the serial murderer’s career. For example, law enforcement personnel may not be aware of other similar murders particularly if they fall outside their division. The serial murderer may have only begun his or her criminal career and therefore there may be no more than two murders committed at that point in time. Egger (1998) states that by waiting until three or f our pe ople ha ve been murdered in similar ci rcumstances, law enf orcement personnel a re pl aced at a di sadvantage as their i nvestigative c apabilities are reduced due to the period of time that has lapsed before they are able to investigate the offences as l inked crimes. Information from pot ential witnesses f or e xample ma y not be available or be inaccurate if a prolonged period of time has passed since the crime took place. A definition that includes a minimum of two murders also allows for a deficiency in the police recognition of related offences, as previous related offences may not have been identified initially. Definitions of s erial m urder c an a lso i nclude pr escriptive guidelines r egarding the relationship between the victims and t he o ffenders. M ouzos (2000) and P into and - 8- Wilson ( 1990) identify that the re is us ually little or no prior personal relationship between the victims and the offender. This is consistent with findings from studies by Fox, Levin and Quinet (2005), who state that the typical victims of serial murderers in the United States of America are prostitutes. However, if this were to be accepted as a definition, men a nd w omen w ho m urder a num ber of pa rtners ove r a p eriod of t ime would not be classified as serial killers: neither would those in medical professions who murder a num ber of p atients, s uch as a nurse. Although i n m any known i nternational cases of s erial m urder, the vi ctim i s c hosen r andomly a nd t hus i s unknown t o t he offender (Egger, 1998; Kidd, 2000) there are occasions in which a relationship with the serial murderer can make a pe rson the target. This is particularly evident in the case of female serial mur derers, as r esearch has s hown t hat t hey frequently ta rget f amily members, f riends a nd acquaintances ( Hale & B olin, 1998) . H owever, as m any definitions of s erial m urder s uch a s t hose pr oposed b y E gger (1998), F ox, Levin a nd Quinet (2005) and P into a nd W ilson (1990) are de rived from s tudies of m ale s erial murderers, this factor is often ignored. Many researchers identify specific motivations that must exist in order for these crimes t o be cl assified as s erial m urders. In a review of s erial m urder i n Australia, Goldsworthy (2002) s tates t hat t he primary motivation of s erial murder is usually the emotional or sexual gratification derived from the act. Egger (1998) claims that serial murderers a re m otivated b y a ne ed f or pow er and c ontrol ove r t heir vi ctims. T hese crimes ar e us ually cons idered to be predatory, i n t hat t he offender’s be haviour of ten includes t he stalking of victims a nd features sexual or s adistic acts (Mouzos, 2000) . - 9- According t o t he Diagnostic a nd Statistical M anual IV-TR ( American Psychiatric Association, 2000, p. 573) which categorises and defines ps ychiatric disorders, sexual sadism: … involves a cts (real, no t s imulated) i n w hich t he i ndividual de rives s exual excitement from t he p sychological o r phy sical s uffering ( including humiliation) of the victim. It is also thought tha t these c rimes a re us ually considered to have u nclear motives (Pinto & Wilson, 1990) or psychological motives (Mouzos, 2000). According to S chleisenger ( 2000b) t he m otivation of serial m urderers a nd s ome ot her serial offenders, relates to the sexually aggressive fantasies entertained by the offender and is driven by the offender’s compulsion to carry out this fantasy which leads to an act such as murder. However, according to these definitions, multiple murders which are motivated by financial or revenge factors, cannot be considered as serial murderers. It is generally accepted that the motives for a m ajority of m ale serial murderers are usually sexually and/or ps ychologically motivated: the of fender aims t o gain sexual pl easure or po wer over the victim (Holmes, Hickey, & Holmes, 1998). Many studies into the phenomenon of s erial m urder f ocus upon s exually m otivated c rimes (Kocsis & Irwin, 1998; McKenzie, 1995) . K ocsis a nd Irwin (1998) outline s ix ps ychological ch aracteristics which they be lieve d efine s erial m urder in the sample s tudied. These p sychological - 10- characteristics are psychopathy, narcissism, sadism, paraphiliac tendencies, and fantasy and compulsiveness. Female s erial m urderers ar e of ten motivated by what H olmes, Hickey and Holmes (1998) refer to as ‘purchases of comfort’. This includes money, insurance and business m otivators. H owever, a ccording t o H olmes e t al. ( 1998) s ome female s erial killers are motivated b y sex, relationship issues or b y revenge. It is important to note that a study by Hickey (1991 in Holmes et al., 1998) found that one in two female serial murderers worked with a male accomplice. In these cases, their motivation may be one of loyalty to their male accomplice rather than other personal factors. The time sequence of the murders is one factor that is usually a greed upon by the majority of theorists as a key identifying factor in the definition of serial murder. It is generally accepted that if the crimes occur over a period of time- hours, days, weeks, months or years- then t hey c an be c onsidered s erial i n na ture (Fox, e t a l., 2005) . However problems arise when the murders are committed within hours of each other as this may indicate that a m ass or spree murderer is at work (Fox, et al., 2005), although other features of the crime can usually help distinguish between these forms of multiple murder. These features include the victim type, method of approach and if a signature is present. In many cases of serial murder, if more than one murder occurs within hours or days of each other, then it can indicate an escalation in the offender’s behaviour. For a series of m urders w hich oc cur ov er hou rs or d ays t o be c onsidered a s s erial m urder, - 11- there m ust be a di stinguishable br eak between the of fences in w hich t he of fender i s considered to have performed ‘normal’ duties such as work (Cook & Hinman, 1999b). Another factor often used to define serial murder is the geographic features of the crimes. Theorists such as Egger (1990b) and Hickey (1997) agree that serial murder occurs ov er a geographical di stance, whether t his relates to the pl ace the vi ctim is selected or where the vi ctim’s remains are disposed. Hickey (1997) also outlines how the offender’s mobility may also be used to typify serial murderers. Hickey claims that there are three typologies that are based upon where, geographically, offenders commit their c rimes. The f irst i s the “ travelling s erial killer”, relating to those w ho commit multiple m urders ove r a w ide geographical range, w hether t his be b etween ci ties, regions or s tates. The s econd category i s t he “local s erial ki ller” w ho may c ommit his/her crimes within a particular city or town. The final category is the “place-specific serial killer” who commits his/her crimes in a specific place such as their home or place of work. The geographical features of serial murder have been the subject of wide debate with t heorists dr awing on va rious e xamples f rom t he U S and Australia. For ex ample Egger (1990b) states t hat s erial m urderers ar e us ually ve ry m obile and are capable of moving be tween s tates. This c ontinuity of m ovement i s a lso i dentified b y Egger a s a feature of m ost know n s erial m urderers. E gger d oes how ever a cknowledge t hat s ome serial murderers operate in a very small geographical area. An Australian case example of thi s limite d geographical mobi lity is J ohn Glover w ho committed his c rimes in a - 12- relatively small area on Sydney’s North Shore. Other theorists such as Canter, Missen and Hodge ( 1996) claim t hat s erial of fenders rarely t ravel l ong di stances and thus usually operate in one particular area. Leyton (1986) has pr oposed t hat w ith i ncreased ur banisation i n de veloped countries such as the United States and Germany, there appears to be a d ecrease in the number of travelling serial killers and an increase in the other two geographical types of serial ki llers. In a s tudy i nto t he di stance t ravelled by s erial m urderers in Germany, Snook, C ullen, M okros a nd H arbort (2005) found t hat 63 pe rcent of of fenders l ived within ten ki lometres of t he l ocation of t heir c rimes. T hus such r esearch s upports the notions proposed by Egger (1990b) and Canter et al. (1996) that many serial offenders commit c rimes r elatively close to their hom es. T his m ay be due t o t he i ncrease i n population in ur ban a reas, which allows f or s erial m urderers t o remain undetected (Egger, 1990b ). T hey are a lso l ess l ikely t o be d etected i n hi ghly popul ated a reas, a s their vi ctims ar e l ess l ikely t o be di scovered. Although there ar e s everal de bates regarding the m obility o f s erial of fenders, t he i nclusion o f t his f actor i s not s een as aiding the identification of a series of related crimes as serial although it may contribute to investigators’ inability to link related crimes (Snook, et al., 2005). With s o m any f actors t o be c onsidered i n t he de finition of s erial m urder, a nd dispute over the relevance of each amongst academics and police and law enforcement agencies, there is no universal definition of serial murder. According to Mouzos (2000) - 13- Australian police agencies accept the Federal Bureau of Investigation’s (FBI) definition which is identified by Cook and Hinman. This definition states that: …there a re 2 or more separate homicide events which oc cur over a period of time…. There is a c ooling-off pe riod b etween episodes. These c rimes a re predatory. The offender frequently s talks hi s v ictims. The m otive is c learly psychological. The o ffender’s b ehaviour an d the cr ime scene ev idence typically i ndicate sexual a nd sadistic features, and may involve torture and mutilation to the victim (Cook & Hinman, 1999b, p. 364). This de finition of a ‘ typical’ or ‘c lassic’ serial murderer w as c onsidered as the most applicable to this research as it is the operational definition used by police organisations in Australia as identified by Mouzos (2000). As previously stated Egger (1998) believes that a definition that in cludes a minimum of tw o related a cts can lead to increased investigative e fficiency. Such a d efinition excludes s everal s eemingly ‘ serial’ c rimes. An example of such cases would include a domestic dispute where several people were murdered over a short period of time consistent with mass murders or cases regarding professional ‘hit-men’ or military-related murders, where cr imes a re com mitted under the guise of a military operation. Although these cases may be labelled in the media as ‘serial mu rder’, there i s a ge neral c onsensus among academics and law enf orcement that t hese c rimes s hould be e xcluded, pr imarily du e t o t he di fferences i n the motivational f orces dr iving the se c rimes f rom those of s erial mur derers ( Pinto & Wilson, 1990). - 14- Several factors affect the applicability of this definition to some cases of serial murder. These include the characteristics of the crime and the offender, which are often referred t o as t ypologies. Typologies are o ften ut ilised to describe a nd ultimately understand the offender, thus aiding the investigation through informing by potentially identifying distinguishing behaviours. In the discussion of investigative tools in Chapter 2, the typologies of the serial murderer will be identified and critiqued. Also excluded f rom t he de finition of a s erial m urderer a re m ass m urderers, when a number of victims are murdered during the one episode (Fox, et al., 2005). Fox et al. (2005) in their discussion of forms of homicide, state that the mass murderer tends to target friends, family, co-workers, students and teachers. These offenders often plan their crimes and are motivated by avenging the injustices that they perceive have been perpetrated against them. These are usually single incident events, or events that occur as pa rt of t he one e pisode w ith no c ooling-off pe riod, w ith t he of fender either committing suicide or being detained by police at the conclusion of the offences (Fox, et al., 2005). One factor that is not identified in definitions of serial murder is the number of offenders. S everal c ases of s erial m urder i nvolving t wo of fenders e merged from t he literature, including Leonard Lake and Charles Ng in California, and Kenneth Bianchi and A ngelo Buono i n Washington, a nd C anadian c ouple P aul B ernardo a nd Karla Homolka (Fox, et al., 2005). - 15- Some definitions of serial crime also require a recognisable modus operandi and s ignature f or c ategorising a s eries of m urders a s s erial m urder ( Davies, 1992; Killmier, 1997) . T he m odus ope randi (MO) can be de fined as t he m ethod of ki lling, whereas the signature is defined as the extraneous actions that were performed during the c ommission of t he murders ( Killmier, 1997) . F or e xample a s ignature of pos tmortem mutilation may vary in that the level and intensity of this act may increase but the key identifying feature of the signature will still remain. As stated by Keppel (2000) although a signature may evolve over time, the core feature of the signature will remain the same. The MO, also referred to as the offender’s ‘calling card’ (Keppel, 2000), in contrast may change over time, either as the serial offender evolves or as a response to law enforcement’s investigations (Keppel & Birnes, 2003). The Female Serial Murderer The proposed definitions of t he s erial m urderer a s previously di scussed a re l imited in that often the female s erial m urderer is not r ecognised a s s uch, a s t he d efinitions are often gendered in nature. Holmes, Hickey and Holmes (1998) state that this may be due to the resistance to the idea of women as violent offenders, rather than the caring and nurturing stereotype t hat cul tural ex pectations are a ccustomed t o. The l ack of recognition may also be due to the extremely rare nature of this type of criminal. Of 47 cases of serial murder in the United States identified by Holmes and DeBurger (1988, cited in Holmes et al., 1998) only three offenders were female. This lack of recognition - 16- as s erial of fenders how ever, m ay a lso b e due t o t he w ay i n w hich t hey commit t heir crimes. This r epresentation how ever, i s i n s tark contrast t o t he i ncidence o f serial murder committed by women in Australia. According to Paul Kidd in his discussion on Australian incidents of violent crime, “…of the 33 cases of serial murder in Australia since 1822, A ustralia has had 10 f emale s erial ki llers; roughly a 30 p er cent average” f(Kidd, 2005, p. 135). This indicates that the incidence of serial murder committed by female offenders may be higher than that acknowledged by the research literature. According to Hale and Bolin (1998) female serial killers differ greatly to their male c ounterparts. O ne of t he m ain di fferences i s t he m ethods t hat w omen us e t o commit the ir c rimes. In their s tudy of 184 female s erial ki llers in the U nited States, Hale a nd B olin i dentify poison a s t he m ost c ommon method employed. As H ale and Bolin a mongst ot hers (for e xample H olmes, H ickey, & Holmes, 1998) acknowledge, the victims of the female serial killer are usually family members or acquaintances, thus allowing for greater opportunities to commit such acts using poison as their weapon. In Australia, the i ncidence of f emale-perpetrated serial c rime is s ignificantly lower tha n that committed by their male counterparts. This pr esence of a p re-existing r elationship be tween vi ctim a nd the female offender w ould p reclude the inc lusion of th ese crimes in the c ategory of ‘ serial’ if traditional de finitional pa rameters w ere followed, s uch as t hose de signated b y t he - 17- definition of s erial mur der ut ilised i n t his t hesis. A s pr eviously s tated, i t i s of ten considered that a d efining f eature o f a s erial c rime is tha t the re is littl e or no prior relationship be tween t he vi ctim a nd t he of fender ( Mouzos, 2000; P into & W ilson, 1990). As found by Hale and Bolin (1998) there is a propensity for the female murderer to target family members, therefore traditional beliefs about what constitutes ‘serial’ are questionable. In a ddition, be liefs a bout how a serial c rime s hould be d efined w here women are not recognised as possible serial offenders, may impact on an investigation. The Serial Rapist The other violent serial crime of interest to this present research is that of serial rape. A definition of serial rape raises several problems as there have been few, if any, attempts by academics to provide a formal definition, and the term is often treated by researchers as t hough it is se lf-explanatory. Further di fficulties a rise w hen considering t he t erm ‘rape’ w hich doe s not ha ve a l egal de finition i n m any states but r ather i s a s ocially constructed term. The term ‘rape’ has been defined as: the penetration of the mouth, vagina or anus by any parts of the attacker’s body or by an object used by the attacker, without the consent of the victim (Easteal, 1992, p. 1). - 18- However, the use of the label ‘serial rapist’ to describe a s erial offender who commits such acts, has been overwhelming both in academic literature, both internationally (for example Money, 1990; Turvey, 2002c; J. I. Warren, et al., 1998) and in Australia (for example Kocsis, 2002). The m ethodology us ed in t his t hesis c onsiders incidences o f s erial r ape t hat have o ccurred i n N ew South W ales onl y. In New S outh W ales, t here i s no legal definition for rape. Rather, such offences are referred to as sexual assault, where “an y person who has sexual intercourse with another person without the consent of the other person a nd w ho know s t hat t he o ther pe rson doe s not c onsent to t he s exual intercourse…” ("Crimes Act 1900 ( NSW)," Section 61I). The definition for serial rape is however, more difficult to operationalise, although Turvey (2002c) states simply that serial rape refers to two or more related rape or sexual assault offences. In simple t erms, serial r ape can be cons idered as r epeated r ape offences (Turvey, 2002c). Thus for the purposes of this research, the term ‘serial rape’ will refer to all s erial of fences w here s exual pe netration is com mitted. The t erm ‘ serial s exual assault’ will refer to cases where penetration was not known to have occurred. This is in line with the legal definitions of the state of New South Wales, as outlined above, as the case studies of serial rape were limited to this region. In both instances, these crimes have largely been considered in the literature to be committed by offenders with little or no pr ior r elationship t o t he vi ctim. T his i s s upported by s ome a cademic s ources w ho - 19- use the term ‘serial rapists’, to usually refer to predatory rapists who victimise strangers (Deu & Edelmann, 1997; Stevens, 1994). Thus rapists who repetitively offend against someone known to them- such as a girlfriend- or those who habitually rape numerous sexual partners over a period of time, are rarely cons idered to be serial r apists. A n e xample of t his i s e vident i n a s tudy conducted by Kocsis, Cooksey and Irwin (2002), in which they excluded multiple rape cases where the offender had a prior relationship with a victim or had raped a victim on a num ber o f oc casions. As w ith serial murder, t he us e of t he vi ctim-offender relationship c an be c onsidered pr oblematic. In i ncidences of s erial r ape, s uch limitations in recognising repeat offenders as ‘serial’ can be considered to affirm myths of the ideal rape victim. This will be discussed in greater lengths in Chapter 1. The Incidence of Serial Crime in Australia Serial Murder According t o t he N ational H omicide M onitoring P rogram ( NHMP) t he num ber of homicide vi ctims in Australia f or th e pe riod of 2006 t o 200 7 w as 266 (Deardon & Jones, 2008). Of these 266 victims only 6 percent were unacquainted with the offender and thus can be considered as stranger killings. However, as identified in several international studies (Fisher, 1997; Jenkins, 1994), the general public’s fear of - 20- being m urdered b y a s tranger is more i ntense than of be ing m urdered by family, intimates or acquaintances which occurs far more frequently. These crimes account for ninety percent of deaths in Australia (Deardon & Jones, 2008). Although t he i ncidence of t hese c rimes at t he h ands of s trangers i s r are, t he phenomenon of the serial offender is rarer still yet is perhaps the most easily recognised due to the productions of popular culture such as films and books (Caputi, 1990). Fox, Levin and Quinet (2005), in their discussion of research considering the incidence and aetiology of s erial m urder, note tha t it i s di fficult to gauge the e xtent of this t ype of crime due t o difficulties in l inking them, and t he t ime and pl ace di sparities be tween these c rimes. These pr oblems a rise from on e of t he ke y d ifferences be tween the majority or homicide or murder incidents, whereby a large proportion of homicides are committed by friends or family members of the victims (Davies & Mouzos, 2007). In Australia since 1900 there have been only 29 recognised incidences of serial murder (Table 1). This number equates to an average of one incident every four years. However, such a f igure is s till pr oblematic in de termining the inc idence of s erial murder as it onl y includes cases where convictions were obtained. It does not include cases w here i t i s s uspected that a s erial m urderer i s r esponsible, for ex ample t he Claremont murders in Western Australia and the murders and disappearances of several women in the Newcastle region in New South Wales. - 21- Table 1: Incidence of Serial Murder in Australia from 1900 Date a 1900-06 1907-09 1930-35 1938-39 1942 1948-52 1946-53 1947-53 1961 1961-62 1959-65 1962-90 1973 1974-90 1976-77 1979-80 1984-86 1986 1987 1984-90 1989-92 1993 1993-96 1998 1998 1998 1993-99 1997-99 2002-03 No. of Victims up to 37 3 4 3 3 2 5 4 2 4 6 2 3 3 6 3 3 4 5 6 7 3 3 2 3 3 11 3 2 Offender/s State Alice Mitchell Martha Rendell Arnold Soderman Albert Andrew Moss Edward Leonski Yvonne Fletcher John Balabana Caroline Grills Leonard Lawson William McDonald Eric Cooke Barry Hadlow Archibald McCafferty Rodney Cameron J.Miller & C.Worrell Patricia Moore Michael Laurence D.Birnie & C.Birnie Josef Schwab John Wayne Glover Ivan Milat Paul Charles Denyer Kathleen Folbiggb Mark Valera Matthew James Harris Leonard John Fraserc ‘Snowtown’d Peter Dupase Francis Michael Fahey WA WA VIC NSW VIC NSW SA NSW NSW NSW WA QLD NSW VIC/NSW SA NSW NSW WA NT & WA NSW NSW VIC NSW NSW NSW QLD SA VIC QLD One murder occurred in Paris prior to the offenders arrival in Australia, three victims were result of a spree Folbigg was also found guilty of the manslaughter of a fourth child c Fraser was also found guilty of the manslaughter of a third woman d Four men were convicted of the murders of these eleven victims however not all offenders were involved in the murder of all eleven victims e Dupas had been convicted for a series of rapes and sexual assaults prior to his murder convictions b An issue that has been a topic of debate, is the perceived or actual increase of the incidence of serial murder. It has been argued that there is an increase in the number of cas es of s erial m urder ea ch decade s ince 1900 (Jenkins, 1994) . Serial m urder i s considered a modern phenomenon, although globally incidences have been discovered to have taken place as early as the fifteenth century (Leyton, 1986). This view of serial - 22- murder as a ‘modern’ occurrence can be explained in part as the result of the changes to the approach of law enforcement agencies to this phenomenon. The term ‘serial killer’ was not c oined unt il t he e arly 1980s ; R obert R essler (Ressler & S chactman, 1992) wished to differentiate these crimes from other ‘stranger murders’, as he had recognised that s ome r elevant cases w ere i n fact com mitted by offenders know n to the vi ctims. Cases of serial murder were, up unt il that period, also grouped under the classification of mass murder (Fox & Levin, 1999). One estimate by Jenkins (1994) of serial murder in the United States up until the end of 1992 lists 447 names of serial murderers. Of these, 82 were active between 1900 and 1970, w ith t he r emaining 365 s erial ki llers a ctive f rom 1970 t o 1 992 ( Jenkins, 1994). These figures appear to evidence a staggering increase in the incidence of serial murder. Y et i t i s i mportant t o not e t hat t hese f igures r epresent i ndividual names of serial killers and not separate incidents. F or example where two or more people have worked in cohort to commit the ir c rimes ( team ki llers), the inc ident w ill r eflect the number of killers. Crimes that are considered mass murders, the killing of a number of people in the one episode, are not included as part of this data. McLaughlin a nd M uncie (2001) have r eferred t o the U nited States as t he ‘natural habitat’ for serial murderers. This label is largely reflective of the high number of cases in the United States as evidenced by research such as that by Jenkins (1994). However A ustralia ha s not be en protected f rom this phe nomenon. In A ustralia, t he numbers of recognised cases of serial murder is not as high as those of North American - 23- statistics. Table 1 demonstrates t he num ber o f recognised cases o f s erial m urder i n Australia from the year 1900 onwards. With the exception of Alice Mitchell, who was not convicted of all the murders that she confessed to, the offenders were convicted of the n umber o f m urders specified although t here m ay ha ve b een ot hers attributed t o these offenders. This table also does not include offenders who were not convicted of two or more crimes as these are not considered to be serial in nature, nor does it include cases that remain unsolved or without a conviction being obtained. During the 1980s and the 1990s, the phenomenon of serial murder in Australia appeared t o be m ore c ommon, a s c ompared t o previous de cades ( Table 1) . In t his period alone at l east 13 cases of serial m urder e merged and challenged investigators (Kidd, 2000; M cGarry, 2005; Pinto & W ilson, 1990). The a ctual i ncidence of s erial crime is c onsiderably lo wer tha n other vi olent c rimes (Goldsworthy, 20 02); how ever, serial crimes inherently have high victimisation rates, often of three or more victims per offender, and t hus c an c reate h eightened l evels o f fear i n t he community, particularly when offences are reported in a sensational way by the media. Serial Rape The i ncidence of s erial rape i n Australia also appears t o be an issue o f i ncreasing importance. Since 1990 in Australia there have b een at l east 23 solved cases of serial - 24- rape a s out lined i n T able 2 3. From a comparison of t he i ncidence o f s erial r ape a nd serial murder in Australia over a similar timeframe, it is evident that serial rape poses a greater focus of attention for investigators. This can be seen from a comparison of the number of vi ctims f or each of fence t ype as out lined i n T able 1 a nd T able 2, w here serial rape has a higher victimisation rate than serial murder. Table 2: Convicted Serial Rapists in Australia Date a 1956-59 1970 1979-90 1982-86 1989-90 1979-92 1984-94 1978-88 1995 1988-96 1988-96 1995-96 1992-97 1995-97 1996-97 1994-98 1995-98 1990-99 1990-2001 1997-1998 1998-2000 1999-2000 2000 1998-2002 2002 1995-2004 2004-05 No. Victims 16 5 6 18 5 6 2 7 4 4 4 12 8b 5 6 6 3 7 11 7 3 4 4 2 7 2 Offender State David Joseph Scanlon Raymond Edmundsa Gary Frederick Sutton George Gerrard Kaufman Terry John Williamson Paul Keating Brett Anthony Major Robert John Fardon Troy Allen Burley George Allan Pryor Quoc Vinh To Graham James Kay Lyall Grant Meizer David Glen Johnson Bruno Presta Donald Wayne Garlett Shane Francis Gray Carlos Delgado-Guerra Shane Ronald Farmer Gilbert Atwell Anthony James Daley Jason Charles Buckley Skaf & co.c Thomas John Armfield Abraham Hzaife Wayne Anthony Trindall Afsheen & Azhar Hussein NSW NSW SA VIC NSW SA SA QLD QLD QLD NSW NSW QLD NSW NSW VIC VIC QLD TAS QLD NSW QLD NSW SA VIC NSW QLD aka “Mr Stinky”. Was also convicted of the double murder of two teenagers Johnson was believed to have committed crimes against 29 women, was convicted of crimes against 18, of which convictions for sexual assault of 8 victims are included c Offenders Bilal Skaf, Mohammed Skaf were convicted of charges relating to all four victims, Mahmoud Sansoussi, Mohammed Sansoussi for charges relating to 3 victims and Tayyab Sheikh for charges relating to one victim. b 3 This is not a comprehensive list of all incidents of solved serial rape cases in Australia, but one where convictions for more than one sexual assault incident on more than occasion could be validated through media reports or court documents. - 25- Despite t his hi gher de monstrated incidence, serial r ape r eceives l ess at tention than serial murder in both the academic research arena and in the general community as evidenced by the disparity in the reportage of these cases through the media and other literary avenues such as documentaries and biographical accounts. One reason for this is that there appears to be less media reportage of such crimes. Another reason may be that the le gal pr ocesses a re de signed t o p rotect t he r ights of t he of fender b y not permitting pr evious c onvictions f or r ape a nd s exual a ssault to be br ought i nto a ny current prosecution for sexual offences. Rarity and the Investigation: Implications As de monstrated the o ccurrence of s erial m urder and serial r ape i n Australia i s relatively rare. The incidence is even rarer when considering the occurrence at a l ocal or s tate le vel. Incidents of s erial c rimes a re inves tigated by the pol ice of the s tate in which t hey o ccur. Table 1 and Table 2 indicate t hat cas es of s erial cr ime ha ve predominately t aken place i n New S outh Wales i n the pa st de cade, w hile ot her Australian states have had fewer, partly due to smaller populations in those states. It is believed t hat t he i nfrequent occurrence of s erial murder and s erial r ape contributes t o inexperience of pol ice i n i nvestigating s uch c rimes. It i s pr oposed t hat f ew of t he detectives investigating the serial crime incidents in Australia would have had previous experience in serial crime investigation due to the rarity of these offences. - 26- What Makes A Serial Killer? A number of academic theorists have attempted to explain the aetiology of the violent serial of fender (see f or e g. E gger, 1984, 1998 ; L eyton, 1986) . These ex planations emerge from various fields of enquiry and include the social and environmental factors, psychological dynamics and biological causes. Although there are a number of relevant theories proposing to provide at least a partial explanation for such offenders only some of these will be discussed in detail as it is beyond the scope or intent of this thesis to provide a comprehensive representation of these theories. Many of these theories have been formulated in response to cases of serial murder with little or no acknowledgment of t he c rimes of s erial r ape. Other theories however, do assign special s tatus t o serial crimes of a sexual nature. Egger (1998) proposes a t heory t hat stems f rom the be lief tha t c riminal behaviour is a function of the socialisation process. The offender’s past may explain his or her aggressive behaviour. This is often a favoured explanation of serial murderers’ motivations, as ma ny s erial mur derers ha ve r eported be ing t he vi ctims of c hildhood abuse (Egger, 1998). The ps ychopathology o f t he of fender i s a nother factor of ten c onsidered in t he attempt t o explain the oc currence of serial mu rder (Pinto & W ilson, 1990) . P ublic perception is that all murderers, particularly serial murderers, suffer from some form of mental illness (Caputi, 1999). However, the evidence does not support this. A study by - 27- Mouzos (1999) on the incidence of mental illness in the homicide offender population in Australia, found that only “4.4 percent (147) of homicide offenders were recorded as suffering from a me ntal di sorder at the time of the hom icide inc ident” over t he ni ne year p eriod reviewed b y t he s tudy (Mouzos, 19 99, p. 4) . M ouzos a lso found t hat a mentally disordered offender was more likely to murder a family member, which is in contrast t o the s erial of fender w hereby most serial m urderers us ually t arget pe rsons unknown to themselves (Mouzos, 1999). There is currently no Australian research that identifies an overwhelming presence of mental illness among serial offenders. According to Egger (1998) the term psychopath is redundant, with the accepted name for the condition being ‘antisocial personality’. In his attempt to develop a theory of t he or igin a nd de velopment of t he s erial of fender, Giannangelo (1996) however, contests tha t ‘antisocial’ and ‘psychopathic’ are not i nterchangeable t erms. Giannangelo proposes that psychopathic serial killers tend to be predatory and exhibit aggressive narcissistic characteristics in addition to antisocial behaviour. Regardless of the act ual de finition and differences be tween t hese conditions, pe ople w ith either disorder are not considered a ble t o e xperience emotions s uch a s l ove and e mpathy (Egger, 1998) . H owever, a s E gger s tates, pe ople s uffering f rom t hese pe rsonality disorders are not insane or mentally ill. Because of the ways in which serial murder differs from other forms of murder, particularly in regard to motivation and the progression of the offending career, it is not possible to ascertain if any or all of the above factors contribute to serial murder. Due to - 28- the of ten s exual na ture of t his c rime, i t i s ha rd t o g eneralise, a s t here has not be en enough research in this area. In their discussion of serial murder in Australia, Pinto and Wilson (1990) warn that “s erial ki llers cannot b e ne atly categorised into a pa rticular behavioural t ype” a nd t hat “ documentation s hould oc cur w ithin a f ramework w hich recognises that serial killing is not committed by specific psychological or sociological types” (Pinto & Wilson, 1990, p. 3). Other e xplanations f or t he be haviour o f t he s erial o ffender consider t hat biological precursors can explain behaviour. One biological theory proffered by Fox et al. (2005) identifies he ad t rauma a s be ing l inked w ith vi olent be haviour. T his i s particularly r elevant as m any s erial m urderers have suffered f rom s ome form of childhood abuse, therefore many may have sustained head traumas (Fox, et al., 2005). As demonstrated, there have been many attempts to identify the causes of serial offending, s pecifically s erial m urder. There i s a clear gap i n knowledge in relation t o the literature surrounding serial rape. Explanations are often sought within the existing literature addressing s exual as sault i n general whilst of ten serial r ape can also be considered as a p recursor t o the s erial m urderer (Schlesinger, 20 00b). These explanations however, do not directly contribute to the knowledge of the investigation of s erial cr imes directly. The aet iology only be comes pertinent pos t-conviction once thorough ps ychological and psychiatric testing of the offender can be carried out. The theories of serial criminal offending can be seen to form the basis of some investigative tools s uch a s c riminal profiling. T he know ledge t hat i s dr awn upon t o c onstruct a - 29- criminal profile of the serial offender, often reflects some of the theories of offending. Terms such as ‘psychopath’ are sometimes used in criminal profiling therefore a level of know ledge a nd und erstanding o f t hese c oncepts i s e xpected t o s ome e xtent of investigators. During an i nvestigation, how ever, ot her m ore s alient f actors a re considered to be able to inform the investigative process. These perspectives of s erial c rimes, being the w ay i n w hich t hese cr imes ar e i nvestigated, have r eceived little i n the w ay o f academic a ttention. A s w ill be i dentified, m uch of t he e xisting r esearch f ocuses on particular aspects o f t he i nvestigation, s uch a s t he us e of c riminal pr ofiling a s an investigative tool . This research w ill c onsider a range of i nvestigative t echniques a nd technologies e mployed during t he i nvestigation w hich c ontribute t o t he investigative process as a whole, and which have little dependence on the causes of these crimes. Why is this research into the investigation of serial murder and serial rape important? From the review of existing literature addressing the violent serial crimes of murder and rape, it is evident that there has be en little focus on the investigation of these crimes. While the t heories of of fending c an i nform t he i nvestigation a bout the ge neral background of t he of fender, t hey cannot ha ve a di rect i mpact on t he i nvestigation. Factors that have a di rect affect are the type of investigative tools and techniques that are utilised in the investigation of serial murder and serial rape. The demonstrated rarity - 30- of t hese of fences, a nd t he w ay t hey di ffer f rom t he m ajority of hom icide or r ape offences, mean that there may not be a developed understanding b y police about how these crimes should be investigated. The evi dence conc erning t he i ncidence of s erial m urder and serial r ape al so substantiates the need for a consideration of the investigative processes of these crimes. The appa rent i ncrease i n t he i ncidence o f s erial cr ime i n recent years, coupled with a greater unde rstanding o f t he of fender’s m otives, br ought about b y t heories of s erial offending, have led to more marked attention on the way these crimes are investigated. These investigations take place within an organisational context which is influenced by the or ganisational c ulture w hich w ill be di scussed i n t he f ollowing s ection. It i s proposed that this organisational culture- in this case, police culture- informs the way in which the crimes of serial murder and serial rape are investigated. Research Aims and Questions This research asks why many incidences of serial murder and serial rape go undetected by police o rganisations for l ong p eriods o f t ime. This t hesis p roposes to de termine which factors impact on the investigation of serial murder and serial rape. In order t o ascertain what t hese factors ar e, several i ssues ne ed to be addressed: the us e o f investigative t echniques, i n particular t he us e of offender profiling as an investigative tool; the use of forensic techniques; the availability of investigators to link crimes; and - 31- finally the issues that impact on the ability of investigators to solve serial crimes. Thus, to simplify these issues we can state that this thesis aims at: • determine how serial crimes are recognised and what impact this has on the investigation • identify if recognition of serial crimes can be improved to reduce further victimisation or the period of criminal offending • establish whether police practices, in relation to allocation of resources, impact the investigative outcome • examine how i mportant the characteristics of t he chosen vi ctims are t o the investigative processes and outcome • identify which i nvestigative t echniques a re us ed i n s erial m urder a nd serial r ape cases and t o what ex tent f orensic technologies contribute t o these investigations In or der t o a ddress t hese research aims of t he t hesis, case s tudy research of incidents of serial murder and serial rape that have occurred in Australia and interviews with f ormer a nd current m embers of N SW Police a nd consultants w ere employed. These m ethodologies have enabled this r esearcher t o gain a s pecific und erstanding of the techniques and technologies employed in the investigation of the serial crimes being studied. T he l evel or de gree of i mpact t hat these i ssues ha d on i nvestigations c an be further ascertained through the an alysis o f t he i nterviews. The r esearch data ha s b een analysed within a framework informed by police cultural understandings. - 32- The i nformation obt ained f or e ach c ase s tudy was analysed i n or der t o determine if certain characteristics are present in relation to the offender and the crime, and the references t o investigative pr actices, issues a nd p ractice. Questions, t his researcher aims to be address through the analysis of the case studies, include: • how inve stigators link ed r elated cases: w hether there was a n intelligence-led system in place t o a chieve t his; w hether a m odus ope randi or s ignature were apparent i n t his c ase; a nd ho w linking r elated c ases contributed to t he investigation of these crimes. • which i nvestigative t echnologies and t echniques w ere e mployed i n t he investigation of t he i dentified cases. W hether t hese t echniques or t echnologies were vi ewed a s contributing to the inv estigative out come; tha t is , the identification of the serial murderer/serial rapist. Which issues surrounding the use of these technologies were apparent. There are several questions and issues that the analysis of interviews will aim to address. Some of these will be aimed at providing a comparison to the material gained from the case study r esearch, whilst some will be aimed at filling the gaps present in such research. Thus generally some of the questions and issues to be addressed are: - 33- • how a s eries of related crimes is identified as serial murder or serial rape. How such a n i dentification i mpacts on a n i nvestigation. W hether c hanges t o t he processes or the ways crimes are investigated have taken place • which investigative techniques and technologies are used in the investigation of serial cr imes. Whether any i ssues i n t he a vailability or pr ocessing of s uch evidence e xist. W hether t hese t echnologies c ontributed t o t he out come of t he investigation. • which pol ice pr actices he lped or hi ndered the i nvestigation. H ow t he investigation changed over time, particularly between related crimes? Whether there was a reduction i n r esources dur ing l ong periods of of fender i nactivity, and if so, whether this impacted on the investigation in any way. This the sis a ims to examine the pr ocesses of inve stigations int o serial crime, and t o determine if police cultural practices – positive or negative – influence the construction of a n investigation, particularly int o serial c rimes. The pol ice c ultural pr actices discussed in this chapter will be used as an analytical framework for this research. Chapter Overviews This chapter has presented an overview and discussion of previous research in the area of s erial m urder a nd s erial r ape, t o c ontextualise t he c urrent s tudy within w hat - 34- represents a large proportion of the existing literature addressing these crimes. There is a large body of research in existence on the topic of serial crime, and more specifically serial m urder, ho wever much of t his ha s i nadequately addressed t he i nvestigation of these cr imes. T his c hapter has al so presented an i ntroduction t o t he t opic t hrough a discussion of the incidences of serial murder and serial rape in Australia, demonstrating the need for further research to be conducted in this country. Chapter 2 will propose that an organisational culture of policing as proposed by Janet C han (1996, 1999b) in he r taxonomy of police c ultural know ledge c omprising four components, informs and influences the progress of serial murder and serial rape investigations. Further, an analysis of t he l iterature s urrounding aspects of pol ice culture that can be considered to influence the investigation of violent serial crimes will be presented. Additionally, this chapter will also raise the issue of how victims are perceived and how the understandings of particular groups in society, and subsequently within the organisational context, increase one’s vulnerability to serial offenders. This factor has also been argued to affect the way in which investigations are conducted, whereby the victimology m ay result in some delay in r ecognising certain people as ‘ victims,’ or influence the way in which some of these victims are perceived by police. Chapter 2 will identify t he range of i nvestigative t ools and techniques t hat are commonly associated with the investigation of violent serial crimes of murder and rape. - 35- The issues surrounding the definition and identification of the serial crimes of murder and rape will also be discussed including a critical overview of the factors considered when attempting to identify the occurrence of serial murder and serial rape is provided. As a m ethod o f i nvestigation, the role o f int elligence-led policing p ractices in the investigation of s erial c rimes will be ide ntified. T he di scussion of t he utilisation o f intelligence-led policing practices within the context of police cultural understandings will show that cultural knowledge informs the investigation of serial crimes. Chapter 3 outlines t he methodological approach e mployed b y t his qualitative research. As previously identified, this is a qualitative study employing case studies of serial crime and semi-structured interviews with members of NSW Police, both former and c urrent. A di scussion of s ome of t he methodological pr oblems encountered in conducting the research is identified and discussed, in particular the issues experienced in researching a police organisation. The following chapter presents a profile and ‘victimological’ timeline of events for e ach c ase s tudy. A s ummary of ke y events r elating t o bot h t he cr imes and the investigation were identified in order to later demonstrate some of the research findings in relation to problems arising from the timeline of events. Chapter 5 sets out the dominant themes that emerged from the thematic analysis of the case studies and the semi-structured interviews. It sets these themes up within the context within which they will be discussed. These themes are grouped into three broad - 36- categories w hich will be di scussed i n s ubsequent c hapters: the m ethods e mployed t o investigate s erial cr imes; the cul tural constructions of t he i nvestigation; and t he reflections on the use of intelligence-led policing practices in the investigation of serial crime. Chapter 6 discusses t he t hemes r elating t o t he m ethods e mployed dur ing t he investigation of s erial r ape and s erial m urder offences as i dentified from t he c ase studies a nd s emi s tructured i nterviews. T he c onstruction of a n i nvestigative t eam a nd the al location of r equired r esources a re di scussed i n t his c hapter. A dditionally, t he investigative t echniques and t echnologies found t o be c ommonly u sed in serial c rime investigations a re ide ntified and the the mes emerging r elating to the impa ct of s uch tools are presented. A theme that emerged in relation to the construction and conduct of serial crime investigations was the attribution of blame to investigators. This blame was often a ssociated w ith a pe rceived f law i n t he investigation a nd t his i ssue w ill be discussed. Chapter 7 presents the themes that emerged from the case studies and the semistructured i nterviews i n r elation t o t he c ultural k nowledge of p olice. There w ere t wo dominant t hemes t hat w ill be pr esented. T he f irst i s t he c ompeting s ubcultures w hich are evident within the police investigation. The second theme is that of the reflection of dictionary kno wledge on t he i nvestigation of s erial crimes. This di scussion will focus on the application and influence of cultural knowledge in the form of stereotypes. - 37- The f inal analysis cha pter presents the dom inating the me o f the di fficulties police e xperience in identifying a nd linking a s eries of c rimes, both in r elation to murder and sexual assault. In Chapter 8 the timeframes of serial crime recognition will be presented and discussed as will the issues arising from the use of the Violent Crime Linkage Analysis System. Additionally, the use of criminal profiling as an investigative tool in the investigation of serial crimes will be discussed, particularly in relation to its contribution to the investigative processes and outcomes. Finally, C hapter 8 pr oposes a t ypology d escribing a nd c ategorising serial murder and serial rape investigations. This typology has evolved from the data analysis from the themes relating to the linkage of related crimes and subsequent recognition of the crimes as serial in nature. The r ecommendations arising from t his r esearch w ill be pr esented i n t he Chapter 9. These recommendations ar e t wo-fold. T he f irst are r ecommendations relating to the inv estigation of s erial c rimes in Australian police or ganisations. Secondly, following qu estions r aised i n t his t hesis, a reas f or further re search a re identified. Finally, t he c oncluding arguments w ill be pr esented. T he c onclusion w ill assess the thematic outcomes of this research within the theoretical framework of police culture w ith considerations of t he role o f i ntelligence-led pol icing pr actices on t he investigation of the crimes of serial rape and serial murder. - 38- CHAPTER 1: CULTURAL CONSTRUCTIONS OF POLICING It i s pr oposed t hat a n important a spect of the investigation of s erial crimes is the organisational ‘sense making’ framework within which an investigation is constructed. This sense making framework developed by Sackmann (1992) in the context of police organisations, has been referred to as ‘police culture’ (Chan, 1996). Cultural practices within a police or ganisation c an a ffect t he outcome of inve stigations in to particular crimes as it is us ed to inf orm the inve stigative r ole a nd the a ctions employed by investigators (Innes, 2002). There ha ve be en s ignificant s tudies i nto t he c ultural m ilieu of the New S outh Wales Police Force. Most notable are the studies by Janet Chan (1999b, 2003a, 2003b) and David Dixon (1999b). The cultural idiosyncrasies of the police as an organisation have been studied by many including Chan (2003a), Dixon (1999c), Innes (2003) and Van Maanen (1978b). Australian researchers C han (2003a) and D ixon’s (1999c) research focus is on police culture in the NSW Police. Janet Chan’s focus included the socialisation of NSW Police recruits in a longitudinal study of the attitudes of recruits to policing. Chan’s (1997) other research includes studies into the policing of minority groups and t he c ultural c onstruction of kno wledge t hat s hapes and i nfluences discriminatory attitudes towards these groups. Chan’s research considers police reforms - 39- undergone du ring and f ollowing t he W ood R oyal C ommission i n N SW 4. Her l ater research also c onsiders t he i mpact of i nformation t echnology on pol icing (Chan, Brereton, Legosz, & D oran, 2001) . D avid D ixon’s (Dixon, 1999a , 19 99b) research examines the regulation of policing in NSW which was also in the shadow of the Wood Royal C ommission’s i nvestigations i nto pol ice c orruption a nd t he s ubsequent incorporation of the new accountabilities brought about as a result of this. According to Chan (2003a) police culture is the term that is applied to describe the ‘ cultural know ledge’ of pol ice or ganisations, s pecifically t he nor ms, va lues a nd attitudes of members of the organisation. Previous research into police culture and its influence on t he i nvestigative process h as be en q uite l imited. The t raditional r esearch into pol ice c ulture ha s not a ddressed t he i mpact of t his c ulture on s erial c rime investigations. Policing the Serial: Cultural Influences on the Investigation The l iterature i n r elation t o pol ice c ulture c an be c onsidered a s f airly broad, addressing a range of issues from the training of cadets, corruption within senior police ranks and the recruitment of pol ice t o w ork w ith s pecific c ommunity groups f rom diverse e thnic a nd cultural ba ckgrounds (Chan, 1996, 1997, 1999a , 1999b, 2003b; Dixon, 1999b, 1999c ). T he di scussion of t he w ork a nd c ulture of de tectives how ever 4 The Wood Royal Commission investigated allegations in relation to, and the extent of, corruption in the NSW Police Service; the impartiality in the investigation and prosecution of a range of criminal activities; and issues relating to the internal promotion systems, and internal commands. See Wood (1997) for full report - 40- has often been restricted to the sphere of corruption and the findings extending from the Wood Royal Commission in NSW (see Dixon, 1999c). In Australia there has been little research i nto pol ice cultural practices i n ot her s tates, although a report b y Chan et al. (2001) analysed t he i mpact of pol ice culture on t he i ntegration o f i nformation technologies i n t he c ontext of Q ueensland pol ice. T here a re s ome general t hemes however t hat emerge from t he general l iterature on police cul ture that ha ve some specific relevance to the i nvestigative pr ocesses of de tectives, particularly s temming from the work of Ericson (1993) and Innes (2002, 2003). Police c ulture is the te rm us ed to describe the or ganisational culture within organisations of policing. According to Schein, the culture of a group or an organisation refers to: A pa ttern of sha red basic assumptions t hat t he g roup l earned as i t sol ved its problems of external adaptation and internal integration, that has worked well enough to be considered valid and, therefore, to be taught to new members as the c orrect w ay t o pe rceive, t hink, a nd feel i n r elation t o t hose p roblems (Schein, 1992, p. 12). The term ‘police culture’ has no universal definition as such and is used to refer to va rious a spects of t he pol ice or ganisation. In g eneral t erms, pol ice c ulture c an be considered t o r efer t o t he m echanisms e mployed b y pol ice t o de al w ith t he pr essures that they face working in a job that can be considered as dangerous and unpredictable (Chan, 1999b) . In hi s di scussion of t he i nfluence o f pr ocess s tructures on - 41- the investigation of hom icide or m urder i n t he United K ingdom, Innes (2002) describes how t he pol ice or ganisation pr ovides m embers w ith pr agmatic, e xperience ba sed knowledge and values which “can be employed in ‘doing’ police work” (Innes, 2002, p. 679). Van Maanen (1978b) in his di scussion of pol ice s ocialisation i n an A merican police de partment, describes t he s eparation of p olice f rom t he m ainstream cul ture as leading to a new collective group. Van Maanen states that: when the policeman dons his uniform, he enters a distinct subculture governed by norms and values designed to manage the strain created by an outsider role in the community (1978b, p. 292). The r ules or characteristics of t his oc cupational culture ar e not eas ily r ecognisable as they are unwritten and informal (Chan, 1999b). As outlined by O’Loughlin and Billing (2000) in their study int o the positive aspects o f police culture in relation to violence and corruption, the term usually conjures up negative images of sinister practices and cover ups . Indeed pol ice c ultural pr actices ha ve been us ed t o pr ovide e xplanations of police deviance and violence (I. Warren & James, 2000). Paoline III however suggests that: Viewing pol ice c ulture as a n oc cupational phe nomena s uggests that o fficers collectively confront situations that arise in the environments of policing, and subsequent a ttitudes, v alues, a nd nor ms t hat r esults are i n r esponse t o those environments.(2003, p.200). - 42- In t he context of m urder i nvestigations, Innes (2002) proposes t hat s uch know ledge relates to how and why such crimes occur and subsequently what investigative methods detectives s hould employ that are mos t e ffective in solving t hese cases. D ue t o t he rarity of serial murder and serial rape, the investigation of serial crime poses different problems f or pol ice i nvestigators. T he know ledge t hat i nforms t he i nvestigation i nto common incidents of single-victim murder or rape cases is not necessarily appropriate for t he i nvestigation of s erial crime. T hus t he pol ice or ganisation c an be s een t o encapsulate t he et hos of i nvestigating m urder ca ses and the as sociated problems t hat may arise and the organisationally appropriate response to these (Innes, 2002). This is c onsistent to some e xtent w ith C han’s (1999b) explanations of pol ice culture, w hich c an be c onsidered a s a m eans of s haring or ganisational knowledge or cultural knowledge. From the perspective of detectives, the combination of formalised training and guidelines coupled w ith t he i nformal va lues and know ledge a bout ho w police work is done, are maintained by this occupational culture (Innes, 2002). Cultural knowledge as a cknowledged b y C han, can be c onsidered a s a s et of a ssumptions consisting of four dimensions as adapted from Sackmann (1992). The four dimensions identified by Sackmann are: • • • • dictionary knowledge, directory knowledge, recipe knowledge, and axiomatic knowledge. - 43- Sackmann investigated how these forms of knowledge were shared by the members of an organisation in the United States. These forms of knowledge were utilised by Chan to elicit an understanding of police cultural knowledge within NSW Police. Dictionary Knowledge Sackmann (1992) describes di ctionary know ledge as be ing d escriptions t hat a re common to a particular organisation. Chan states that dictionary knowledge “provides definitions and labels for persons, things, and events that police encounter in the course of their work” (2003a, p. 31). This knowledge is used by police to develop a routine for assessing pe ople a nd s ituations t o e nable t he t imely j udgement of a s ituation. This dictionary kno wledge i s s imilar t o t he t ype of know ledge Van Maanen (1978b) describes as being p assed on f rom ‘ war s tories’ during t he s ocialisation process. This experience based knowledge is a set of perspectives of people and situations that police will be in contact with that Van Maanen considers to be collectively held. As outlined by Chan (2003a), such knowledge and subsequent judgements can be based on class or race. The Australian Law Reform Commission (1992) produced a report into the impact of the laws and the legal system on those from ethnically diverse communities and from various c ultural ba ckgrounds. A mongst i ts f inding, t his r eport c oncluded that pol ice organisations w ere engaging i n ethnic s tereotyping, p articularly of Aboriginal p eople and their involvement in criminal activity, which is apparent in the knowledge of police (Chan, 1999b, 2003a). - 44- Chan’s discussion of dictionary knowledge however, refers onl y to the way in which pol ice us e t his know ledge t o m ake de cisions on t he c riminality o f a pe rson or group. It is s uggested that t hese s imilar j udgements c an be m ade i n regards t o t he victims of c rime. Such unde rstandings of t he us e o f c ollective unde rstandings or dictionary knowledge, have been limited to its application to street level police. Neither explanation however was discussed by Chan in relation to the use of this knowledge in the investigation of crimes by detectives. Whilst such labels may be used to inform the understandings of s treet level pol ice, t here has b een no publ ished r esearch examining whether they exist or how such labels are used at the level of detectives. The impact of such judgements will be of consideration to this research, as consideration will be given to how such knowledge impacted on the ability of police to investigate serial crimes. Directory Knowledge Dictionary know ledge as out lined a bove i nfluences t o s ome extent, di rectory knowledge, which provides descriptive understandings about the appropriate responses to the situations identified through the use of dictionary knowledge (Sackmann, 1992). Therefore in the context of a police organisation, directory knowledge refers to the way in w hich “ operational w ork i s r outinely c arried out ” o r how t hings a re t o be done (Chan, 2003a , p. 32) . F or pol ice, t he j udgements f ormed f rom t heir di ctionary knowledge pr ovide t hem w ith a g uide t hat enables them t o m ake s nap de cisions a nd follow up with quick action based on these labels. For example youths congregating in - 45- shopping centre areas are targeted with proactive policing measures (Chan, 2003a). The use of both dictionary and directory knowledge are described by Schein as necessary to ensure the safety and protection of officers and reduce anxiety inherent in this risky and dangerous occupation (Schein, 1985, 1992). Recipe Knowledge Recipe knowledge is a f ramework of “recommendations and strategies for coping with police work” (Chan, 2003a, p. 33) . This form of knowledge is described by Sackmann (1992) as pr oviding r ecommended a ctions or r ecipes f or a chieving s uccessful resolutions. Examples provided by Chan (2003a) describe the learnt behaviour of police to do the m inimum amount of work required of them, where m ore a ctive officers are looked down on. Because there is a lack of reward for working hard, particularly in a job considered risky and unpleasant, police are encouraged to do the minimum required of them. Devery (2003) identified a pe rvading c ulture a mongst police trainees w here they observed incidents where reports were not being followed through because it was not de emed necessary, thus r educing t he pa perwork r equired. T he t rainees t hen internalised this knowledge, assuming this strategy in their own policing practices. This recipe knowledge also extends to the “code of silence” amongst police when faced with accusations o f m isconduct. A s di scussed b y C han (1999b, 2003 a) the f ear of not - 46- receiving s upport or ha ving s omeone ‘ watch your ba ck’ i n a da ngerous s ituation i s a deterrent to informing on fellow officers. Axiomatic Knowledge Finally, the explanations for why events occur are referred to as ‘axiomatic knowledge’ (Sackmann, 1992) . In t he c ontext of pol ice or ganisations, a xiomatic know ledge i s considered as a s et of u nderlying assumptions a bout t he r ole of t he pol ice i n s ociety (Chan, 1999a , 2003a ). The t raditional r ole o f police i s one of c rime fighting a nd protecting l ife and pr operty, a pe rception w hich i s e ncouraged b y pol ice t hemselves. This leads to greater public demand for dramatic police action thus contributing to the distinction between ‘real’ police work and routine duties (Chan, 1999b, 2003a). Habitus of Policing Work According t o C han (2003a) these f our di mensions of c ultural know ledge o ccur a nd inform t he habitus of p olicing, a c oncept de veloped b y Bourdieu (1990). B ourdieu describes habitus as a ‘product of history’ in that it is a system of characteristics which are f ormed t hrough t he integration of pr evious experiences. T his habitus is a s et of structures w hich m ay b e a dapted t o t he out comes w ithout t he ne ed f or c onscious consideration of t he pr ocesses i nvolved i n a chieving t he de sired out come. T his i s because t he de sired end has al ready be en determined by pr ior experience. Thus - 47- Bourdieu c onsiders t hat, a s i t i s a p roduct of p ast e xperiences, ha bitus e nsures t hat correct pr actices are consistent ove r t ime, and are cons idered to be m ore r eliable i n achieving the desired outcomes than other formal rules and norms (Bourdieu, 1990). As a product of history however, habitus would be limited by the scope of such history. This r esearch argues t hat t he habitus would be limi ted in relation to serial crimes due t o t heir r arity. Therefore t here w ould be i nsufficient p revious e xperiences upon w hich t o de velop this know ledge. T his l imitation w as r ecognised b y Bourdieu (1990) when he states that habitus is limited by the social and historical contexts from which it was developed. Bourdieu theorises t hat t he ha bitus pr ovides t he c ultural s ense making framework within the field; that is, the structural conditions within which one operates. When a pr ofessional e nters a ne w f ield he or she s till r etains t he pr evious ha bitus produced by the field they existed in previously. Therefore for new police recruits Chan states tha t the y ne ed to constantly e valuate the ir a ctions unt il the y a djust to the ne w habitus of the field in which they are situated (Chan, 2003a) . A ke y c oncept of t he f ield i s c apital, w hich m ay b e e conomic, social, informational or cultural. Chan (2003a) explains that in policing, the field is a reflection of t he s ocial, legal and pol itical capi tal. The ‘ currency’ of s uch capital i ncludes a number of resources both derived from the individual and from the organisation itself. - 48- Individual r esources i nclude r ank and experience, while organisational r esources include the opportunity for promotion and budgetary considerations. These c oncepts how ever a re not i ndependent of t he pl ayers w ithin the organisation. Chan (2003a) states that the police officers themselves have an active role in the development and transmission of cultural knowledge. As such, they can respond to and incorporate any changes to the field within which they operate. The development and transmission of police culture can be seen to be a learning process. As outlined in Chan (1999b) this of ten occurs through the te lling o f ‘war s tories’ by s enior, m ore experienced officers. As de scribed by Van Maanen (1978a) such anecdotes r elate t o mistakes ma de b y ot her of ficers a nd le ad to the r ecognition that a lthough mistakes occur on e ne eds t o p rotect ot hers t o e nsure t hat t hey are pr otected f rom t he consequences of such mistakes. There are a num ber of different i nfluences or a spects t hat c ontribute t o t he nexus t hat i s pol ice cul ture. More s pecifically, some of t hese cul tural as pects c an be seen to influence the way in which an investigation is conducted or how investigative techniques or t echnologies are us ed. This research aims t o consider t he way i n which cultural know ledge c an be s een t o i nform t he i nvestigation of s erial c rimes b y detectives. A s ha s be en i dentified, t he di ctionary kno wledge w hich i nforms investigative practice is often informed by what Van Maanen (1978a) described as war stories. It is proposed however, that since serial crimes in Australia are considered to be extremely r are, t hat s uch know ledge w ould not be a vailable. T herefore i t i s e xpected - 49- that the dictionary knowledge that is employed will be reflective of the ‘war stories’ of street level policing not of serial crime detection and investigation. One Part of the Whole - Police Subcultures Contrary to some traditional perceptions of police culture, Janet Chan (2003a) proposes that pol ice c ulture i s not a uni form c oncept but one w hich i s di verse a nd that can be influenced by a range of factors, including the socialisation process (p. 20). In an earlier study, Chan comments that different extant cultures are identifiable by the position of officer in the organisational hierarchy (1996). Thus consistent with Bourdieu’s concepts of field and habitus, this culture can be considered as a reflection of the types of capital one has, or the response to the challenges that the field poses to them as an individual or group (Chan, 2003a). O’Loughlin and Billing’s (2000) research into police cultural expectations among new recruits at the Victorian Police Academy, showed that while these subcultures may be identifiable based on a n officer’s rank, these subcultures can be overlapping and naturally discernable. Subcultures can be based on ot her aspects of the police organisation and one’s place in it such as the nature of one’s role in police service ( investigative o r ge neral dut y), t he geographic l ocation or gender. For t he purposes of t he c urrent r esearch, t he e xistence of pol ice s ubcultures i s c onsidered primarily in relation to detectives. Ericson (1993), in an observational study of Canadian detectives in their general investigative duties, commented that detectives can be seen to work within an ideology. - 50- This ideology shapes their decision making by providing explanations for their actions, consistent w ith r ecipe k nowledge. A s a cknowledged i n a n e xploratory s tudy i nto t he investigative processes of homicide investigations in England, the detective is seen by the general community as the “hard-working, hard-drinking, fornicating, and swearing crime-fighter” (Innes, 2003, p. 15) . This picture however, has been seen to evolve and soften over time, although, as identified b y Innes (2003), the extent of this perceptual change in perception is debatable. Consistent w ith the transfer of pol ice cul ture i n general, detective cul ture i s developed t hrough e xperience a nd t he s ocialisation pr ocess i nvolved i n be coming a detective. Innes (2003) discusses how ne w de tectives e xperience c ontinual m entoring by ot her m ore ex perienced detectives. These de tectives can be s een to share bot h the directory and recipe knowledge required at that level to conduct the work required. Although de tectives m ay be considered to be a s ubculture w ithin t he organisation as a whole, it is an expectation of this research that subdivisions may exist relating to a de tective’s pos ition within s pecialised i nvestigation uni ts or l ocal ar ea commands. This would be consistent with O’Laughlin and Billing’s (2000) contention that s uch s ubdivisions e xist. These di visions a re pr esent across pol icing ins titutions internationally, as Ericson (1993) provides an example where friction existed between two ‘teams’ of detectives, where there was a l ack of information sharing. Ericson also found t hat t here w as f riction e vident be tween d etectives f rom s pecialised uni ts a nd - 51- what he t ermed ‘general i nvestigation detectives’, s uggesting t hat s ubdivisions w ithin detective teams may be reflective of different knowledge capital. A long-term field study of two New York City police precincts led to a different perspective on pol ice s ubcultures (Reuss-Ianni & Ianni, 1983 ). T his s tudy i dentified two cultures: that of street cops and management cops. While both these cultures share organisational a ims of combating crime a nd m aintaining c ommunity s afety, R eussIanni and Ianni b elieve t hat s treet cops and m anagement cops have di fferent i deas on what this means and how this can be achieved. From this perspective, the street cop is the ‘professional’ cop who has the experience, the ability and the gut instinct to identify unwelcome a nd de viant pe ople a nd s ituations. Decisions a re m ade b ased on l ocal knowledge a t t he l ocal l evel, r ather t han using organisational goals an d associated procedures. Reuss-Ianni and Ianni (1983) contend that in contrast, the management cop relies on an ope rational c ulture i n publ ic a dministration a nd organisational management. T he pr oblems f aced by s treet cop s ar e cons idered on the br oader, city wide scale. Management cops must prioritise resources based on t his broader problem and consider other factors such as the political, economic and social limitations evident (Reuss-Ianni & Ianni, 1983). What’s Mine is Mine - Ownership of Information Another i nfluence s temming f rom pol ice c ultural know ledge i s t he c oncept of information ow nership, w hich i s r eflective of B ordieu’s pr oposition of c apital. T his - 52- issue w as r aised in E ricson’s (1993) study of d etectives in which general de tectives reported that detectives from specialised units were seeking to take credit for cases that fell within their jurisdiction. The homicide squad was singled out as the one most likely to t ry t o c laim unde served c redit b y t he ot her de tectives. In hi s s tudy of C anadian police, Ericson however also described how this concern over ownership of a case also occurred amongst general de tectives but t o a far l esser ex tent than e vident w ith detectives f rom s pecialised squads. One i ncident outlined b y Ericson describes how a detective be came i nvolved i n t he que stioning o f a nother d etective’s s uspect, an act considered to be poor form. This ill-feeling in such cases may stem from the measures of accountability in place, with one detective observed in the study expressing concern that he ha d not reached the required target b ecause a num ber of c ases ha d be en allocated to his partner (Ericson, 1993). This pr oblem w ith c ase ownership w as not a lways e vident. Ericson (1993) records several situations where other detectives were asked for assistance or willingly forwarded on cases deemed to relate to other investigations that were underway. Innes (2003) also describes t he s ocial construction of murder s quads or t ask forces w here a number of de tectives a nd s upport s taff a re b rought t ogether as one t eam i n or der t o accomplish an efficient and effective i nvestigation. In s uch a s cenario ow nership of information or ownership of the case as such, could be considered as collective as there is a shared investigative responsibility. As this research considers cases that have been investigated by bot h general de tectives a nd t hose f rom s pecialised uni ts, the i ssue of ownership will be explored. - 53- Case Theories and the ‘Art’ and ‘Craft’ of Policing In hi s cons ideration of detective p ractice and t he w ay i n w hich de tective s kills ar e utilised in the investigation of crime, Reppetto (1978) states that there is the perception: …of the detective as a n a rtist, the individual of brilliant insights, a master of interrogation a nd ot her s kills, w ho e ngages i n a n i ntuitive e xercise w hich ultimately leads to the solution of a crime (p. 8). In their discussion of the ways in which detectives carry out police work, Tong and Bowling (2006) contend that: The ‘art’ of detective work concerns intuition, instinctive feelings and hunches towards problem solving in an investigative capacity (p.324). This art is gained through experience and cannot be taught or developed by the police s ervice its elf, allowing de tectives to practice the c raft of pol icing. The experience gained allows detectives to read not only criminal behaviour but also that of other actors involved, such as witnesses (Innes, 2003; Stephen Tong & Ben Bowling, 2006). Those w ho have at tained the abi lity t o practice t he ar t of d etective w ork are considered to have an elevated cultural s tatus (Stephen T ong & B en B owling, 2006) . According to Tong and Bowling the recognition of this status and quality of the art is - 54- not s ubject t o e xternal scrutiny but r ather i nternalised a nd a cknowledged b y ot her detectives. The a rt of de tective w ork i s e vident i n how a n i nvestigation pr ogresses. One strategy employed in an investigation is the use of a ‘case theory’ which directs the line of i nquiry du ring an investigation (Dixon, 199 9b). Referred to by Innes (2003) as ‘playing a hunch’, a case theory can be seen as a decision making process formed from knowledge of a cas e and linked to the pe rceived ability o f a de tective to ‘read’ t he criminal element and the motivations of suspects. Although s uch a pr actice m ay be originally i nformed b y traditional s ources of information s uch a s witness a ccounts, D ixon (1999b) and Innes (2003) both acknowledge that dogmatic dependence on a case theory can have a negative impact on an i nvestigation. D ixon states t hat ot her l ines of i nquiry t hat are s upported b y other evidence m ay be ove rlooked or not i nvestigated f urther. Both a uthors r ecognise t hat evidence that is inconsistent with the case theory may be minimised or ignored. Innes (2002) discussed the way in which murder investigations are constructed in the context of the law, the police service, and the crime under investigation in the United Kingdom. Innes concedes that the existence of an incorrect case theory and the following of false leads are an important aspect in the development of major crime investigations such as murder inve stigations. H owever, a ccording t o Innes, “ such di fficulties t end t o be ‘down-played’ in any police accounts of the investigation” (Innes, 2002, p. 682). - 55- The ne gative i mpact of an incorrect cas e t heory i s evi dent i n the A ustralian context of t he investigation of s erial cr ime, with the cas e of H arry B lackburn. Dixon identifies a cas e of suspected serial rape in NSW, where an incorrect case theory saw evidence inconsistent with the theory distorted or suppressed (Dixon, 1999b). This case involved the investigation of a police officer for two separate series of sexual assaults. The case theory that the police officer, Harry Blackburn, was responsible for the series of assaults stemmed from another officer’s belief that Blackburn fitted the description provided b y s ome o f the vi ctims. As i ndicated by Dixon (1999b), t he R oyal Commission Inquiry into the case found that evidence was falsified to support the case theory and t hat e vidence s upporting B lackburn’s i nnocence w as s uppressed b y investigators (Lee, 1990). Dixon goes on t o state that the a bility of case theories to lead investigators astray in cases such as Blackburn’s, are “extremes on a continuum which runs back to everyday, often reliable hunches felt and assumptions made by officers in the course of their inve stigations” (Dixon, 1999b, p. 48) . Incorrect c ase t heories can l ead to an inability to identify those responsible for the crimes committed. In the Blackburn case, the m easures t aken to ensure t hat t he case t heory w as s upported, i ncluding t he tampering with evidence, led to the inability of others to use this evidence in any new criminal inve stigation. T hus, the t wo s eries o f r apes t hat w ere unde r i nvestigation remain unsolved (Lee, 1990). - 56- Innes (2003) acknowledges t hat a l ack of c hecks and balances i n the investigative process means that an incorrect case theory or an investigator with “tunnel vision” i s not r eadily i dentified. D ixon (1999b) identifies s everal reports int o police investigations which have acknowledged this problem and identified the need to have greater supervision of criminal investigations and greater scrutiny of the case theories. In discussing the reasoning behind police decision making and the case t heory that they employ in the United Kingdom, Keppens and Schafer (2004) state that the use of case theories is inevitable and the problem lies not with their use of case theories but with t he w ay i n w hich t hey r estrict t he s cope of t he i nvestigation. T his pr oblem i s considered to be reinforced by the police culture of police organisations as: …Work is done properly, and a case solved, if a suspect gets convicted. This orientation t owards positive r esults f avours a n “ inductive” e thos, where t hose pieces of evidence that points towards the guilt of the main suspects are seen as more valuable than those that would “falsify” the leading hypothesis (Keppens & Schafer, 2004, p. 110). This i s s upported b y C hatterton (1995) who states t hat de tectives w ould use t he information that they received to achieve the case outcomes that they considered to be appropriate. In a serial murder or serial rape investigation, such actions could influence the use of various techniques and technologies. In relation to the Blackburn serial rape case, this was detrimental to the achievement of the arrest of the serial offender in this - 57- case as no further investigation could be carried out due to the incorrect handling of the evidence. Investigating Stereotypes As previously referred to, the application and use of stereotypes in police work, a form of dictionary knowledge, can be considered an inherent measure of dealing with police work. A ccording t o a s tudy b y Sanders a nd Young (2002) conducted in the U nited Kingdom, police culture contains components which are sexist and racist, which affects the way in which police view society when coupled with the stereotypes held of people and groups. T he act of a ttributing a l abel t o a gr oup ba sed on t heir r ace a nd t he subsequent us e o f t his l abel a s a s tandard i n po lice de cision making i s referred t o as racial profiling (Engel, Calnon, & Bernard, 2002). In a discussion of the use of profiling in policing, Cameron (1999) claims that such pr ofiling i s t he r esult of pol ice e xperience, w here t hey d evelop a ‘ sixth s ense’ about different groups of people, which is thought to have “kept many of them alive” (p. 4) . In order to “protect s ociety from th e sinister e lements” pol ice ne ed some recognisable s igns as t o where the c riminal be haviour ma y ma terialise (Cameron, 1999). People who fall within marginalised groups in s ociety are often categorised as ‘worthless’ or vulnerable (Egger, 1998; Fattah, 1997). This labelling of certain groups - 58- as w orthless b y society can al so be i nternalised, as de monstrated by the pol ice’s attention or la ck of a ttention to the vi ctimisation of t hese groups. A ccording t o Innes (2003) many of these victims were previously involved in criminality or had lifestyles considered a s de viant. T his l eads t o c onflict in t he c ultural unde rstandings of t he victims’ place or worth. As a member of a marginalised group they would ordinarily be considered w ith di sdain, how ever as a vi ctim they a re g iven greater priority (Innes, 2003). This pr oblem w as recognised a s e arly as 1978 when a c onference i nto police practices and civil r ights w as he ld in Washington DC. Jean O’Leary (1978) of t he National G ay Task F orce d escribes how s erial r apes and as saults on gay men and lesbians were not i nvestigated or followed up b y pol ice. This was found to be due t o police di scretionary po wers t hat did not i dentify t he serial at tacks o n lesbians as warranting an investigation. To add to this, only one of the six rapes and eight assaults was entered into the police’s central computer: this information, if entered, would have meant a sudden increase in crime would have been detected. The decisions not to report correctly were made by the local police officers. A similar attitude towards homosexuality was historically evident in the NSW Police Force during t he 1960s as ackno wledged by Finnane (1999, pp.11 -12) quoting the Commissioner’s comments at the time that offences related to “homosexuality and other forms of perversion” had the potential to provide a serious threat to the youth in the community. - 59- These s tereotypes i nform the di ctionary know ledge t hat pol ice us e t o decide how t o respond t o s ituations they encounter and thus how t o ‘do pol ice work’. In t he investigation of s erial m urder and serial r ape, t hese s tereotypes are reflected in the police response to the victims of these crimes. As will be examined later in this chapter, the victimology of serial crimes and the social stereotypes that pertain to some victim groups can be seen to be internalised in the police cultural knowledge. The outcome of this us e of know ledge i s cons idered t o impact on the de lay in the ide ntification of crimes as serial in nature. A Language of Their Own From a study into police socialisation practices o f NSW Police, Devery (2003) found that there were incidents where police used the specialised terminology or language of operational police. This specialised language can be seen to reflect common stereotypes through the use of derogatory terms to refer to groups of people such as derelicts and those w ho r esided i n H ousing C ommission properties, for e xample. This s pecialised language is a form of dictionary knowledge as previously discussed. The us e of t his f orm of dictionary knowledge is not r eserved f or of fenders or undesirable types of victims. Devery (2003) also out lined a n i ncident w here a n ol der woman who raised concerns about unusual noises in her complex was dismissed as just another ‘ old l ady’ who ‘heard noi ses’. T he op erational pol ice w ho a ttended t hat c all - 60- placated the woman and left without adequately searching the area, later explaining that it was “not uncommon for ‘old ladies’ to complain about this sort of thing because they were afraid.” (Devery, 2003, p.120). Shortly after it w as found that her neighbour had been the victim of a home invasion. This labelling of the reporter as an ‘old lady’ and the s ubsequent j ustification f or i naction b y pol ice took pl ace in the pr esence of a Student Police Officer, demonstrating the ability of such stereotypes to be transmitted through t he s haring of c ultural know ledge. These s tereotypes o f di fferent gr oups ha s been s hown t o i mpede the i nitiation of i nvestigation i nto pa rticular c rimes (Innes, 2003). According to Innes (2003) the use of formal and informal language is a means of dealing with the complex and sometimes ambiguous nature of policing. In relation to the c rime of mu rder, Innes ide ntifies s even t ypes of m urder a s c onstructed b y t he operational l anguage of police. These s even types as d escribed by Innes (2003) are: domestic hom icides 5, a rgument m otivated or c onfrontational hom icides 6, c hild murders 7, c riminal c ause m urders 8, s exual m urders 9, stranger m urders 10 and serial murders. These cat egories ar e cl osely aligned w ith classifications d erived from academic r esearch and are l oosely b ased on the motivational f actors inf luencing the offending be haviour, the vi ctim-offender r elationship, t he ‘ type’ of vi ctim a nd t he 5 Act stems from an argument between people in a relationship, for example husband and wife Death was an outcome of an argument between two people, such as young men in public settings using violence 7 Recognised as being difficult cases emotionally for police 8 Death occurs when either the victim or offender is taking part in criminal activity 9 Homicides considered to be sexually motivated 10 Victim is unknown to attacker. This can often coincide with other categories 6 - 61- offence i tself. As di scussed b y Innes, t his c lassification f orms a s et of s hared understandings by police, which is used to inform the use of investigative strategies. Who Could Be Bothered? Results of Accountability in Practice In NSW the Luster Inquiry of the early 1980s and the Royal Commission inquiries of the 1990s identified problems in a number of areas of policing including management, training a nd, most s ignificantly to police pr actice, corrupt pol icing pr actices (Chan, 2003a). T he r ecognition of c orrupt pol icing practices i n N SW a nd ot her pol ice organisations l ed t o c hanges i n t he m easures us ed t o s crutinise pol ice pr actice. In conjunction w ith t his c ame t he i ncreasing ne ed for pe rformance b ased accountability measures. According t o C han (2003a), i n or der t o achieve t hese f orms of a ccountability these measures needed to be internalised in the dictionary and directory knowledge of police. O ne m ethod of achieving t his w as t hrough a n i ncrease i n t he c ompulsory completion of forms which detailed police work carried out. This measure of increasing paperwork t o s upport accountable pr actices c an be considered t o h ave b een c ounter productive w here t his ne ed f or a ccountability can be s een t o i nfluence t he a ctions of police. The c onstraints of pol icing a lso created problems f or the m easures of accountability in place. As outlined by Ericson (1993) detectives were often confronted - 62- with of fences w here t here was i nadequate i nformation or evidence with w hich t o proceed, however the detectives took the time to contact and interview the victim. This was referred to as a publ ic relations exercise by the detectives interviewed in Ericson’s study. The concept of consolatory police action as a ‘public relations exercise’ was not reflected as such in the paperwork submitted by detectives. Ericson comments that the paperwork produced by detectives was shaped by the organisational guidelines to make it appe ar as t hough real i nvestigative work ha d be en c arried out . T he occurrence of these publ ic r elations e xercises appe ared t o be limited to relatively m inor of fences however, and not crimes such as murder therefore it is not expected that such practices are relevant in serial crime investigations. Although t here a re many f ormal m odes of a ccountability f or police organisations, the police can also be subject to informal modes of accountability from external sources. One of the most significant of these sources of informal accountability as identified by Dixon (1999a) is the media. Ericson (1993) illustrates the influence that such a m ode of i nformal a ccountability c an ha ve on a m urder i nvestigation, w here resources are allocated based on the predicted public response to a crime. Discretionary powers An i ntegral pa rt of be ing a p olice of ficer is the pr actice of us ing the d iscretionary powers that are inherent in the job. These discretionary powers relate to the application - 63- of the laws and regulations and are often shaped by time restrictions, limitations in the availability of r esources a nd a n of ficer’s know ledge of s uch l aws a nd r egulations (Corsianos, 2003) . As i ssues of time a nd resources ar e of conc ern here, pol ice discretion may be considered as the enacting of recipe knowledge as identified by Chan (2003a). Police officers have the power to act in a given situation, but also the power to choose inaction (Ericson, 1993). As outlined by Corsianos (2003), the ability to use discretionary powers is not reserved for hi gh ranking of ficers but oc curs at all l evels of the pol ice or ganisation, from street cop to detective. The accountability of such discretionary decision making however varies at the different levels in the police organisation. According to Ericson (1993) the nature of detective work occurring away from the usual supervisory controls experienced by street cops means that a low-visibility autonomy exists for what Ericson terms as ‘specialized’ police. Corsianos (2003) suggests that the use of police discretion can lead to unequal treatment of pe ople or g roups i n c omparable s ituations. T his s uggests t hat pol ice dictionary knowledge and subsequently directory knowledge can be used to guide the exertion of this discretionary power, an idea which is supported by Ericson’s judgment that di scretion i s a c omponent of t he or ganisation a nd not i ndividualistic (Ericson, 1993). As such, the stereotypes subscribed to by pol ice may also influence the use o f this power when situations occur as one group of people may be seen as more deserving of police attention than another. - 64- (Over) Accountability The use of discretionary powers as a means of avoiding the processes of accountability can also be s een in police cultural practices. T he a voidance of pa perwork i s a n issue raised by Devery (2003), who cites incidents not being followed through as they were not deemed ne cessary and i t reduced t he amount of paperwork they had t o do. T hese incidents oc curred in the pr esence o f t rainees, thus reinforcing and c ontinuing t his culture. As outlined in Chan (2003b) the high levels of bureaucratic and accountability measures were s een to be n egatively r eceived b y police r ecruits. The ne ed to be accountable w as considered to generate a h eavy w orkload a nd e xtra pr essure on t he recruits. Similar a ttitudes w ere e xpressed in a n umber of s tudies in cluding, but not limited to, those b y E ricson (1993) and Innes (2003). The m easures o f accountability are bl amed by police f or di stracting th em f rom real pol ice w ork (Chan, 2003b) . O ne detective f rom E ricson’s (1993) study e xpressed t he vi ew t hat t hey w ere no l onger evaluated ba sed on c ase c losure r ates but on t he a mount of pa perwork t hat w as submitted. - 65- Victims of Violent Serial Crime The c onsideration of the r ole of the vi ctim in the c riminal jus tice pr ocess ha s become a c entral f ocus for academic r esearch (Zedner, 2002) . A ccording t o Zedner (2002) the amount of research into the impact crime has on victims and their associated needs, and t he wide use of vi ctims s urveys have moved t he t opic of t he ‘victim’ into both criminological and political arenas, making the victim a key player in the criminal justice system. Zedner (2002) claims that although the focus on the victim is a relatively recent phenomenon, the acknowledgement of the study of victimology can be traced back to 1949 when psychiatrist Frederick Wertham r eferred to such a s tudy as a ‘ science’ (Wertham, 1949 c ited in Zedner, 2002). The term ‘victimology’ has been described as referring t o “the s tudy o f v ictims” (Turvey, 200 2d, p.137) . In general terms thi s c an refer to victims of natural disasters o r a ccidents (Turvey, 2002d ) or victims of crime. There ar e many as pects that a re c onsidered t o inform t he s tudy of vi ctims. These include the of fence tha t was c ommitted against the vi ctim, the vi ctim’s r esponse, the characteristics of the victim and why the victim was targeted (McLaughlin & Muncie, 2001; Turvey, 2002d). A randomly chosen victim can tell investigators much about the type of offender that they are seeking (Turvey, 2002d). Egger (1998) raises a nother r eason t hat i nvestigators s hould consider t he victimology of serial crimes. Egger argues that often physical evidence from the crime - 66- scene o f an incident s uch as s erial m urder m ay be l acking or non -existent. T hus, the only information that investigators may have to go upon is that pertaining to the victim. Egger (1990b) claims that the consideration of victim information is essential in a serial crime investigation. One of the most common features of victims of serial crime is that they a re unknow n t o t he s erial of fender (Egger, 1990b) , t hus t his pos es problems f or investigators a s t here m ay appear t o be no c ommon l ink be tween vi ctims unt il s ome considerable t ime ha s elapsed. In a n i nvestigation, i nformation a bout t he vi ctim c an also yield a n i nsight i nto how , w here a nd s ometimes w hy t he of fender has s elected particular victims. Additionally, police knowledge of some groups of victims can also pose a p roblem f or t he i nvestigation of s erial c rime a s recognition of the vi ctim a nd their place in a series may not readily occur. A Reflection of the Offender As a bovementioned, t he c onsideration of t he vi ctims of s erial c rime c an be us ed t o draw i nferences a bout the unknow n of fender (Turvey, 2002d) . It h as a lso be en suggested that some victims may initiate contact with the offender in what is described as victim precipitation (Egger, 1990b). For example, when a hitchhiker asks a s tranger for a lift or the victim a sks someone for directions. Victim precipitation has been the focus of s tudies i n r elation t o t he vi olent c rimes i ncluding r ape a nd s exual as sault (Marciniak, 1998; P olk, 1997; T hornton, R obbins, & J ohnson, 1981) . Egger (1990b) states tha t vi ctim pr ecipitation is c onsidered t o be a contributing factor to many homicides. However, the e xtent to which this a ct of pr ecipitation could play in serial - 67- crimes is unknown. Victims may also be targeted because of what they represent to the offender (Egger, 1990b). This may be their physical appearance or because they evoke symbolic s imilarities w ith something or s omeone tha t the s erial of fender c onsiders significant (Fox, et al., 2005). Serial m urderers o ften deliberately choose vi ctim-disposal me thods tha t w ill make discovery of the crime difficult. Serial murderers are likely to use extreme forms of vi ctim di sposal, s uch as bur ying bodi es i n i solated pl aces w here t heir discovery i s unlikely (Levin & Fox, 1999) . T his i s un like many ot her c rimes s uch a s s pousal murder, where in most cases there is no a ttempt to conceal the victim, who is usually discovered relatively soon after death (Fox & Levin, 1999). The method employed by a serial of fender t o di spose of t he vi ctim’s bod y c an a lso reveal the or ganisational and planning methods used by the offender (Fox, et al., 2005). As previously outlined, the categorising of an offender as organised aids in the development of an offender profile (Ressler, Burgess, & Douglas, 1988). Rape Myths In cases of serial rape the issue of identification of crimes as serial is still difficult even though it is the victims that report these crimes, enabling the police to gain information about t he of fences w hen r eported. H owever i t i s s till a cknowledged t hat t he r ate of reportage i s r elatively low i n c omparison t o ot her of fences (Lievore, 200 2). A lower rate of reporting can mean that the number of cases investigators are able to link may - 68- not represent the total number of incidents; this may also lead to a failure to link related crimes. One of the greatest difficulties investigators face in relation to the victimology of serial rape crimes is identifying victims who have not previously contacted police. An issue tha t ma y im pact on the inve stigation of s erial crimes is the w ay in which these crimes are defined, specifically in regard to the relationship or lack thereof between vi ctim a nd of fender. T herefore i f offenders w ho do not c onform t o t hese definition of serial r ape, it may be deduced that both the offender and the victims are viewed differently by the legal s ystem. If these offences had escalated to murder then the offender would have been considered a s erial m urderer and allocated the r equired resources to investigate these crimes. Therefore it must be asked if the characteristics of the vi ctim a nd t heir r elationship t o the of fender should de termine i f a s eries of r apes should be considered as serial. It ha s be en stated that s erial r apists ar e pr edatory i ndividuals, seeking t o gain control a nd dom inance over t heir vi ctim (Stevens, 1994 ). T his de scription how ever could a lso a pply t o of fenders w ho c hoose vi ctims of t heir acquaintance t herefore t he definitional r equirement t hat no pr ior r elationship s hould e xist be tween vi ctim a nd offender should not deter police from investigating these crimes as they would strangerserial rape cases. It should also be asked that if a series of rapes occurs and is deemed not to warrant an investigation as serial in nature due to a relationship between victim and offender could this be seen as a return to the acceptance of some forms of rape such as that once held for intra-marital rape. - 69- Perceptions of the rape victim’s role in the actual offence have been the subject of debate. According to Abrams et al. (2003) extant ‘rape myths’ can attribute blame to the victim. According to Burt (1980) rape m yths are stereotypes and false beliefs that can be p rejudicial t owards t he c rime, t he vi ctim, a nd t he of fender. T hese m yths are based on unf ounded a ssumptions a nd c ommon e xamples i nclude ‘ only bad girls get raped’ and ‘women ask for i t’(Burt, 1980 ). These r ape m yths c an include t he general belief in Lerner’s renowned (Crome & McCabe, 2001) ‘Just World Hypothesis’ where bad t hings onl y h appen t o t hose who de serve i t. Lerner (Lerner & M iller, 1978) examined the attributional process of blaming the victim and called it the ‘Just World Hypothesis’ where, “individuals have a need to believe that they live in a world where people generally get w hat t hey de serve and deserve w hat t hey get” (Lerner & Miller, 1978, p.1030). It w as ar gued by Lerner t hat pe ople generate t he be lief t hat a vi ctim’s failure to take pr ecautions a nd nor mal pr eparations t o a void ne gative circumstances leads to their outcome; that is, their victimisation. This ide a tha t the vi ctim ha s s ome c ulpability in their ow n victimisation however, can also be s een t o relate t o a person’s st atus i n society. Lerner s tates t hat those who are considered as ‘deviant’ or part of the minority including “the foreigner, and the mentally ill” (Crome & McCabe, 2001, p . 399) are considered as deserving of the events from which they become the victim. - 70- Similarly, Miller and Schwartz (1995) interviewed prostitutes in a Midwestern city i n t he U nited S tates a bout t he r ape m yths pr evalent a mong c lients, pol ice a nd people t hey w ould e ncounter ge nerally. It w as f ound t hat c ommon rape m yths contributed to the violence against these women. Emerging from this study were four key m yths t hat w ere found t o be us ed t o s upport t he w ilful i gnorance of vi olence against prostitutes, while incorporating a fundamental belief that prostitutes are public property, and as such, open to abuse. The four key myths identified that: prostitutes are unrapeable; no ha rm i s done t o pr ostitutes w hen a ssaulted; pr ostitutes de serve t o b e raped; a nd, all pr ostitutes ar e cons idered the s ame (Miller & S chwartz, 1995) . Miller and S chwartz s tate t hat due t o t heir pos ition i n s ociety w hereby t hey op enly s ell s ex, prostitutes are confronted by a stronger collection of rape myths, the strength of which is beyond that experienced by other women. The r elationship be tween t he r ape vi ctim a nd of fender c an also i nfluence people’s perception of victim blame. The study by Abrams et al. (2003) found that in some cases people were more likely to attribute blame to the victim when a relationship existed between the victim and the offender. Those findings were dependent upon t he levels of benevolent sexism 11 exhibited by participants. Such findings are supported by other s tudies s uch a s t hat of H azelwood a nd B urgess (2001) into t he i nvestigation of rape of fences s uggest t hat pol ice of ficers us ed f actors s uch as t he exi stence of a relationship be tween t he vi ctim a nd of fender t o de termine t he s trength of t he c ase. Additionally, if the victim were considered to have been forced into the situation where 11 Benevolent sexism refers to attitudes which perceive women conforming to stereotypical roles in a positive light (Abrams et. al. (2003). - 71- the r ape oc curred t hen t he c ase w as c onsidered to be s tronger b y pol ice t han i f t he victim had willingly gone with the offender. In a r eview of t he l iterature addr essing pol ice attitudes t o rape, Page (2007) suggests that in general, police officers accept rape myths whereby decisions on victim credibility and the ensuing investigation were based on t he conformity of the victim to stereotypical r ape behaviour and the assault con form to ‘ideal’ ch aracteristics of rape. As discussed by Page, police culture can be seen to maintain the “rigid attitudes about the roles and actions of men and women in society” (Page, 2007, p.23) which reinforces the continued acceptance of rape myths by individual police officers. Thus, what police officers consider the ‘ideal’ victim and offence shapes the dictionary knowledge of rape offences a nd i nforms t heir di rectory know ledge of w hat investigation, if a ny, is necessary. There are also myths that specifically focus on the victimisation of those within the c ontext of s ex r oles a nd hom osexuality. M any of t hese m yths a nd s tereotypes surround male sexual assault, as outlined by Mc Adam and Fitts (1999) in their research into the responses to the reporting of male sexual assault. One common myth identified was t hat m ales cannot b e cons idered as vi ctims of s exual as sault b ecause an y sexual contact i s a pos itive experience for t hem and t herefore not c onsidered as ‘assault’. Another m isconception i s t hat hom osexuality i s a f eature of t hese of fences w here t he offenders a re hom osexual a nd t he vi ctims ‘ become’ hom osexual, a m yth s trongly refuted by McAdams and Fitts. As argued by Mc Adam and Fitts, such myths pervade - 72- both t he g eneral c ommunity a nd t he pr ofessions that deal w ith m ales s exual as sault victims, such as police. According to Stewart’s (1999) report into the response of NSW Police to cases of sexual assault, there was dissonance in the way in which incidents were reported and subsequently addressed. T his us ually r eflected the t ype a nd s everity o f t he s exual offence r eported, and the t ime t hat pa ssed be tween the of fence and when it w as reported to police. Stewart s tates tha t thi s inc luded a te ndency not to take s tatements from victims; this was i n accordance with police protocols at the time of some of the offences. Additionally, there was a t endency to refer victims to sexual assault services only when there was forensic evidence to be collected. There was an apparent failure to provide referrals for victims of indecent assault or attempted sexual assaults. The Invisible An issue acknowledged both in Australia and internationally, is that the death of some members of t he community can oc cur w ithout cr eating an alarm. Specifically v ery young children and the elderly are two groups whose deaths can be attributed to natural causes without the necessary evidence to support such a conclusion (Kraemer, Lord, & Heilbrun, 2004) . T he de aths of e lderly p eople, pa rticularly t hose kno wn t o ha ve suffered f rom a l ong-term illness, are less likely to be queried as their d eath is at the least, not unexpected. This factor is often exploited by the ‘angel of death’ female serial murderer, who uses this to their advantage (Kelleher & Kelleher, 1998). - 73- Similarly, the deaths of young children, when under 18 months of age, may not be considered as suspicious. This is particularly evident if the actual cause of death is not determinable, it is often attributed to Sudden Infant Death Syndrome (SIDS). This practice, and the serial murderer’s ability to take advantage of these circumstances, was recently uncovered in Australia with the conviction of Kathleen Folbigg. Folbigg was convicted of murdering three of her four children over a period of time. The children’s deaths were originally attributed to SIDS, until investigators found evidence suggesting that their deaths were deliberate. There is currently no national system in place in Australia to track the deaths of children, particularly w hen the cause of de ath h as be en r uled t o be ‘natural’. Some Australian states h ave monitoring s ystems in place; the NSW C hild D eath R eview Team (CDRT) provides a benchmark for other systems. The CDRT was established in 1996 12 to m aintain a r egister of de aths of c hildren unde r t he a ge of 17 years (NSW Commission for Children and Young People, 2008). This s ystem how ever, d oes not a ppear t o h ave an y i nvestigative cap ability or use. As stated in their recent annual report, the CDRT has six specific functions: • • • 12 maintain the Child Death Register, classify deaths according to specific criteria, identify patterns and trends emerging from the data, This was introduced following the Folbigg case. - 74- • • • undertake research into prevention and reduction in child deaths, monitor recommendations proposed, and identify ar eas w here f urther r esearch is r equired (NSW C hild Death Review Team, 2007). In addition, t he r eport s tates t hat t he C DRT m akes r ecommendations t o pr event or reduce child de aths e merging f rom t hese f unctions, for impl ementation by bot h government a nd non -government a gencies a nd t he c ommunity i n general. W hilst t he NSW O mbudsmen r eviews de aths where t he child ha s ha d a c onnection t o t he Department o f C ommunity S ervices 13, ha d b een i n de tention, or ha d a di sability a nd had lived in some form of residential accommodation, and have the capacity to review suspicious de aths. T his however i s s eparate f rom t he C DRT. T here i s no a pparent mechanism f or t his s ystem t o be acc essible b y NSW P olice or an y ot her A ustralian police agency. Similar s ystems a re in place in most A ustralian states, although there is currently no formal review mechanism in existence in the Australian Capital Territory (Commissioner for Children Tasmania, 2006; NSW Child Death Review Team, 2007). These systems have similar functions to the CDRT and are restricted to reviewing the deaths of the particular s tate’s residents, which m eans that there is a gap in recording the de aths of children w hose de ath oc curs i n a s tate ot her t han i n w hich t hey r eside. 13 Provides support and services for children and families experiencing a range of difficulties including, but not limited to, child protection - 75- Again, as with the CDRT, there appears to be no i nvestigative mechanism available to state police agencies. This lack of available monitoring mechanisms that are accessible to police could also pr ovide one explanation f or t he l ow incidence o f r ecorded s erial m urder b y females. Although there a re moni toring s ystems in place, there is no national mechanism to assist in determining previous in cidents in other s tates o r j urisdictions. This ha s the c apacity to result in many cases g oing undetected o r unsuspected. Therefore t he actual i ncidence of female s erial m urder may b e m uch hi gher t han generally be lieved, as m any t arget t he groups de scribed above (Holmes, et al., 1998). Reduced investigative capabilities d ue to a la ck of a vailability of moni toring s ystems available to police may be a compounding factor. Without a Trace- The Missing An i mportant issue t o be cons idered is tha t, in A ustralia a lone, a pproximately 3 5,000 people a re r eported m issing e ach year (James, A nderson, & P utt, 2008) . James et al . found t hat a m ajority o f pe ople a re l ocated, w ith 90 percent of missing pe rsons i n Victoria located within seven days and found to have disappeared of their own accord. For t he s mall pe rcentage t hat a re not l ocated w ithin t welve m onths of be ing r eported the pol ice do attempt t o trace t hese pe ople. H owever with pol ice r esources b eing - 76- restricted, investigators need to prioritise ea ch case b ased upon t he i nformation t hey have received and their general knowledge and experience. Research into the investigation of missing person reports has been limited and erratic, particularly in Australia. There has been little current research in Australia that extends beyond the consideration of the incidence of missing person therefore much of the ava ilable acad emic r esearch is da ted. An early research brief t hat cons idered the findings of two reports into missing persons, considers the investigative responsibility as resting generally with local police in most Australian states and territories (Swanton & W ilson, 1989) . This report also considers th e ide ntification of ‘ at r isk’ mis sing persons and how this was determined. According to Swanton and Wilson (1989) police categorise m issing pe rsons, w ith t hose considered m ost a t r isk be ing t hose una ble t o protect t hemselves. T his i ncludes t he e lderly, young c hildren a nd t hose w ho ha ve a mental or physical disability. The police response to missing persons is driven by both formal a nd i nformal r isk c riteria us ed t o a ssess t he a ppropriate r esponse (Swanton & Wilson, 1989). This ear ly research into missing p ersons also recognises t hat t here were assessments made which proved to be inadequate. It was suggested that the only way to ensure an increase i n e ffective as sessments w ould be t o treat all c ases as ur gent (Swanton & Wilson, 1989). However Swanton and Wilson contend that such a measure would result in a waste of resources, considering that the majority of missing persons are l ocated within days of t heir di sappearances. The ne ed for a na tional r egister of - 77- missing persons was recognised as being a possible tool to assist in the investigation of missing pe rsons acr oss s tates, however t he cos t ef fectiveness of s uch a m easure w as identified as an issue yet to be resolved (Swanton & Wilson, 1989). Research following this report was largely absent for nearly a d ecade. A report by H enderson a nd H enderson (1998) considers t he i mpact of m issing pe rsons on t he general community in Australia. An issue that was revisited in this report was the need for a n ational a pproach to t he i nvestigation of missing pe rsons. T he e stablishment of the N ational M issing P erson U nit ( NMPU) was t outed a s a m eans of c oordinating national ef forts, while a f ormalised crime l inkage s ystem w as r ecognised as an investigative t ool t hat can be us ed t o p rovide s hared resources (Henderson & Henderson, 1998). The Henderson report also discussed the level of satisfaction of the public with the pol ice r esponse t o t hose r eporting s omeone missing. A s urvey o f t he f riends and families of a sample of missing persons was conducted to obtain a representation of the levels of s atisfaction with both police a gencies a nd other or ganisations tha t a ssist in locating m issing p ersons (Henderson & H enderson, 1998) . T wo i ssues emerged from this s tudy: t he t imeliness of pol ice a ction onc e t he r eport ha d be en m ade a nd t he feedback pr ovided b y police dur ing th e inve stigation. However, there w as little discussion of t he i nvestigation i tself. M any of the r eports w hich followed t hat of Henderson and Henderson, were based on the findings of this research and therefore do - 78- not c ontribute a nything further t o t he unde rstanding of t he i nvestigation of m issing persons. A r ecent r eport int o the inve stigation of mis sing p erson reports b y Victoria Police f ocused on t he l inkage be tween uni dentified r emains and out standing m issing person cases (Office o f P olice Integrity, 200 6). T his w as i n r esponse t o t hree complaints about the failure of Victoria Police to identify a reported missing person as deceased. Of significant interest is this report’s suggestion that: the e stablishment of a na tional da tabase w ould be t he be st w ay f orward. A national da tabase w ould ens ure a ll A ustralian Missing P ersons R eports (irrespective of w hich state t hey w ere r aised in) co uld be m ore ef ficiently matched to corresponding unidentified bodies (irrespective of which state they were found in) (Office of Police Integrity, 2006, p.10). In this 2006 r eport, there is no m ention of the existence of a national database and its intended purpose for the monitoring of both long term missing persons and unidentified remains, a s i dentified b y P utt a nd H iggins (Putt & H iggins, 1997) . W hile t his r eport indicates t hat a f ormal l inkage s ystem ha s no t b een incorporated as an investigative tool, it specifies the need for another tool, a forensic odontology database to provide an accessible c ollection of dental r ecords of m issing p ersons (Office of P olice Integrity, 2006). - 79- During t he pa st de cade t here ha s be en limited research into the p olice investigation of mis sing pe rsons int ernationally, w ith the mos t s ignificant r eport produced by Geoff Newiss (1999). This study looks at nine police forces in the United Kingdom and examines the pol icies r elating to missing persons r eports. Few of t hese reports offered any models that could lead to more effective policy and practice in the investigation of missing persons. An issue that may impact on the risk assessment of some individuals is that not all mis sing pe rsons a re residents of A ustralia. According t o t he Department o f Immigration and C itizenship (Department o f Immigration a nd C itizenship, 2007) just below 46 4 00 vi sitors t o A ustralia ove rstayed t heir vi sa as of December 2005. W hile this equates to less than one percent of all visitors for the given period, it can still be considered as a significant number of people. This acknowledged level of visitors who overstay t heir vi sas can be s een t o c ontribute t o the di ctionary knowledge of police. With large numbers of tourists voluntarily becoming untraceable, it may be considered that t he r eported pol ice pe rspective that ma ny tourists te nd to e xtend their s tay b y choice, and as such do not require the allocation of investigative resources. It should be noted that according to the Department of Immigration and Citizenship, police agencies are not considered r esponsible f or l ocating vi sitors w ho ha ve ove rstayed t heir vi sas (Department of Immigration and Citizenship, 2007). In addition to immigration factors, a study conducted by Henderson, Henderson and Kieren (2000) found that less than 1% of people in Australia were still missing 18- - 80- 24 months after they were initially reported missing to police which equates to two of 505 pe ople. This m eans t hat a pproximately 118 of t he 3 5,000 pe ople t hat go m issing every year in Australia remain so after 18 months. Thus this could be seen to give rise to the understandable expectations of police that a majority of missing persons will be located i n t ime or when t hey s ee fit or are a ble t o not ify f amily and friends of t heir whereabouts. The Worthless A factor that may contribute to the difficulty faced by law enforcement in identifying a series of r elated m urders how ever, i s t he vi ctim s election m ethods of t he i ndividual serial killer. Often victims from marginal groups are targeted, such as the homeless or prostitutes, as they may not be reported missing as readily as other members of society (Fox, et al., 2005). According to Fattah (1997), some groups of victims are considered “worthless” w hich he de fines as be ing t hose w ho are m arginalised by s ociety du e t o their lifestyle choices which do not conform to those of the society in which they live. Fattah identifies pr ostitutes, the ho meless and hom osexuals, a s j ust some of t he ‘outcasts’ that can b e i ncluded. E gger (1990b, 1998) states th at the se ma rginalised groups are often at a greater risk of being victimised by serial offenders, because their lifestyle br ings the m int o contact w ith greater t hreats. This fa ctor w as validated in 2003; G ary Leon R idgway appeared i n a S eattle c ourt a fter pl eading g uilty t o t he murders of 48 females i n the S eattle area s ince 1982. T he majority of the se vi ctims were prostitutes. Ridgway (quoted in BBC, 2003) stated that: - 81- I also picked prostitutes as victims because they were easy to pick up without being noticed. I knew they would not be reported missing right away and might never be reported missing. I picked prostitutes because I thought I could kill as many of them as I wanted to without getting caught. Not onl y doe s s uch a choice i n vi ctims m ake i t di fficult t o i dentify a cas e of s erial murder, but the response by the community to such crimes may also be reflected in this. It ha s b een argued that some as pects of a com munity, particularly t he media, do not consider the de aths o f pr ostitutes a s w arranting a hi gh l evel of m edia c overage (Fineman & M cCluskey, 1997) . T his l ack o f media attention received can be s een t o hamper an investigation, because the public is less aware that such crimes have taken place. The la ck of a ttention that the vi ctimisation of prostitutes r eceive f rom s ome police w as hi ghlighted i n a n A merican s tudy of Miller and Schwartz (1995). Interviewing a ci ty vi ce officer as pa rt of t he s tudy i nto rape m yth prevalence am ong clients of prostitutes, police and other, they were told that: …it was a common belief among local police that when a prostitute was found dead: “Who really cares? Who really cares? [You’re dealing with] a girl who’s zero” (Miller & Schwartz, 1995, p. 14). It w as con cluded that such beliefs b y t hat p articular pol ice d epartment ha d been indicative of the a ttribution of e xtreme le vels o f vi ctim bl ame (Miller & S chwartz, - 82- 1995). The prostitutes interviewed in the study of Miller and Schwartz stated that their crimes of prostitution were considered to be more serious by the criminal justice system than t he c rimes of r ape a nd vi olence c ommitted a gainst t hem. A lthough t his s tudy conducted ov er t en years a go, i t doe s d emonstrate t he a ttitudes e xisting in t he pol ice department studied. Similar attitudes are evident in a later study conducted in Australia, suggesting that such police attitudes are not restricted to America alone. Reflecting on their interactions with prostitutes in Australia, referred to as sex industry workers in their paper, Bridgett and Robinson (1999) comment that prostitutes are one group in society that are marginalised and thus vulnerable to acts of abuse and violence. B ridgett a nd R obinson e xamined a NSW Standing C ommittee of S ocial Issues report (Legislative Council, 1996) which found that there was an extremely high incidence of violence among this group, although the reporting of these crimes is low. It w as a lso f ound t hat those w ho of fend a gainst pr ostitutes f requently ha d a p rior history of sexual offences and would target prostitutes as they believed that there was a lesser chance of conviction. As discussed by Bridgett and Robinson (1999), these vi ctims of sexual assault often f eel unf airly t reated b y both t he l egal and he alth s ystems. T hose w ho ha ve reported an assault describe t he di sbelief with w hich their r eport w as r eceived by police. Bridgett and Robinson also identify that there have been claims where a woman has reported a sexual assault to police and were told that it was, “part of their job and consequently w hat they de serve” (Bridgett & R obinson, 1999, p.4) . Victims ar e al so - 83- faced w ith inadequate outcomes in the c riminal jus tice s ystem, whereby attitudes expressed can be seen to reinforce the myths that prostitutes deserve to be victimised or that it is not as serious. It i s of ten cont ended by academics i n t he field, and by at l east on e s elfconfessed serial of fender (Anonymous, 1998) , that t he s erial of fender d epersonalises his victim. However according to Egger (1998), victims of serial crimes such as murder are often depersonalised by society as well. Particularly when victims are drawn from the ‘ worthless’ cl asses, many pe ople justify the mur derer’s a ctions b y deeming t hat victims ‘ had it c oming t o them’ (Lerner & M iller, 1978) . T he vi ctim i s no l onger o f interest to society but becomes, according to Egger, ‘less-dead’ in that the undesirable groups have been cleansed from society by the offender. According to Turvey (2002d) investigators who share this perception of victims may not deem their victimisation as warranting a thorough investigation. Chapter Summary Police c ultural know ledge c an be s een t o i nform pol icing pr actice a nd s hape decisions m ade i n t he f ield. T his know ledge and t hus i ts r esulting a ctions a re not uniform a s t here are id entifiable s ubcultures e xistent in policing o rganisations. T he cultural knowledge and more specifically the dictionary knowledge inherent within the organisation can l ead t o recognisable outcomes in t he w ay i n which an investigation may be condu cted. The pol ice pe rception of t he vi ctims a nd t he kno wledge t hat - 84- informs t heir unde rstanding of t he va rious g roups, c an l ead t o t he ov erlooking o f valuable information by police and the delay in the recognition of these crimes as serial. As a consequence i nformation from the f irst in cident of w hat b ecomes a s eries o f related crimes can be lost and impact on the investigation of subsequent crimes. In Chapter 2 the issues relating specifically to the investigation of serial crimes will be discussed. There are a range of investigative issues that are of consideration to this the sis. The a bility of inve stigators to identify c ases of s erial c rime ha s be en considered as having a direct impact on t he progress of an investigation. Such an issue may also be compounded by the type of victims in such offences which can hinder the ability of police to link such crime. The use of technological aids and other techniques will also be discussed in relation to their role in the investigative process. - 85- CHAPTER 2: INVESTIGATING SERIAL CRIME There ha ve be en r ecent s cientific d evelopments i n pol ice w ork t hat c hallenge t he traditional approaches to policing (Tong and Bowling, 2006). In their presentation of a typology of t he l ogic driving t he way which detective work i s done, t hey identify the profiling of offenders and the use of forensic science and new information technologies as developments which have an influence on the way that detective work is carried out (Stephen T ong & B en Bowling, 2006) . T his t hesis c onsiders t he i nfluence of t hese practices on the inve stigation of vi olent s erial crimes in Australia. The way i n which investigative t ools a nd t echniques a re ut ilised i n t he i nvestigation of t hese c rimes, particularly criminal profiling and forensic evidence will be examined. Additionally, w ith t he i ntroduction of ne w m odels of pol icing c ome investigative tools to enhance the investigative capabilities. This new model is referred to as int elligence-led po licing, a nd r efers t o t he us e of i nformation t echnology a nd scientific methods in the analysis of the crime scene (Tilley, 2003) and in investigative practice. T he concept o f intelligence-led pol icing and its r elevance to s erial c rime investigation will a lso be a ddressed in this c hapter. It will be de monstrated that intelligence-led pol icing pr actices c ontribute t o problems i n t he w ay in which pol ice culture informs the investigation of serial crime. As pr eviously s tated, t he prevalence of s erial c rime is di fficult to determine. This m ay be due t o t he na ture of s erial c rimes, such as s erial m urder, where t he - 86- geographic m obility o f t he of fender a nd t he a bsence of a r elationship between the victims a nd t he of fender, f urther ha mpers t his i dentification pr ocess. Due t o t hese factors, the a bility of inve stigators to detect the serial nature of t hese murders is subsequently affected. Serial r ape c ases m ay also be di fficult t o detect as r ape and sexual assault incidents are generally recognised to be under-reported crimes (Lievore, 2002). Thus investigators may not be aware that related incidents exist. This is further complicated when related crimes oc cur ac ross di fferent pol ice com mand areas- either regional or national. One description of a serial murder investigation proffered by Ogan and Alison describes it as: …combining a s it d oes the involvement a nd s crutiny of the pr ess, the complexity of m ultiple m urders, c ontradictory a nd competing i nformation, concerns around t he worries of t he local community and t he i nformation overload involved in the case (Ogan & Alison, 2005, p.24). This de scription though a ptly relevant to the in vestigation of s erial crimes today , is describing the W hitechapel or ‘ Jack t he R ipper’ m urders of t he 1880’ s. T his can be seen t o hi ghlight t hat t he challenges confronting pol ice i n t he i nvestigation of violent serial crimes remain the same more than a century on. - 87- Traditional Investigative Techniques and the Serial Problem Traditional i nvestigative t echniques, such as the int erviewing of the vi ctim’s family and friends, often provide little information in serial cases, due to factors such as the lack of victim-offender relationship and the motivation that drives the offender. In order to identify a suspect or gain information about a suspect, investigators often use interviews. T his i nformation i s una vailable w hen i nterviewing f amily a nd f riends of serial victims. Canter (2003) in a discussion of geographical profiling, comments that often i n a m urder i nvestigation, the vi ctim is often considered a s t he ‘starting poi nt’. Questions about the victim’s lifestyle and habits are posed by investigators, looking for the situation in which the victim may have encountered their killer. Due to the frequent absence of a victim-offender relationship prior to a serial offence and therefore a lack of r ecognisable m otive, the us ual s trategies em ployed by i nvestigators seldom contribute to an investigation (Fox & Levin, 1998). Investigators often need to rely heavily on information from the public. This is often a chieved t hrough m edia announcements, requesting information a bout s uch crimes t o b e r eported t o i nvestigators. Investigators mus t a lso utilise a va riety of techniques a nd t echnologies i n t heir s earch f or t he s erial of fender, a s s uch e vidence may b e t he onl y i nformation w hich c an a id a n arrest and c onviction. In m ost m urder investigations, the investigators can often gain a wealth of information from the crime scene (Rossmo, 1995). However, in many serial murder cases, there is often no known crime scene as the victims are found at a different location, referred to as the disposal - 88- location. The ‘Truro’ murders, which occurred in South Australia, illustrate this point. The offenders di sposed of t heir vi ctims’ remains a t va rious l ocations unr elated t o t he actual crime scene (Hoysted & Kidd, 2002). Scientific and Technological Advances A factor considered important in the investigation of serious crimes such as murder is the col lection of evi dence f rom t he crime s cene (Mouzos & M uller, 2001) . T he exploratory study of Mouzos and Muller (2001) considered the factors which contribute to the a bility of inve stigators to solve c riminal c ases, i ncluding t he us e of f orensic technologies. According to Mouzos and Muller, investigators consider the securing of a crime s cene o r s cenes as ne cessary t o solve a crime s uch as m urder. T he presence o f forensic s pecialists at t he cr imes s cene/s w ere also considered as an important f actor (Mouzos & M uller, 200 1). T his e nsured t hat e vidence w as pr eserved, c ollected a nd uncontaminated. C onversely, one o f the m ajor i mpediments to a murder investigation was considered to be the length of time taken for the forensic analysis to be completed (Mouzos & Muller, 2001). Improvements i n f orensic t echniques s uch a s DNA a nalysis and f ingerprint technology have i ncreased the ability o f pol ice i nvestigators t o l ink r elated c ases and thus identify cases of serial murder (Evans, 1996) or serial rape. Such techniques have become increasingly important in linking cases using evidence found at crime scenes, - 89- and could thus pr ovide a m echanism f or cl assifying r elated cases as s erial m urder or serial rape. Fingerprinting i s pe rhaps one of t he ol dest f orensic t echniques used by p olice investigators a nd h as b een us ed f or ove r one hu ndred years i n A ustralia (New S outh Wales Police, 2005). There are commonly two forms of prints. The first, patent prints, are visible to the naked eye such as fingerprints left after contact with ink, blood or oil (Genge, 2004). The second, latent prints, are the other form of prints and cannot be seen with the na ked eye, and r equire pow der o r ch emical t reatments t o make the m vi sible (Genge, 2004). According to Genge (2004) improvements in fingerprint technology has allowed f or t he de tection of f ingerprints f rom s urfaces w hich w ere pr eviously considered to be fingerprint ‘black holes’. The technology of fingerprint identification is still the onl y means of uni quely identifying an i ndividual (PMSEIC, 2000 ). The N ational Automated F ingerprint Identification S ystem ( NAFIS) w hich ha s be en i mplemented unde r t he national CrimTrac or ganisation aims to improve the a bility t o m atch a nd i dentify fi ngerprints. This system ideally aims at making information available between police organisations, which increases the effectiveness of fingerprint analysis and pos sibly i dentification in cases w here a s uspect or vi ctim is f rom a r egion or jur isdiction other t han where t he crime took place. NAFIS was declared operational in 2001, and allows police to do an immediate s earch of ex isting r ecords when processing an of fender af ter an arrest (Mobbs, 2001). - 90- In their discussion paper on serial murder in Australia, Pinto and Wilson (1990) discuss t he us e o f de veloping f orensic t echnologies. Deoxyribonucleic acid (DNA) profiling i s a m ethod t hat a nalyses m aterials s uch a s bl ood, s emen, or other hu man tissues to identify a suspect biologically once samples are given for comparison. This is a t echnology t hat i s c onsidered t o h ave i mproved t he i nvestigation i nto s erial c rimes (Pinto & Wilson, 1990). Traditional DNA testing of humans analyses the nuclei DNA. However hum ans ha ve a s econd f orm of DNA r eferred t o a s m itochondrial ( MDNA) which can be t ested when there is insufficient n uclei D NA pr esent (PMSEIC, 2000) . According t o the P rime M inister’s S cience, Engineering a nd Innovation C ouncil (PMSEIC), the acquisition of mitochondrial DNA techniques has been considered since 2000 (PMSEIC, 2000 ). However, t he recent t rial of the accused murderer of British tourist, Peter Falconio, highlighted the absence of such technology: samples were sent to t he United K ingdom for te sting by t he F orensic S cience S ervice w ho pe rformed a new D NA an alysis p rocedure (AAP, 2007 ). C onsistent w ith N AFIS, the N ational Criminal Investigation DNA Database (NCIDD) was operational from 2001 which is a central d atabase and matching s ervice of DNA p rofiles s ubmitted from a ll A ustralian police agencies (Mobbs, 2001). The an alysis o f f ibres is anot her forensic t echnique that has be en gaining increasing prominence in the criminal justice arena over the past few years. The science of trace fibre analysis is based upon the assumptions of Locard’s Principle which states that “every contact leaves a t race” (Erainçlioglu, 2004). In simple terms, every time an offender comes into contact with the victim or the crime scene, they will leave traces of - 91- themselves and take traces from the victim or scene with them. For example brushing up against the victim will lead to a transfer in fibres between both parties. However as outlined by Erainçlioglu (2004) although a trace may be pr esent there is no guarantee that t he t race w ill be de tected. O ne of t he m ost c ommon f orms of t race e vidence i s fibres (Lane, 2004). Lane (2004) identifies two types of fibres- natural and man-made. Natural fibres include those of animal, vegetable and mineral origin, while man-made fibres can be identified as natural polymers, synthetic polymers or inorganic fibres. There are several different methods that can be employed to analyse fibres. The most common method is through the use of a comparison microscope. However, due to the increase in the number of man-made fibres produced and their similarities, further testing of the se f ibres were of ten r equired. In order t o do t his, doubl e r efraction or birefringence, infra-red s pectrometry and p yrolysis g as c hromatography can be u sed. These m ethods ana lyse di fferent ch aracteristics, such as l ight or i nfra-red a bsorption, and gas components (Lane, 2004). According to the PMSEIC, “Australia has a high quality capability but limited capacity i n forensic cr iminalistics. For s ome t race m aterial t here i s a cr itical s hortage both of c apability and capacity” (PMSEIC, 20 00, p.10) . T hus although t here a re forensic techniques available to analyse trace fibres, it is uncertain to what extent these technologies are available or in use in Australia. - 92- A t echnique t hat has b een us ed for s ome d ecades i s f orensic odont ology. Forensic odontology refers to methods of identification, being the matching of teeth to pre-existing record to identify an individual; and the comparison of teeth to bite marks that may have been left by an offender (Genge, 2004). The use of forensic odontology is particularly useful in cases where remains are unidentifiable by other means such as DNA pr ofiling or vi sual i dentification. H owever one of t he pr oblems w ith f orensic odontology is that it relies on records of the victims teeth being in existence in order to gain a comparison (Evans, 1996) . Forensic odo ntology c an a lso be ut ilised t o a id investigators in identifying pos sible vi ctims pr ior to records b eing available (Evans, 1996). For ex ample, a person has been found but more information is required before police c an i dentify a po ssible vi ctim, a nd t hen forensic odont ology can pr ovide an estimate of age. Another t echnology u sed in the inve stigation of s erious c rimes is ba llistic technology, which refers to the science of firearms and bullets including the motion in which t hey t ravel (Evans, 1996) . F rom t he m arkings on f ired bul lets, i t c an be determined which weapon it was fired from because the barrel of each firearm is unique (Evans, 1996). This form of identification is used to identify a suspect and also to link crimes where the same weapon was used to commit various crimes. Although ballistic technology is well presented in the literature, in Australia few crimes are gun-related. A recent r eport i dentifying t he us e of w eapons i n hom icides i n A ustralia f ound t hat firearms were used in relatively few incidents (Mouzos & Houliaras, 2006). Less than one in ten female victims were killed with a firearm, while 19 percent of males were - 93- killed using this same weapon (Mouzos & Houliaras, 2006). The incidence of firearms to commit s erial c rimes, particularly s erial mur der, is r are. Within the s cope of thi s current study, there was only one case of serial murder that involved a firearm and the use of ballistic identification. The e stablishment of D NA da tabases i n s ome countries s uch as t he U nited Kingdom and Australia, has allowed for the linkage and prosecution of cases that were considered t o b e be yond t he i nvestigators’ ability to close. A r ecent example of t he successful use of this technique is the case of Steve Wright who was prosecuted for the murders of five women in England in January 2008 (Rayner, 2008). DNA evidence was presented to link the offender to three of the victims. DNA linkage was also evident in a series of rapes in the United States. The offender was identified after the federal DNA database l inked a D NA sample f rom t he of fender w ith evidence obt ained in at l east nine r ape cas es ac ross d ifferent j urisdictions (Snyder, 2005) . Some o f t hese c ases, as identified by Snyder (2005) date back to 1974. Improvements in the use of technologies for the processing of fingerprints, trace fibres and ballistic evi dence ha ve positively impa cted on the s olvability of c riminal cases (Evans, 1996; G enge, 2004) . S uch t echnologies may al so enable a r eduction i n the linkage blindness often experienced in cases of serial murder. The contribution of such systems to the investigation of serial crimes in Australia has not been addressed by previous research; the current study aims to ascertain how such techniques are used in cases of serial crime in Australia. - 94- Media as an Investigative Tool Due t o t heir rare a nd violent na ture, s erial m urder cases command considerable attention in the media (Fox, Levin and Quinet, 2005). Factors such as the age and sex of the vi ctims, t he num ber of vi ctims a nd t he a mount of or t ype of vi olence evi dent increases t he publ ic i nterest i n such cases and t herefore results i n increased m edia coverage (Canter, 2003). This is supported by Innes (2002) who states that crimes of an unusual na ture of ten attract i ntense m edia i nterest. These i nclude t hose t hat i nvolve high levels of violence, child victims or where the crimes are considered to be linked to other offences, thus identifying the crimes as newsworthy (Greer, 2003; Innes, 2002). The media however can have various roles extending from their main function of acting in t he public i nterest; how ever, Innes pr oposes t hat s uch interest pl aces a s ignificant amount of pr essure on s uch i nvestigations. T his pr essure i n t urn e mphasises t he ne ed for detectives to achieve a good result (Innes, 2002). According t o Innes (2002) police i nvestigating these c rimes c an be s een to collude with the media, encouraging them through the use of what he terms “rituals of display” (Innes, 2002, p.676). S uch r ituals i nclude p ress a ppeals f rom the vi ctim’s family, crime r econstructions a nd interviews with the s enior inve stigating of ficer. These practices were considered by detectives studied by Innes to have two purposes. The f irst of t hese w as t o manage t he m edia r eportage of the c rime, e nsuring tha t the coverage di d not damage t he progress o f t he i nvestigation. Detectives considered t hat the second purpose was to use the media as an investigative resource, whereby contact - 95- can be m ade w ith pot ential w itnesses a nd i nformants i n a t imely and cost-efficient manner (Innes, 2002) . The m edia’s coverage of hi ghly publ icised i nvestigations impacts upon the amount of information produced from these murder enquiries. Innes (1999) comments t hat w hen pol ice are c onfronted b y a m urder investigation that is not progressing towards identification of an offender, they may use the media through a range of strategies as part of their wider investigative practice. This use of t he m edia f or t he i nvestigation of s erial cr imes ha s al so been referred t o a s ‘police public relations’ (Chavez & Gibson, 2004; Gibson & Chavez, 2004). According t o C havez a nd G ibson (2004) an important pa rt of a n investigation into serial crimes is the management of public relations. They describe public relations during a s erial c rime inve stigation a s ha ving t hree pur poses w hich are c onsidered t o have goal-oriented functions. T he first of these is media relations, which refers to the interactions w ith t he m edia pe rsonnel. T his i ncludes t he r elease o r w ithholding of information relating to the case at hand to media outlets. This can be driven by the need to achieve a specific purpose. Chavez and Gibson describe the media relations during the investigation into the serial murderer known as the Son of Sam in New York City. In this case the media was manipulated by both the police and government in order to “facilitate t heir i nvestigation, c ontrol t he f low of ne ws t o t he publ ic, a nd i nfluence public perceptions of the case” (Chavez & Gibson, 2004, p. 13) . Thus this can be seen to obtain specific outcomes from the police-media relationship. - 96- The s econd f unction of publ ic r elations i s crisis m anagement. C havez a nd Gibson (2004) explain that this is a response to the public reaction to crimes like serial murder which gain widespread attention and create concern in the general public. The third function of public relations identified by Chavez and Gibson is to provide public information. T his c an be di sseminated i n va rious w ays a nd can t arget s pecific information at particular audiences. Public relations exercises can be driven by different motivators. In what may be seen as positive stereotyping, Ericson (1993) describes the priority designation of some cases to be investigated based on the judgment that the prioritisation of such cases was good publ ic r elations. E ricson out lines t wo t ypes of c ases generally d esignated hi gh priority due to public relations reasoning. The first are cases involving high profile or special vi ctims where p ressure is applied from the pol ice hi erarchy to prioritise them. The ot her t ype of cas e i s where t he s uspect is high pr ofile a nd c ould m ake e ffectual complaints to both the police hierarchy and the public. Media s trategies of ten used in serial murder i nvestigations, as out lined b y Gibson a nd C havez (2004), can i nclude ne ws r eleases, t elevision pr ograms, publ ic meetings, news conferences, radio announcements and media briefings. The use of such media s trategies can ho wever cr eate pr oblems f or i nvestigators. Chavez a nd G ibson (2004) highlight the di fficulties inve stigators f ace when ba lancing t he needs of t he investigation a nd t he interest of t he publ ic. A ccording t o C havez a nd G ibson, pol ice aim to receive t he maximum a mount of i nformation a bout a s eries o f crimes w hile - 97- releasing t o t he publ ic onl y a limited amount of inf ormation from the ir inve stigative files. At the same time however, they need to ensure that the minimal information that they do r elease t hrough m edia out lets adequately notifies t he public of the pr oblem posed by the serial offender without creating public panic. When divulging to the media information pertinent to the crimes, investigators often attempt to withhold specific de tails r elating to the c rimes. These unpubl icised details remain only known to the investigators and the offender and therefore are often referred to as ‘polygraph ke ys’ by C havez a nd G ibson (2004). S uch pol ygraph k eys would be us ed b y i nvestigators t o v erify i nformation t hat they m ay receive f rom a suspect; that is, in a confession. As outlined in Canter (2003) media coverage of high profile cases can be seen to place p ressure on the i nvestigating pol ice t hrough t he m edia’s uninformed speculations as to motives and possible offender. Canter discusses the murder of a high profile t elevision pr esenter i n London, w here s uch m edia s peculations w ere widespread. In this cas e t he m edia w as s een by pol ice t o be s preading s o much misinformation that they were impacting on the public’s ability to share information of relevance to the investigation. The detective interviewed stated that he felt the need to strongly reject such claims publicly in order to ensure that the public did not receive a distorted view of the ‘facts’. - 98- The i mpact of t he m edia need not have a ne gative i mpact on a n i nvestigation. Turvey (2002b) notes that the dissemination of information through the media can be a useful tool for gaining information from previously unidentified victims and witnesses. Without some of the information about the crimes being m ade available to the public through the press, members of the public who possess valuable information would not know its value and t herefore w ould not c ontact pol ice (Chavez & G ibson, 2004) . Although, a s C anter (2003) points out , t his c an a lso l ead t o i nformation ove rload f or police i nvestigators, where l arge am ounts of i rrelevant i nformation w as supplied b y members of the public, disguising the ‘crucial’ information from immediate attention. The media can also be actively used by investigators to ‘communicate’ with the serial of fender. T urvey (2002b) emphasises t he need f or i nvestigators t o m onitor t he media coverage t o obs erve i f t he offender’s behaviour reflects any of t he i nformation presented in the media. Canter sees this as indicating not only whether the offender is watching t he i nvestigation, but whether h e or she i s learning or r eacting t o that information about the investigation disseminated through the media. In addition Fox et al. (2005) discuss a c ase w here an investigator p ublicly t aunted the s erial o ffender i n the m edia, pr ovoking a response f rom t he of fender i n t he f orm of f urther of fending behaviour which led to his arrest. When media outlets gain access to confidential information that is considered as key t o i dentifying a n o ffender t his c an s eriously impact f urther i nvestigations. S uch information used t o s ecure a s coop, can be gained t hrough b ribing p eople i nside t he - 99- investigation, such as the police, for information that they can use as an exclusive scoop (Fox, e t a l., 2005) . The disclosure o f k ey i nformation can l ead t o changes i n the offender’s be haviour s uch a s r elocating f rom a n a rea he or s he learns i s be ing s taked out by police (Chavez & Gibson, 2004). The r easons or j ustifications f or s uch r eporting a re va ried, a lthough i n a ll examples di scussed the crimes c an be s een as ‘ newsworthy’ as de scribed by Innes (2002). A s tudy conducted b y C hermak (1998) of c rime s tory s alience found t hat t he number of victims af fected by a cr ime w as one predictor of t he s alience of a story in newspapers. This could thus explain why a case of serial murder, which by definition involves a number of victims, receives such detailed attention in the mass media. In his analysis o f t he m edia i n t he c ontext of m urder i nvestigations, Innes (2002) considers that hi gh pr ofile c ases result in media a ttention that s crutinises the r eputation and competence of t he i nvestigating pol ice or ganisation, be coming a s ignificant s ource of pressure on the investigation and the judicial process. - 100- Resources Another important aspect of the investigation of serial murder and serial rape is the w ay i n which resources ar e allocated. A s tudy c onducted b y M ouzos a nd M uller (2001) on factors related to the solvability of murder investigations in Australia found that one of the main issues raised by investigators was the lack of resources available. This i ncluded a l ack of a ppropriately e xperienced i nvestigators, s upport s taff, a nd analysts. Another i mportant pr oblem, which was i dentified b y M ouzos a nd M uller a s hampering in vestigations, was t he l ack of t ime de voted t o t he investigative process. This was due to the practices that an investigator can work on only one murder case at a time. T his s tudy how ever, di d not f ocus on s erial m urder or s erial c rimes i n g eneral. Thus the relevance of this study to these crimes although not directly related, may be inferred and aid in the development of a research strategy. According to Innes (1999, p.270), “wider moral conceptions of the seriousness of murder” are responsible for the substantial resources available for the investigation of such crimes. This however is seen by Innes to be dependent upon the progress of an investigation a nd t he l evel of publ ic c oncern about the c rimes be ing inve stigated. According to Innes, when an investigation is not progressing significantly, investigators are placed under increased pressure to reduce the allocated level of resources available. One w ay i n w hich t o s ecure a nd s ustain a dequate l evels of r esources t hroughout a n investigation, i s t o de velop a hi gh m edia pr ofile f or t he case t hus creating a ‘ public concern’ (Innes, 1999). - 101- In his e xamination of th e s olvability of hom icide c ases in the U nited States, Marché (1994) suggests that the allocation of police resources can be dependent upon the number of victims and the race of the victims. This is seen as being due to the level of community interest in the solution of such cases when there are a number of victims. One e xplanation i n r elation t o r ace s uggests t hat a pr eference t owards w hite vi ctims may be du e to economic f actors in a c ommunity (Marché, 19 94). E ven i n predominately non -white ar eas, Marché s urmises t hat t he com munity conc ern for sustaining its white tourist or commuter presence would be reflected in the availability of pol ice r esources as t here w ould be a greater c hance o f s olving t he case due t o the community interest (Marché, 1994). Therefore i t w as s uggested that gr eater r esources may be allocated to the investigation of c rimes a gainst w hite vi ctims due to the hi gh level of community concern in both white and non-white communities. The pe rspective on police r esources dur ing a s erial m urder i nvestigation i n Australia is cons idered b y H agan (1992) when he s tates t hat s uch cr imes r aise enormous challenges in this regard. Hagan suggests that the forming of a task force is an effective and efficient appr oach to dealing with the di fficulties tha t s erial c rime investigations pos e. T his s upport f or f ormation o f t ask f orces ha s be en c ountered b y Fleming (1994) who cites s everal s erial m urder cas es i n America w here t ask forces were considered to have performed poorly. F leming i dentifies a r ange of r easons f or this lack of performance, including the decentralised nature of some of these task forces and the poor focus demonstrated at an operational level. - 102- Intelligence-led Policing in the Context of Serial Investigations The t raditional vi ew of policing a s a c raft ha s be en e volving w ith t he i ntegration of increasingly c omplex t echnologies and s cientific m ethods i nto t he i nvestigation of crime (Stephen Tong & Ben Bowling, 2006). Tong and Bowling (2006) suggest that a new pol ice pr ofessionalism ha s e merged, c alling f or de tectives t o w ork w ith other branches in the police organisation. This has led to a more scientific approach to doing police work that requires detectives to possess specialised knowledge and expertise. The relatively n ew con cept of i ntelligence-led pol icing offers m ore e ffective ways of investigating crime. Technologies such as the development of forensic science techniques i ncluding D NA a nalysis (Crim T rac, 2008a ), and computer a pplications such as the Crime Linkage Analysis System (Mouzos, 2001), play an increasing role in the inve stigation of ma ny c rimes. However a greater e mphasis on s uch t echnology exists when the crime being investigated is serial in nature. Tilley (2003) comments tha t there i s no p recise philosophy that underpins t he notion of intelligence-led policing. Tilley states that: Intelligence-led policing is essentially about doing the practical business of policing m ore s martly, i ncorporating m odern i nformation t echnology a nd modern methods (Tilley, 2003, p.321). - 103- The introduction of intelligence-led policing is considered to be the response to calls f or m ore ef fective and cost-efficient policing pr actices (Ratcliffe, 2003) . Additionally, e xternal p ressures, such as t he growing d emands o f globalisation a nd technological enhancements and the inability of traditional reactive models of policing to respond adequately to these pressures, furthered this drive towards a smarter way of policing (Ratcliffe, 2003). Within the domain of policing, Ratcliffe (2003, p.2) asserts that in the early 1990’s in the UK, there was a recognition of reduced public confidence in police who were seen to be “losing the battle on the streets” coupled with an increase in the reliance on the private security industry. As out lined by T illey (2003), intelligence-led p olicing c an be s een i n some respects a s a r eturn t o t raditional pol icing, w here t he de tection a nd a pprehension of offenders is a priority within a policing framework which considers law enforcement as the key function of police organisations. The focus of this model is to identify offenders and either i ncapacitates t hem or ef fect changes that reduce the ir ability to commit crime. T his c an be a chieved t hrough t he us e of c omputer te chnologies, w hich Tilley states has enhanced capabilities to manage information concerning criminals and their milieu. This e nables pol ice to enact a s marter r esponse to criminal of fending (Tilley, 2003). This s ame t rend ha s be en reflected on an i nternational s cale. Intelligence-led policing is s aid t o ha ve or iginated i n t he United K ingdom and was introduced i n Australia in the late 1990s (Ratcliffe, 2003). It has been well established internationally - 104- since the year 2000 (Sheptycki, 2004). In his discussion of this model, Ratcliffe (2003) states tha t int elligence-led pol icing, a s pr ovided b y t he U K N ational Intelligence Model, f ocuses on f our m ain e lements: the ta rgeting of a ctive o ffenders; the management of crime and disorder hotspots; the application of preventative measures; and finally, the investigation of linked series of crimes. Tilley (2003) expands on t his, stating t hat i n or der t o achieve t hese f our elements of focus, intelligence-led policing is informed by intelligence products. Four types of i ntelligence p roducts are out lined b y Tilley: s trategic assessments; tactical assessments; target pr ofiles; and pr oblem pr ofiles. S trategic a ssessments i nform t he allocation of pol icing resources by pr edicting long t erm t rends that are considered t o have some predictive function, whereas tactical assessments are considered to be short term. The offender descriptions or target profiles are used to inform the police actions aimed at these offenders. Finally, Tilley identifies problem profiles as a measure for the identification of a series of offences or crime hotspots. These intelligence products can be i nformed b y a r ange of s ources w hereby i nformation i s not onl y t raditionally received but actively sought, and thus may not be in the public domain. According to Tilley, this c overt ga thering of int elligence inf ormation has be en formalised by the intelligence-led policing paradigm. As de scribed b y S heptycki (2004) police ha ve al ways u sed some form of intelligence t hrough t he gathering of i nformation or know ledge a nd t he t ranslation of such knowledge into action. This is consistent with beat police gathering knowledge at - 105- a local level although this has become increasingly difficult with urbanisation and the greater geographic r esponsibility of pol ice: thus ‘ local’ pol icing can be considered a misnomer. S heptycki not es t hat over t he pa st f ew de cades a r ange of a nalytical and investigative t echniques ha ve be en used more f requently. T hese i nclude surveillance devices a nd computer facilities thus le ading to this r eliance on intelligence-led information ga thering processes. T he m ove t owards a m ore s ystematic f orm o f intelligence gathering has been demanded by the increased sophistication and increased mobility of offenders (Tilley, 2003). According to Ratcliffe (2003) intelligence-led policing aims at attaining greater efficiency i n pol icing. H owever, w ith i ts i ntroduction, i t ha s be en c oupled w ith ot her efficiency m easures w ithin a policing organisation which can be s een t o conflict with intelligence-led pol icing. T hese m easures of accountability, as pr eviously m entioned, attempt to quantify many aspects of policing such as police response times with the aim of achi eving m easurable i mprovements (Ratcliffe, 2003) . T hese d emands f or accountability could, according to Ratcliffe, lead to a reduction in the effectiveness of intelligence-led pol icing a s i t gets l ost i n the pr ocesses of a ccountability a s responsibility f or va rious f unctions get s hifted between t hose i n t he o rganisational structure. Using Intelligence-led Policing in the Investigation of Serial Offences - 106- One aspect of intelligence-led policing is of most interest to this research. That is, the identification and investigation of a s eries of crimes linked to the one offender, a key focus identified by Ratcliffe (2003). This emphasis on t he targeting of serial offenders has been demanded by the growing recognition of such offenders. As demonstrated in the ea rlier di scussion of t he i ncidence of s erial murder and serial r ape i n Australia, serial offenders pose a number of issues for investigators which are often exacerbated by factors such as offender mobility and the victimology of such offences. In order to meet t hese de mands pl aced on i nvestigators, t he de tection of s erial c rimes led to the increased use of intelligence-led policing measures. As previously examined, Australian police organisations can be seen to utilise a number of w hat can be considered as ‘ intelligence-led’ a pproaches. T his i ncludes t he national fingerprint identification system, (NAFIS). This new system was introduced in 2000 f ollowing t he r ecognition t hat t he pr evious na tional f ingerprint s ystem ha d reached its capacity (Crim Trac, 2008b). The initial system which was designed in 1986 however, w as s uperseded b y those ut ilised i n Europe, t he U nited S tates and New Zealand. NAFIS has since b een heralded as incorporating the latest technology (Crim Trac, 2008b) . The f ormat of t his s ystem a llows f or t he e xchange of i ntelligence bot h nationally and internationally (Mobbs, 2001). Another similar national comparison and matching system is the National Criminal Investigation DNA Database (NCIDD) which is a warehouse of DNA profiles (Crim Trac, 2008a). As t he c ases o f s erial m urder and serial r ape considered by t his research ha ve been investigated over a relatively long period, being up t o fifteen years, it is expected - 107- that the pr ogressive impl ementation of int elligence-led policing w ill be vi sible in the investigative practices employed. Particular attention will be given to the incorporation of ViCLAS into police practice, a system purported to reduce the problems associated with investigating serial crimes, as will be examined below. Police Subcultures and Intelligence Gathering The integration of intelligence-led policing initiatives into police practice is not without problems. S tudies have f ound t hat s ome of t he di fficulties i n i ncorporating t he us e of technologies i nto pol icing pr actice can be ex plained within the context of the occupational culture. One study considered the i ntegration of intelligence-led policing systems i n pol ice or ganisations i n s everal c ountries, i ncluding t he U nited K ingdom, Canada, Sweden and the Netherlands (Sheptycki, 2004). Sheptycki (2004, p.32 2) comments tha t “ the int roduction of c riminal intelligence a nalysis int o the pol ice s ector casts ne w li ght on the occupational subcultures t hat c omprise i t”. S uch oc cupational s ubcultures c an b e w ithin a n a gency but s ubcultures c an a lso e xist be tween a gencies. W hilst S heptycki r ecognises t he detective and street cop s ubcultures, he al so discusses t he i ntelligence analyst. T he position of crime analyst is considered a ‘new profession’ which has been instituted in an e stablished or ganisation a nd t hus t his ne w r ole n eeds t o be a ccommodated. Sheptycki d escribes one analyst as b eing t reated inappropriately, assigned t o do tasks unrelated to the analysis of intelligence. This was considered as due to the entrenched - 108- attitudes and behaviours within the detective and street cop culture, which were unable to adapt to the needs of an intelligence-led approach. These findings by Sheptycki were supported by the Chan, Brereton, Legosz and Doran report i nto t he i mpact of i nformation t echnology on pol ice pr actices i n Queensland (2001). T hey found t hat t here were di stinct c onflicts be tween t he operational culture and the new information technology culture. One of the key issues in relation to this conflict that emerged was the perceived lack of understanding of the needs of operational police. Chan et al. (2001) also found that there was a shift in perceived status of the role of ope rational pol ice. One r espondent i n t heir r esearch e xpressed t he vi ew t hat operational pol ice w ere now s upporting i nformation t echnology a nalysts when “ they should r eally be s upporting us ” (Chan, e t a l., 2001, p.99) . T hese i ssues of c ultural conflict ar e f urther ex acerbated by i ntra-cultural r elations be tween pol ice a nd c ivilian information technology workers. Chan et al . (2001) revealed that pr oblems as sociated with data ent ry and communications tha t e merged during th e s tudy w ere a ttributed to the c ivilian information t echnology analyst’s l ack of knowledge about pol ice work. The i nclusion of c ivilians in this r ole also raised issues o f a vailability of te chnological s upport a s civilians w ere not available out side nor mal w orking hou rs unl ike ope rational pol ice working shifts. Chan et al. (2001) also found that whilst civilians could be considered - 109- useful f or f reeing up po lice t o c onduct ope rational dut ies, t he t rend t owards ut ilising civilians was devaluing the role of police. Further conflicts have been found to exist between the detective subculture and that of the inf ormation te chnology or int elligence an alyst. A s out lined b y Sheptycki (2004) there is a l evel of personal prestige that comes with a detective making a ‘ good pinch’ and as s uch can l ead t o advancements i n their careers. In contrast, analysts do not r eceive ade quate ac knowledgment of t heir r ole. A s s tated b y S heptycki hi gher status i s g iven t o c rimes t hat ha ve be en de tected r ather t han t hose t hat ha ve be en prevented, thus acknowledgment of the analysts’ contribution to crime prevention can be m inimised. A dditionally, t he oppor tunity f or car eer adv ancement ava ilable t o detectives who have received acknowledgment for a cas e is not similarly available for the analyst which can lead to the creation of friction within the organisation (Sheptycki, 2004). A r esult of s uch conflict be tween pol ice s ubcultures i s i nformation or intelligence hoarding (Sheptycki, 2004). According to Sheptycki information hoarding is a symptomatic of the culture which allocates prestige and status to those who make a good pinch. Therefore there is a motivation to hoard intelligence as it is considered as a valuable com modity t hat can be pr esented when “it can realize i ts be st r eturn” (Sheptycki, 2004, p.321) for t he hol der. T he ho arding o f i nformation m ay not b e a deliberate a ction but m ay r esult f rom a di slike f or t he pa perwork i nvolved i n intelligence-led a pproaches (Sheptycki, 2004) . This w ould be c onsistent w ith C han’s - 110- recipe knowledge which can be seen to inform police actions as a means of coping with the demands of police work (Chan, 2003a). This hoarding of information is contrary to the demands of the intelligence-led policing m odel w hich relies on t he s haring o f i nformation b etween officers a nd organisations t o pr ovide a n accurate r epresentation of t he l aw and or der pr oblem (Sheptycki, 2004) . T hus S heptycki s tates t hat t his l ack of i nformation e xchange hampers the a bility o f int elligence-led policing systems t o make an accurate and efficient analysis of the intelligence it holds. Whilst S heptycki’s r esearch w as in organisations, t he p rocess of the c ontext of int ernational p olice change i s s omething experienced b y a ll pol icing organisations tha t a re i ncreasingly reliant on int elligence-led approaches. S heptycki describes the integration of various intelligence-led approaches into police practice and the r esponse a pparent i n t he pol ice c ultural a ttitudes t o t hese ne w s ystems. It i s expected that s imilar r esponses ma y be e vident in this r esearch of serial cr ime investigations. - 111- Joining the Dots: Crime Linkage A c ommon problem for inve stigators is be ing able to identify s imilarities be tween related cases. The ina bility to discover s uch similarities is r eferred t o a s l inkage blindness as defined by Egger (1990a, pp.164-165): ...the ne ar total lack of sharing and coordinating investigative information and the lack of a dequate networking by l aw enfor cement ag encies. This lack o f sharing or ne tworking i s a pr evalent c ondition o f today’s law e nforcement officers and their ag encies. Thus, linkages ar e rarely established among geographic ar eas o f t he c ountry be tween similar c rime pa tterns o r m odus operandi. A factor which has been considered as paramount in the successful investigation of s erial cr imes is the a bility of inve stigators to link these cr imes at an early s tage (Egger, 1998) . There are s everal w ays i n which this can be achi eved. However t he efficacy of t hese m ethods r elies on a num ber of de pendent i ssues. T raditionally, a method of de termining i f s everal c rimes a re related i nvolves c omparing any evident modus operandi a nd/or s ignature (Keppel & B irnes, 2003; K illmier, 199 7). H owever this can raise difficulties when there is not the means to compare such information. Another f actor w hich affects a n i nvestigation i s t he l ack of i nformation exchange be tween the v arious pol ice areas. This means that some serial crimes could - 112- go undetected. In order to counteract the problem of linkage blindness that is associated with s erial c rimes, i ncluding how t o i dentify possible vi ctims, t here has be en a movement b y A ustralian police tow ards int elligence-led pol icing. A ustralian pol ice organisations introduced the ViCLAS system in various states, to address the problems that arise when cases occur across jurisdictional boundaries. This introduction occurred in the mid-1990’s following the highly publicised cases of Ivan Milat and John Glover in NSW and Paul C harles D enyer i n Victoria (Killmier, 1997; M ouzos, 2001) . This intelligence-led a pproach a ims a t targeting s erial of fenders a nd c losing t he gap i n t he shared knowledge amongst police organisations nationally. The Violent Crime Linkage Analysis S ystem (ViCLAS) was developed by the Royal C anadian M ounted P olice, a nd r ecords a r ange of i nformation o n a r ange o f offences i ncluding: all s olved a nd uns olved hom icides, pr edatory s exual as saults, unidentified remains, outstanding missing persons and non parental abductions (Putt & Higgins, 1997) . T he term “ outstanding mis sing pe rsons” r efers to those s till mis sing after 60 da ys or t hose t hat ha ve di sappeared u nder c ircumstances c onsidered t o be suspicious (NMPU, 200 5). The i nformation c omes f rom an “ analysis of victimology, suspect i nformation, m odus ope randi, f orensic and be havioural da ta” (Mouzos, 2001, p.6). The data recorded can then be com pared to determine if any c ases share s everal common factors, according to a range of criteria set b y the analyst. According to Putt and H iggins (1997) the Vi CLAS includes data r elating to hom icides and pr edatory sexual a ssaults f rom 1975 i n a n a ttempt t o a id i nvestigators i n c learing up t hese offences and to identify possible cases of serial crime. - 113- According t o M ouzos (2001), the A ustralian ViCLAS int elligence a nalysts provide assistance to detectives with the collection and analysis of information during an investigation. ViCLAS analysts are also able to liaise with other police organisations and other a gencies i n a timely and effective m anner. This ef ficient ac cess t o relevant information is considered important to the outcome of an investigation (Mouzos, 2001). Although t here i s a n i ncreasing r ecognition of f emale s erial m urderers as acknowledged by Hickey (1997), little has been done to target this problem. As will be discussed, female s erial m urderers us ually t arget vi ctims w ho a re know n to them in some cap acity. T his pl aces l aw e nforcement at a pos sible advantage, a s t hey m ay be able to identify r elated crimes e arlier. In Australia, the Vi CLAS was impl emented in order to increase the ability of investigators to link similar crimes believed to be part of a s eries of of fences b y t he one of fender. However, the t ypes of of fences t hat ar e recorded by V iCLAS ar e r elatively un related to the t ypes of of fences t hat f emales, in particular f emale serial m urderers, w ould be l ikely t o commit. T here i s c urrently no national registry tracking the questionable deaths of children and elderly people. These two demographics are perhaps t he m ost vul nerable t o female s erial murderers; t he deaths of some groups of vulnerable victims are not easily monitored or tracked as they are the responsibility of different agencies. Although ViCLAS wa s originally ope rational i n N SW, c ooperation between this program and NSW Police has since been limited. NSW Police and Victorian Police withdrew their pa rticipation s ome t ime a go and t hus were no l onger s ubmitting cases - 114- for entry in the system. According to at least one source, ViCLAS was being actively used by Queensland police in 2007 dur ing a sexual assault investigation (Lion, 2007). However, at the time of w riting it a ppears tha t thi s s ystem h as now be en decommissioned na tionally although t here ha s b een no public r ecognition that thi s is the situation. To date this system has not been formally replaced with another s ystem and a s s uch t he onl y m eans a vailable t o l ink r elated c rimes a re t hrough the s tandard police r ecording s ystem ( such as C OPS i n N SW) and through t he us e of f orensic techniques s uch a s DNA a nd f ingerprints. A s t he V iCLAS s ystem w as ope rational during t he t imeframe c overed b y t his r esearch, t he c ontribution of t his s ystem t o t he investigation of violent serial crimes remained a consideration. While many s upport t he us e of a f ormalised l inkage a nalysis s ystems (Egger, 1990a; K illmier, 1997) Canter, M issen a nd H odge (1996) are of the opi nion that systems such as ViCLAS are too expensive. They argue that the resources allocated to such systems would be better invested in traditional investigative practices. Canter et al. state that in countries where police jurisdictions are geographically large, serial crimes will usually occur within that jurisdiction thus rendering a linkage system unnecessary. However C anter et al. focus upon t he a bility of t hese l inkage analysis s ystems t o identify an offender, offering an example of a case where the offender was not recorded in the linkag e s ystem u sed in that p articular ju risdiction. They f ail t o acknowledge, however, that linkage analysis s ystems are designed to identify other crimes or events that exhibit similar characteristics and thus may be investigated as serial. - 115- The Picture of Evil - Serial Offender Typologies and Criminal Profiling Is There a Typical Serial Offender? There are several typologies for categorising serial murderers, but only a few typologies for serial rapists, one of which has recently been amended to apply to serial murderers as w ell, a lthough there a re specific t ypologies derived for t hese o ffenders. S ome o f these typologies will be discussed along with the issues that the uses of such typologies raises. It m ust be not ed that a lthough t hese t ypologies appear t o di fferentiate c learly between t he di fferent types of s erial of fenders, no of fender c an be f itted e xclusively into one c ategory (Hazelwood & Burgess, 2001 ). Such t ypologies are o ften used by investigators and c riminal pr ofilers t o pr ovide a f ramework w ithin w hich t hey c an operate to identify a possible suspect. Such typologies, such as the serial rape typologies proposed by Groth, Burgess and Holstrom (1977) and the Massachusetts Treatment Centre Rapist Typology Version 3 (MTC:R3) (Knight, 1999; Knight, Warren and Reboussin, 1998) are derived from the motivations of t he of fender. T he t ypologies developed f or s erial m urderers a re primarily organised according to the offender’s motivational factors, and generally refer to male s erial m urderers. According to Hickey (1997), this i s t o enable a greater understanding of why serial m urderers commit t hese a cts. T he m ost c ommonly - 116- referenced typology s pecifically for m ale of fenders w as de veloped by H olmes and DeBurger (1998). However, a t ypology that has also been recognised in the academic literature i s t hat of K elleher and Kelleher ( 1998) w hich is t argeted at f emale s erial murderers and derived from a study of 100 offenders. Organised versus Disorganised The U S F ederal B ureau of Investigation (FBI) h as f ormulated a cl assification system for s erial of fenders t hat c ategorises of fenders based on t heir a ctions and a pparent motivations. This c lassification system identifies of fenders a s e ither or ganised or disorganised and is a pplicable t o a s ubgroup o f serial m urderers. For t his s ubgroup, crimes a re cons idered as m otiveless and ar e i dentified as s ex-related cr ime (Godwin, 2000). T hese ca tegories ar e us ed to aid the c riminal pr ofiling pr ocess, a s the y a re derived f rom s uch pr actices. T he t erms ‘organised’ and ‘disorganised’ a s out lined b y Ressler et al (1988), refer to both the characteristics and personality of the offender and the characteristics of the crime scene. Ressler et al. (1988) state that t he or ganised m urderer i s us ually i ntelligent, skilfully e mployed and able t o interact s ocially, of ten as a popular m ember of t heir community. They can usually sustain sexual relationships and are sexually competent. Although a s a child t hey m ay h ave h ad a r elatively s table c hildhood w ith steady financial r esources, di scipline w ould ha ve be en i nconsistent or a bsent. D uring t he commission of t he crimes whose t rigger m ay b e stress, this offender m ay ha ve be en - 117- under the influence of alcohol yet been able to control his mood. After the crime he will follow the news to determine his situation and may change jobs or move if he believes he is in danger of detection or to avoid detection. The organised murderer usually plans his c rime. H e ga ins c ontrol ove r t he vi ctim t hrough t he us e of r estraints a nd conversation. The body of the victim is usually hidden in a location separate from the murder scene and little evidence and no weapon are left nearby. In c ontrast, t he di sorganised m urderer a ccording t o R essler e t a l. (1988), i s of low int elligence, and may be e mployed in an unskilled area if e mployed a t a ll. He is unable to interact socially, cannot maintain sexual or social relationships and is usually sexually incompetent. He has had an unstable childhood, where discipline would have been harsh. At the time of the crimes he would be anxious but not under the influence of dr ugs or a lcohol. H owever, after t he c rime his i ntake of dr ugs and a lcohol m ay increase. T he c rimes are us ually unpl anned a nd s pontaneous. V iolence i s us ed t o subdue the victim. After death, the victim is usually left where he or she died without any attempt to conceal the body. The weapon is usually left with the body with a large amount of evidence. Criminal Profiling Once a case of serial crime has been identified, there is often a large number of suspects because of the e xtent of oppor tunities to of fend (Turvey, 2002a ). Therefore investigators need to utilise investigative techniques which may help reduce the range - 118- of suspects. Often, the traditional method of deduction is the most useful tool in which suspects can be eliminated due to lack of opportunity. However, in some serial crime investigations, suspects are few, although there is an enormous number of people who were capable of committing such an offence. The or igins of c riminal pr ofiling are di fficult to identify. The f irst p erson considered t o pr offer a criminal pr ofile i s fictional c haracter, S herlock Holmes, t he creation of Sir Arthur Conan Doyle (Petherick, 2005). As a tool used by detectives in the investigation of crimes, one of the first attempts to develop a profile of the offender was with the kidnapper of Charles Lindburg Jr. in 1932. This attempt was deemed to be rudimentary a nd not ve ry successful (Evans, 19 96). T he f irst s uccessful us e of t his technique came from an attempt to identify an offender following a series of bombings in N ew Y ork C ity w hich s tarted dur ing 1940 a nd c ontinued ove r a pe riod of s ixteen years (Evans, 1996; Petherick, 2005). A profile was developed by Dr James Brussel, a psychiatrist, who predicted many things about the offender, including that he would be wearing a doubl e-breasted s uit w hen a rrested. T he pr ofile pr ovided b y Brussel l ed t o the i dentification and arrest of G eorge M etesky, r eferred to as t he ‘Mad B omber o f New Yor k’ (Petherick, 2005) . T his l ed t o t he i ncreased us e of t his t ool i n t he investigation of serious crimes. The t echnique of criminal pr ofiling c an he lp reduce t he num ber of p ossible suspects or help identify a possible suspect. Criminal profiling has been defined as the technique of interpreting behavioural patterns from a crime so as to construct a list of - 119- ‘characteristics’ or a ‘profile’ of t he pr obable of fender (Kocsis, 1999) . According t o Wilson (foreword, Kidd, 2000, xi): criminal pr ofiling i s increasingly be coming a major source of attention in the popular m edia. T his i s i mportant, a s pr ofiling i s one of t he t ools t hat o ffers great promise to the investigation of not only murder, but all forms of violent and serial crime. Keppel and Walter (1999) state tha t a c riminal profile c an aid investigators to identify: the s ignature o f the offender t o enhance cr ime l inkage; w hat characteristicsphysical, be havioural or de mographical- that m ay be pos sessed b y an uni dentified offender; and ways in which the offender may be apprehended. Once an offender has been apprehended Keppel and Walter posit that t he c riminal pr ofile can be u sed b y investigators to develop useful interview techniques and identify areas where evidence may be found. There are s everal di fferent m ethods us ed to form a cr iminal pr ofile. These methods i nvolve sourcing different i nformation a nd a nalysing t his i nformation according to different frameworks. The difference in methods results in the production of c riminal pr ofiles w hich, i n s ome c ases, m ay differ t o pr ofiles f ormed us ing ot her methods. T hese v arying forms of of fender pr ofiling are of ten referred to as ‘of fender profiling’, ‘psychological profiling’ and ‘criminal investigative analysis’ (see Petherick, 2005 for discussion). There are two main ways in which a criminal profile is derived. These t echniques a re di stinct f rom t he f rameworks of i nformation us ed to pr ovide a - 120- profile. T he t wo m ethods us ed t o f orm a pr ofile a re r eferred t o as t he i nductive a nd deductive methods (Turvey, 1998). According t o T urvey (1998) inductive m ethods i nvolve t he pr ofiling of behaviour, crime scenes and victims from known behaviours, that is, derived from other similar inc idents. This method of pr ofiling is often based on statistical inf ormation derived from previous cases of a similar nature. The resulting profile is therefore based on generalised information, not all of which is specific to an individual case, which can lead to inaccuracies and may implicate innocent people more frequently. However, due to its more user-friendly nature, inductive profiling methods are the most common form of profiling. The de ductive m ethod of pr ofiling r equires t he a nalysis of s pecific c rimes, crime s cenes and victims, exclusively based on evidence obt ained -forensic and otherwise (Turvey, 199 8). This m ethod can more accu rately i dentify the s erial murderers MO and signature, enabling the linkage of other related cases. The signature must be i dentified and established i n order t o i dentity t he offender, us ing this profile. As it is based on information specific to the offences investigated, the deductive profile avoids generalisations, l eading t o greater a ccuracy of t he profile. The pr ofile can also be modified and r edeveloped as t he of fender’s b ehaviour changes and d evelops ove r time. The f ormation of s uch a pr ofile r equires t rained s pecialists who ha ve t he specialised knowledge a nd s kills t o interpret i nformation f rom t he cr ime s cene. The - 121- expertise required and the fact that it is case specific means that it is often considered as the most accurate method of profiling (Turvey, 1998). The pr ocess of of fender pr ofiling is ba sed upon t he a bility t o c lassify these offenders into understandable concepts. These categories can be as simple as sex, race or occupation, and be as complex as the methods of serial murderer and the organised or di sorganised be haviour pa tterns of t he s erial murderer o r ra pist. Criminal profiling has be come popul arised be cause i t a ppears t o pr ovide a s imple m ethod f or l aw enforcement personnel to track down serial offenders who terrorise the community. These f orms o f c riminal pr ofiling bot h rely on pr evious e xperience- that is previous acts of serial crime in the form of typologies. Once an incident of serial crime has be en r ecognised, t he t argeting o f pos sible s uspects be comes di fficult due t o t he large number of offending opportunities in the general population, thus profiling is used to r educe t he s cope of the i nvestigation. The pe rceived usefulness of t he cr iminal profiling process and outcomes, by police investigators of serial murder and serial rape in Australia will be considered at a later stage in this thesis. Geographic Profiling Another f orm of pr ofiling u sed i n t he i nvestigation of s erious c rimes i s g eographic profiling, also referred t o as ‘spatial m apping’ (Hickey, 2002). Thi s form of profiling - 122- seeks t o identify geographic s imilarities be tween related crimes. As pr eviously discussed, the distance serial offenders travel to commit their crimes varies, indicating their le vel of mobi lity (Egger, 1990a ; H ickey, 1 997). G eographic pr ofiling c onsiders these f actors of distance, t he mobility of of fenders a nd t he de mographics of t he locations of the ir c rimes (Hickey, 2002 ). G eographic pr ofiling also i dentifies ‘mental maps’, the cognitive images that the offender might have about the surroundings of his or her crimes (Hickey, 2002). According to Hickey (2002) the more comfortable the offender becomes when committing hi s or he r crimes, the f urther he or s he is w illing to travel to of fend. By mapping w here t he s erial of fender’s c rimes ha ve oc curred a nd t heir s equence, i t i s possible to identify a possible area to which the offender may be attached. This may be an area i n which he or she l ives, w orks or ha s some c onnection t o. T his c an a id an investigation by identifying an area where investigators can commence door knocks for information or to reduce their suspect sample by determining who has a c onnection to that region (Hickey, 2002). Geographic pr ofiling ha s be en credited a s a ke y investigative t echnique t hat contributing to the arrest of Joseph Thompson, a serial rapist active for 12 years in New Zealand (Canter, 2003; Manning, 1997). According to Detective Inspector Manning, it had become apparent that “traditional criminal investigation methods were unlikely to be s uccessful i n this p articular case” (Manning, 1997, p.2) , a nd research i nto t his technique ha d l ed pol ice t o de velop a pr ofile b ased on t he p rinciples o f ge ographic - 123- profiling, and c riminal pr ofiling t o s ome e xtent. C anter (2003) describes tha t w ithin three months of researching and developing this technique, the offender was identified and arrested. Manning d escribes t he us e of t his profiling t echnique a s “ an e xtremely viable a nd va luable i nvestigative t ool” (Manning, 1997, p.8) in the inv estigation of serial rape offences. Chapter Summary As described in this chapter, there are a number of issues that can be considered to have an impact on the out come of a s erial c rime inve stigation. The a bility of pol ice to identify a nd l ink i ncidents of s erial m urder a nd s erial rape a ppear t o be of upm ost importance to achieving a positive investigative outcome. However, as discussed, there are numerous other factors that can also influence the progress of an investigation. The i ncreased reliance on investigative t echnologies such a s t hose in t he forensic sphere can be seen to aid in the ability of investigators to not only link offences but a lso i dentify offenders. T echniques s uch as c riminal pr ofiling are impor tant tool s complementing traditional investigative practices. The relatively n ew mod el of pol icing, intelligence-led policing, has al so be en discussed. T his m odel has be en increasingly prevalent in policing p ractices bot h nationally and internationally. This model of policing can be seen to be influenced by - 124- police cultural knowledge and practice, particularly in the police ability to link related crimes, a supposition that will be explored in this research. - 125- CHAPTER 3: METHODOLOGY There are several gaps in the current knowledge about the investigation of serial murder and serial rape. Previous studies into the different aspects which can contribute to the investigative pr ocess h ave be en limited and few ha ve considered s erial cr ime as a separate investigative issue. As previously stated, serial crime poses several issues for investigators tha t may n ot be a pparent i n ot her investigations a nd t hus t his r esearch aims to identify how serial crime investigation can be seen to differ. Research Question The r esearch que stion directing this r esearch was how can i ncidents o f serial cr ime remain undetected for long periods of time? In order to answer this question, the use of a range of investigative tools, techniques and technologies was examined to determine how their use contributed to achieving the investigative outcome. - 126- Research Aims First and foremost, this r esearch aimed to compile and systematically analyse a previously disparate body of procedures, practices and experience-based knowledge of investigators of serial crimes. Thus, this research aimed to bridge the gap in academic knowledge of t he m ethods e mployed t o i nvestigate s erial c rimes i n A ustralia a nd t he issues and problems that arise in the practices and techniques employed. The use of a range of investigative tools, techniques and technologies in the context of serial crime investigation were considered. This has also been deliberated on within the framework of pol ice c ultural know ledge. B y e xamining t he use of i nvestigative t ools, t echniques and technologies i n the cont ext of pol ice cu lture, the ex istence o f a di ctionary knowledge pe rtaining t o s erial m urder a nd s erial r ape was identified. H ow t his knowledge is used was examined in light of the decisions made by detectives in relation to the way in which these crimes were investigated. This r esearch also aimed to ascertain how police c ulture impacts on t he investigation of t he s erial c rimes of m urder a nd r ape. It ha s be en argued t hat police culture is a mul tifaceted construct (Chan, 199 9b), t herefore consideration has be en given t o i ts va ried n ature a nd how di fferent o utcomes c an a rise f rom the c ultural influences t hat pervade and s hape pol icing. T hus t he w ay i n w hich aspects of pol ice culture r eflect on an investigation of s erial cr ime has be en identified and examined. Police pe rceptions of the i nvestigative pr ocess are important s ources of i nformation, allowing such reflections to emerge through analysis. - 127- Of pa rticular int erest is the pol ice c onstruction of vi ctimhood a nd ho w t his relates t o t he br oader arguments a bout oc cupational c ulture a nd t he ba rriers t o investigative pr actice. T he be aring of t his c onstructed know ledge on t he i nvestigative response will be c onsidered. As di scussed i n t he pr evious c hapter, international research has indicated that the victimology of serial crimes can impact on the progress of the inve stigation (Turvey, 2002d) . This r esearch attempts to pr ovide a n A ustralian perspective on this phenomenon. The research literature suggests that characteristics of convicted serial offenders can be us ed t o i dentify offenders s till a t l arge (for e xample E gger, 1998 ). T he us e of intelligence-led policing practices to identify and proceed with an investigation against serial offenders has been examined and the effectiveness of these practices, in particular investigations, was ex plored. The timeframe of t his research covers the lead up t o the implementation of s ome f orms of int elligence-led policing pr actices, specifically ViCLAS. O f i nterest w as how t hese w ere i ncorporated i nto br oader investigative practice. Methodology This research has used a qualitative research approach to address the abovementioned research ai ms and r elated issues. According t o Denzin and Lincoln (2005) qualitative research is the study of a range of empirical materials through interpretive practice to - 128- gain a n und erstanding o f a s ubject. A s t his w as a qua litative s tudy w hich s ought t o answer questions raised by the research literature surrounding the investigation of serial crimes, a grounded theory approach was utilised by this research. Developed by Glaser and Straus, grounded theory uses inductive strategies to analyse data and is considered a com parative m ethod of ana lysis (Charmaz, 2003). Grounded t heory m ethods enable the r esearcher t o a nalyse t he da ta w ithout pr ior h ypotheses, i n or der t o g enerate a theory. There were t wo methods t hat w ere em ployed i n this qua litative s tudy: cas e studies a nd s emi-structured i nterviews. T his us e of t wo m ethods, i n c onjunction w ith the extensive literature review, triangulation, was designed to ensure depth and detail of data was obt ained. A ccording t o G olafshani (2003), triangulation is a s trategy employed t o i mprove t he va lidity a nd reliability of r esearch findings, a nd c an b e defined as t he c ombining o f m ethods or d ata t o achieve t his. B urns (1997) considers that t he us e of m ultiple sources of data “makes a case s tudy report m ore convincing” (Burns, 1997, p.86) through t he c orroboration of t he findings. Breitmayer, A yres a nd Knafl (1993) state t hat t riangulation pr ovides c ompleteness a nd c onfirmation of information. According to Richards (2005) the nature of qualitative records means that there is no consistency in the detail, style, source and reliability. As such, more than one form of data should be used. This is particularly important, when the research is attempting to “learn from the data, rather than test a t heory already arrived at” (Richards, 2005, p. - 129- 35). Therefore, t riangulation provides for s ystematic comparison of data sources. It is for t hese pu rposes t hat t his m ethod w as c hosen, a s i nformation r elating t o t he investigation of s erial c rimes w as not r eadily available f rom one s ource but f rom various sources. Selection of Cases The t imeframe w ithin w hich bot h m ethodologies were dr awn was from 19 90 t o the present d ay. Due t o the r arity o f s erial m urder and serial r ape i n Australia, a br oad timeframe w as de emed appropriate i n or der t o e nsure t hat a s ufficient s ample s ize of cases from which to select was possible. This also allowed for the researcher to gain a sufficient sample for both the interviews and case studies. For incidents to be included the inve stigation must h ave occurred within the stated timeframe al though the act ual crimes may have occurred outside this period, as it was the investigation of the crimes that was of interest, not the actual crime itself. Through a ne gotiation p rocess, pe rmission t o a ccess N SW P olice c losed c ase files and personnel for i nterview were obt ained. During t his process, contact had also been made w ith South Australia P olice a nd Victorian Police to attempt to negotiate similar access. This was not successful. I was advised by South Australian police that they would not consider my research proposal. This was partially due to the very recent nature of the S nowtown case w hich at tha t time w as s till the s ubject of le gal proceedings against some of the offenders. Victorian Police advised that their approval - 130- processes would take some time- up to twelve months. As will be outlined later in this chapter, extensive de lays ha d already b een experienced with NSW Police and given time restraints it was decided that this avenue would not be pursued. Stage 1: Case Studies Incidents of serial murder and serial rape in Australia were identified for analysis based on their suitability for thematic case study analysis, based on the information available. This information included a range of documents and records that were either publicly available or accessed from NSW Police. Case studies provide information in relation to the i ssues s urrounding i nvestigative t echniques a nd pol icies. The cas e s tudy analysis provided information a bout t he pr essures on po lice t o s olve t he c rime through t he utilisation of secondary data such as biographical and media accounts of serial crimes which detail f actors r elating to the investigation. The use of cas e studies as a f orm of research allows for the focus upon several factors, resulting in often highly detailed and extensive i nformation (Neuman, 2003) . A ccording t o N euman (2003), case s tudies allow f or t he a nalytical investigation of an i ssue. F urthermore, case s tudies can help link the micro to the macro. This is consistent with the proposed use of case studies, in that the indi vidual c haracteristics of the inve stigation into specific inc idents of s erial crime ( micro) will be s tudied i n r elation t o i ts contribution t o t he und erstanding of investigative practices of serial crime as a whole (macro). - 131- There a re di fferent t echniques t hat can be em ployed to conduct case s tudy research, from obs ervations t o interviewing to doc ument a nalysis (Burns, 1997) . T he technique t hat t he cur rent r esearch utilised is d ocument a nalysis. Document analysis was f avoured here as i t was de emed the m ost ap propriate method to enable ac cess t o the case study data being sought. Similar s tudies in America examining the ps ychological a nd o ffender characteristics of serial murder have also used case study research. Giannangelo (1996) used case s tudies o f American serial ki llers to determine w hich psychological characteristics were prevalent, while Egger’s (1998) use of case studies enabled him to develop a n understanding of a r ange of i ssues r elated to serial m urder i ncluding offender characteristics and investigative issues. These case studies were conducted to provide further evidence in relation to the issues surrounding investigative techniques and policies, as many biographical accounts of serial crimes detail factors relating to the investigation. By comparing investigators’ opinions of such practices and having accounts of these practices presented by a third person, t he c onsistency and r elevance of s uch o pinions w ere gauged f rom t his da ta. Such information was also utilised as an aid in the development of interview questions as w ell a s f urthering t he f indings f rom s uch i nterviews. Themes ide ntified from the research literature and strongly r epresented in the cas e s tudy da ta w ere i ncorporated into t he i nterview s chedule i n t he hope of g aining m ore de pth i n i nformation. Additionally ma ny is sues tha t w ere p ertinent to an investigation m ay no t ha ve be en - 132- apparent i n t hese a ccounts a nd t hus t he i nterviews e nabled a richer, m ore i nclusive picture of t he i ssues relating to investigative pr ocedural and r esource i ssues. Specifically, indications a s to why pa rticular inve stigative te chniques w ere e mployed over others may not be evident from the case study data. Sample In order to determine which cases were to be included in the analysis the population of offenders who could be considered as serial had to be determined. This was achieved by i dentifying c ases i n Australia w here a n of fender or t eam of of fenders ha d be en convicted of c rimes ( rape/sexual a ssault or m urder) a gainst t wo or m ore i ndividuals. This w as a ccomplished by ut ilising th e ma terials lis ted in ‘Sources o f Information’ section which follows. As previously discussed in the introduction, there are criteria recognised by law enforcement agencies in Australia, which are used to determine whether crimes can be defined as s erial i n nature (Cook & H inman, 19 99a; M ouzos, 2000; P into & W ilson, 1990). T hese d efinitions i n c onjunction w ith t he i dentified t ypologies of s erial offenders, were used to determine the suitability of cases for inclusion in the population of known convicted serial murderers or serial rapists in Australia. These were presented in Tables 1 a nd 2 i n t he i ntroduction. The de cision t o us e bot h t he formal definitions and the typologies as guidelines in this research, stemmed from the number of problems that arise w hen relying on traditional de finitions of s erial c rime a lone. Additionally - 133- cases s uch as t he S nowtown murders which are recognised as ‘ serial’ i nternationally would ot herwise ha ve b een e xcluded. A s discussed pr eviously, there ha s be en m uch critical di scussion of s ome of t hese ba sic d efinitional f rameworks (Fox, e t a l., 2005; Hickey, 1997). Cases were excluded if there were inadequate convictions recorded, if offenders were de emed t o b e ‘ mass’ ki llers, pr ofessional or c ontract ki llers, o r i f t hey di d not conform to the definitional criteria or typologies previously discussed. For instance, the case of P eter D upas w as or iginally excluded f rom t he popul ation a s onl y one m urder conviction had been obtained. This however changed in 2004 when he was convicted of a second murder and is currently awaiting trial for a third (AAP, 2006). From the ten recognised incidents of serial murder that have occurred during the stated timeframes, five of these cases were selected for inclusion in this research. There were a number of reasons why particular cases were selected as outlined in Table 3. - 134- Table 3: Selected Incidents of Serial Murder for Case Study Analysis Offender Name No. of Victims State Reason for Inclusion Paul Charles Denyer 3 VIC • Availability of doc uments a nd Kathleen Folbigg 3a NSW • Availability of doc uments a nd records • Example of investigation from state other than NSW records • Example of female serial murderer John Wayne Glover 6 NSW • Availability of doc uments a nd Ivan Marko Milat 7 NSW • Availability of doc uments a nd records, including cl osed case f ile from NSW Police • Example of investigation in NSW records • Example of investigation in NSW • Example of m issing pe rsons investigation Snowtown a 12 SA Convicted of the manslaughter of a fourth child • Availability of doc uments a nd records • Example of investigation from state other than NSW • Example of m issing pe rsons investigation Three of these o ffenders com mitted their crimes in New South Wales, na mely John Glover, Ivan Milat, and Kathleen Folbigg. Paul Denyer committed his crimes in the s tate of V ictoria while the s eries of c rimes r eferred t o of ten a s t he S nowtown murders oc curred i n S outh A ustralia. T hese c ases ha ve be en chosen on t he ba sis t hat the investigations occurred within the determined timeframe and based on the amount of i nformation a vailable on e ach c ase. Although there were s ome other high pr ofile serial murder cases that occurred during the timeframe were considered by this current - 135- research, the cases were excluded based on the lack of available information about the investigation. These ca ses that f ormed the s ample were al so selected due t o the differences be tween each of t he cas es: t he vi ctims, the of fender, the i nvestigations at different poi nts of t ime b y different pol ice f orces a nd t he w ay i n which t he m edia portrayed the investigation of these crimes. By conducting a comparison of these different cases, it was considered possible to gain a greater understanding of how various aspects of a serial crime can influence the w ay in w hich t he crimes a re i nvestigated. A s h as be en di scussed i n pr evious chapters, the phe nomenon of f emale s erial cr ime of fenders ha s be come i ncreasingly recognised. Therefore i t w as al so considered important t o include a cas e of a f emale serial m urderer, t o de monstrate how t raditional pol icing t echniques could ove rlook these i ncidents. T he num ber of c ase s tudies c hosen ha s be en s upported b y previous studies of s erial c rime tha t aimed t o identify the i mpact of s pecific of fender characteristics on issues of int erest to those r esearchers (Egger, 1984; G iannangelo, 1996). As illustrated in the review of the research literature, it appears that serial rape is at the very least as prevalent as serial murder in Australia. Therefore, the inclusion of such cases was considered as paramount to understanding the investigative techniques employed by A ustralian police, as t hese cr imes a re considered t o present comparable issues to investigators. F our c ases of s erial r ape w ere i ncluded in this c ase s tudy research, all of which were committed in NSW. The selection of cases was based upon - 136- the quantity and quality of information available on each and it was coincidental that all cases o ccurred i n NSW. As discussed pr eviously, cases of s erial r ape rarely gain t he same coverage as serial murder through biographical accounts and the like. Sources of information i ncluded newspaper ar ticles and records, t he r ichness of da ta f rom t he newspaper articles w as often not com parable to that available for s erial murder cas es. The of fenders o f i nterest t o t his r esearch w ere Wayne Anthony Trindall, D avid G len Johnson, Anthony James Daley and Bruno Presta. Table 4: Selected Incidents of Serial Rape for Case Study Analysis Offender Name No. of Victims State Reason for Inclusion Anthony James Daley 7 NSW • Availability of doc uments a nd David Glen Johnson 8 NSW • Availability of doc uments a nd Bruno Presta 5 NSW • Availability of doc uments a nd Wayne Anthony Trindall 6 NSW • Availability of doc uments a nd records records, i ncluding partial closed case file from NSW Police records, i ncluding c losed case f ile from NSW Police records, i ncluding partial closed case file from NSW Police As can be seen in Table 4, all cases chosen were from NSW. This was primarily due to the quantity and quality of the information available about these cases. - 137- Sources of Information In order to conduct case study research, a variety of materials were drawn on to provide details of these crimes. When conducting a document analysis, there are two forms of evidence, being documents and records (Hodder, 2003). Documents and records can be distinguished b y t he pu rpose f or w hich t hey w ere c reated. As out lined b y H odder (2003) records ar e of ficial t exts and refer t o a f ormal and often legal t ransaction, whereas doc uments are texts t hat are f ormed f or pe rsonal us e. Examples of r ecords include m arriage ce rtificates and drivers lic ences w hilst doc uments a re in formal te xts such as letters, memos, and field notes (Hodder, 2003). Documents The doc uments u sed in t his r esearch included publicly a vailable biographical te xts, newspaper articles, radio transcripts, television documentaries and previous case study research, de pending upo n t he a vailability o f s uch r esources. In hi s di scussion of c ase study m ethodology Burns (1997) contends t hat w hile s ome doc uments m ay not be entirely accurate or unbiased, they provide a means to corroborate evidence from other sources. B urns a lso s tates t hat doc uments c an pr ovide greater d etail a bout e vents a nd issues than would otherwise be obtained from interviewees. - 138- General Internet s earches us ing s earch engines s uch as G oogle and Alta V ista were also used to obtain information on cases of interest, although care was needed to ensure the accuracy of the information presented. It is interesting to note that there was little inf ormation available on the Internet on Australian cases w ith the ma jority of sources referring predominately to cases in the United States and the United Kingdom. One A merican-based website, Tru tv ( www.trutv.com) for merly known a s C rime Library (www.crimelibrary.com) and C ourt T V ( www.courttv.com), does i nclude accounts of some of the more notorious Australian cases of serial murder such as that of Ivan Milat, John Glover and the Snowtown murders. Other Internet sites provide brief accounts l acking de tail. Many of t he A ustralian Internet s ites w ere m edia-based and related to television programs, media reports and the like. However incidents of serial rape w ere largely absent f rom s uch s ources w hether t hey b e A ustralian or A merican cases. Information on serial r ape cas es w as l argely d rawn from ne wspaper a rticles, biographical texts, and in some cases court judgements, as indicated in Table 5 below. The us e of Internet s ources c an identify i ssues r elating t o the credibility a nd reliability of the information presented, as there are few regulations to ensure that the information c onforms t o a ny a cademic s tandard (Scott, 1996) . In a n a rticle a dvising how to assess Internet sources, Scott (1996) states that when considering a source one should: … C heck reliability a nd c redibility by v erifying the author, his or her affiliation, date, and the source of publication; check perspective by assessing - 139- biases p resented in t he information o r its s ource; a nd c heck t he pu rpose by determining its scope, coverage, and level (Scott, 1996, p.46) This pr ocess w as c arried out w hen de ciding whether to i nclude Internet s ources. Several w ere excluded w hen t he a uthor’s c redibility could not be a scertained or t he website on which the information was posted was considered to have a purpose that did not support an unbiased, academic approach. This included serial killer fan websites. The us e of s econdary s ources s uch as t he doc uments out lined a bove ha s been accepted as app ropriate materials f or c ase s tudy r esearch. In pa rticular, Giannangelo and K eppel a nd Birnies (2003) have pr eviously accepted t his us e of s econdary materials in studies of serial crime. The recent study by Mouzos and Muller (2001) also used secondary m aterials. A ccording t o Egger (1990b) media r eports ar e a comprehensive source for gaining information of serial crimes which have been used in several studies into serial crime including his own (Egger, 1984 cited in Egger, 1990b). Newspaper indices were used to identify and obtain relevant newspaper articles for t he s elected cases of focus. The i ndices of contribution i nclude Lexis.com and t he Proquest ANZ Newsstand and the databases of t he Sydney Morning Herald, the Daily Telegraph, ABC Online and BBC News Online. Some ar ticles w ere from t he author’s own collection. Similar to the review of Internet sources, the newspaper articles and other media sources w ere j udged on their credibility. A l arge proportion of t he newspaper articles - 140- quoted t he r elevant pol ice or ganisation a s t heir s ource and t hus t hese articles w ere deemed to be cr edible. This w as r einforced by similarly w orded articles appe aring at the s ame time f rom va rious ne wspapers indi cating tha t th e inf ormation came f rom a published m edia r elease f rom pol ice. T his a pplied t o bot h br oadsheet and tabloid newspapers, w hich of ten pr esented t he s ame i nformation i n t heir r eports, a lthough stylistically they may have differed. Documentaries and biographical t exts w ere as sessed similarly. M any of t he authors to such texts were journalists who had conducted their own research in addition to the inf ormation gleaned from inve stigators th emselves. Although the writing s tyle would vary- some more sensationalist than others- the basis for credibility came from the s ources t hey cited. S ome of t he doc umentaries i ncluded i nterviews w ith t he detectives and consultants that had been part of t he investigative team on these cases. Therefore the information obtained from these was deemed to be credible. Records A number of databases were also used to obtain court decisions relating to the cases of interest. These databases include the Australasian Legal Information Institute, CaseLaw NSW a nd Lawbook O nline. T hese d ecisions often r elated t o t he appeal a gainst convictions or s entences a nd t hus w ere s elective i n t he amount of case s pecific information that was detailed within. As information regarding the investigative process - 141- was of i nterest, these m aterials only c ontributed if a poi nt r egarding the inve stigation had been questioned during an appeal. Access t o official pol ice records i n t he form of closed case f iles and Briefs of Evidence from N SW P olice w ere obtained i n a number of ways. A l ist of c ases of interest was forwarded to the research contact at NSW Police who then forwarded these on to NSW P olice A rchives. The cases t hat w ere av ailable at N SW P olice A rchives were t hen relayed and access a rranged at t he s torage f acility. For a num ber of cas es, such files were not present at NSW Police Archives and the actual location could not be determined. One case, that of Ivan Milat, was not present at NSW Police Archives and for s ome t ime t he e xact location of t he f iles c ould not be de termined. Initially it w as advised that t he ent ire cas e cl osed case f ile ha d been lost al though there w ere reassurances t hat i t w as ‘ on disk’ how ever not a t t he N SW Police A rchives t herefore access would need to be at another location. While I expressed a willingness to review this case file at any location at which it would be available, access was not forthcoming. It was la ter thou ght that the cl osed case f ile h ad been ‘found’ w ith a f aculty w ithin NSW P olice and although access w as t o be a rranged through t he r esearch contact at NSW Police, this never eventuated. It is still unclear as to why this was the case. After 18 months of waiting and due to time restrictions on conducting the methodology, this was not pursued any further. Access to other files, usually Briefs of Evidence, were gained primarily through another r esearch contact at t he B ehavioural C rime U nit of N SW P olice. This cont act - 142- had been conducting their own PhD research into similar cases of serial crime and thus had been i n possession of a num ber of r elated files. These f iles w ere accessed at t he NSW P olice H eadquarters at P arramatta w here t he Behavioural C rime U nit i s ba sed, although there were extensive delays in gaining approval to access the files once a new unit c ommander w as i n place. M any of t he files accessed were br ief an d incomplete with documents, reports or pages of information absent. Other Briefs of Evidence were available to view f rom the le ad investigator a t the loc ation that the y w ere c urrently positioned at. It was noted with interest at the time that for some cases there appeared to be no closed case file or Brief of Evidence in existence. Whilst one would assume that they h ad e xisted a t one t ime, t hey w ere now c onsidered e ither l ost or simply ‘held somewhere unknown’. Due t o t he highly c onfidential na ture of t hese f iles, doc uments c ould n ot be removed and all files had to be viewed at their location. Prior to gaining access to any documents, U WS E thics A pproval w as r equired b y NSW P olice ( See Appendix A ). NSW P olice di d not ha ve an ethics r eview pr ocess i n place, instead accepting t he veracity of the processes of the university ethics review s ystem. In order to allow for later analysis of the information obtained from these files, notes were taken for coding and analysis. The records obtained from NSW Police would be accurate and credible sources of inf ormation relating to the inve stigative techniques tha t contributed t o t he investigative outcome, and with Briefs of Evidence, to the prosecution. However, these - 143- sources would not p rovide a c omplete r epresentation of aspects of t he i nvestigation. Information relating to inve stigative te chniques tha t w ere not p resented i n t he prosecution of the case would not be included in Briefs of Evidence. Also missing from such s ources w ould b e i ndications a s t o w hy s ome c ases w ere not i nvestigated at t he time they were reported. Such sources would also not detail opinion as to why specific methods were not used, or the usefulness of particular investigative methods. Availability of Sources Table 5 details the various sources utilised to inform this research. Not all sources were available for all cases, for example court decisions were not accessible for all relevant cases. This contributed to a disparity in the amount of detailed information available on comparable cases of serial crime. Table 5: Sources of Information Used in Case Studies Sources of Information Documents Offender Name Snowtown* Milat Folbigg Glover Denyer* News Articles Y Y Y Y Y Documentary Y Y N Y Y Book 3 2 1 1 1 Book Chapter 3 4 1 Y 2 Web 1 1 1 1 1 Trindall Johnson Daley Presta Y Y Y Y N N N N N N N N N N N N - - 144- Records Police Court Case Brief of Decisions File Evidence Y N N Y N N Y N N N Y Y Y N N N N Y Y N Y Y Y Y N N N * P olice cas e f iles o r B riefs o f Evidence w ere n ot s ought f or t hese o ffences as t hey o ccurred i n s tates o ther t han NSW an d w ould h ave r equired s eparate U WS E thics A pproval an d r esearch ap proval f rom t he s tate police organisations The sources out lined a bove w ere not ex haustive. Several ne wspaper articles w ere excluded due t o t heir comparable content t o ot her articles from t he s ame print da y or commercially r elated paper. Some books w ere also excluded due t o t he lack of ready access t o these s ources over prolonged p eriods of t ime m aking a comprehensive thematic analysis of the source not possible. Case Study Analysis The a im of thi s c ase s tudy analysis w as to identify if t he ke y t hemes t hat ha d been identified from the research literature relating to the investigation of serial crime, were evident in the documents and records obtained on e ach case. The identification of the types of resources and techniques utilised by police during a serial crime investigation was of importance. In or der t o c onduct t hese c ase s tudies w ith uni formity, a c oding s ystem w as established ba sed on t he r esearch l iterature t o identify i ssues t hat t he records and documents m ight a ddress. S imilar t o a s urvey q uestionnaire, t his f ramework of br oad thematic clusters set out the issues and themes related to the aims and questions of this thesis and this framework was consistent across all eight case studies (see Appendix B for e xample of t hematic c lusters). S uch a f ramework a lso a llowed f or ease i n c oding - 145- variables i ndentified as t hemes f rom t he r esearch l iterature, in the da ta a nalysis s tage which will be discussed below. The i nformation obt ained f or e ach c ase s tudy was a nalysed i n or der t o determine i f cer tain characteristics w ere pr esent i n relation to the of fender and the crime, and the references to investigative practices, issues and practice. The information derived from these case studies of serial crime w ere ana lysed and used to contribute t o the development of interview questions, as details about the investigation are discussed in such sources. The information from the case studies was essential to addressing the a ims of thi s s tudy, s pecifically in regard to the us e, efficiency and accuracy of the various techniques and technologies considered, such as criminal profiling. Methodological Issues: Case Studies One of t he m ain p roblems t hat ar ose f rom the r esearch was t he contradictory information t hat w as pr esent. A lthough t here w as a general c onsensus of i nformation across the various sources, a few occasions existed where information from one source contradicted or w as ab sent f rom a nother. This ma de it di fficult to determine the accuracy of information, as it was unknown if this additional information was the result of poe tic l icence of a uthors or j ournalists or good i nvestigative j ournalism. Any information that could not be confirmed by at least one other source or by the interview - 146- data w as not c onsidered b y this r esearch. T his di d not a pply t o records w hereby information in Briefs of Evidence and the closed case files was often not disclosed to the media or other outlets as they are official records of NSW Police. Another issue which had to be addressed when using this methodology was the lack of availability of records. At the outset it was hoped that coronial reports and court transcripts coul d be ac cessed. H owever co ronial findings w ere of ten de emed not available to the public and court transcripts were too expensive to obtain for all cases reviewed. A s pr eviously di scussed, s elected c ourt t ranscripts w ere a vailable f rom t he Australasian Legal I nformation Institute website (www.austlii.edu.au) however m any of t hese were r elated t o a ppeals and t herefore of ten di d not contain the de pth of information about the investigation of these crimes required for this research. However i t w as l ater f ound t hat t he i ncidents of s erial m urder a nd s erial r ape were not presented to the Coroner therefore access was not required. Additionally, from reviewing and analysing the cour t t ranscripts t hat w ere av ailable i t be came cl ear t hat there was very little information relating to the actual investigative processes employed, other t han i f t hey contributed t o t he pr osecution c ase. E xcluded f rom t ranscripts w as information c oncerning the r ange of i nvestigative pr actices ut ilised r egardless of t he contribution that these had to the outcome. Therefore it was decided that as information from this source was limited it would not be used further. - 147- The r ecords av ailable f rom N SW P olice, B riefs of E vidence and closed cas e files, a lso pos ed s ome pr oblems. T he B riefs of E vidence onl y i nclude i nformation relating to the investigation that can be used to obtain a conviction. Other information regarding t he i nvestigative pr ocesses em ployed that f ailed to culminate i n a us able result is not included in these files. For example if police had interviewed a number of people who were unable to provide any relevant information, this fact is not included in a Brief of Evidence, resulting in an incomplete picture of the investigation. Similarly, the c losed case f iles accessed were no t al ways comprehensive with some ‘ files’ c ontaining the B rief of E vidence a nd a s mall a mount of a dditional paperwork. Other f iles were s imply i ncomplete, w ith some o f t he c ontents of a box appearing to be missing or some of the boxes/files to be absent. It was also difficult to gain access to some files. There were a number of cases that had been identified to me as being in the possession of the Behavioral Crime Unit. However, when I spoke to the contact with the unit, the files were not always accessible. As her research had been on the us e of criminal pr ofiling i n t hese i nvestigations, t his doc torial r esearch m ay h ave been seen to compete with her own research interests. As was illustrated in Table 5, not all cases were available to be reviewed. The records for these cases were not stored at the NSW Police Archives, and there was no available s ystem t o determine w here s uch cas es would be. This i ssue ho wever i s not unique t o NSW P olice. Similar ex periences ac cessing pol ice files w ere r eported by - 148- Brookman (1999) in her doctoral research into the patterns of masculine homicide and violence in England and Wales. Brookman (1999, p. 48-49) states that the idea that: police murder files would be neatly filed in some secured and centralised area for the respective forces could not have been further from the truth. Therefore it ma y be s uggested tha t thi s ma y b e a c ultural p ractice existent a mongst some other police forces and not unique to NSW Police or even to Australian policing organisations. N otably a bsent in this pr esent s tudy w ere the f iles of the inve stigation into Ivan Milat. As detailed above, these were reportedly ‘lost’. However it is surmised that one reason for the unavailability of these files may be that Milat is currently under investigation for ot her cases. Therefore t here may ha ve be en concern about t his researcher gaining access to relevant materials. Stage 2: Interviews with NSW Police Detectives and Professionals Members of t he NSW P olice and ot her p rofessionals, such a s c riminal pr ofilers who have worked on serial crimes in an investigative or consultative capacity, were selected for i nterview. An i nterview format w as s elected to ascertain the police pr actices and the use of investigative tools used in the investigation of serial murder and serial rape, as s tated b y t he i nvestigators t hemselves. T he opi nions of t hese i nvestigators a re essential i n e stablishing how be neficial s uch t ools are, and if pol ice pr actices as previously outlined, impacted on the investigation and to what extent. - 149- Interviews of t he investigators o f s erious c rimes ha ve be en used i n nu merous studies. T he s tudy b y M ouzos a nd M uller (2001) that aimed to ide ntify f actors tha t impacted on the a bility of inve stigators t o s olve m urder cas es u sed t his t echnique, i n addition to using secondary resources. Many issues may not be apparent in these case study accounts and thus the interviews will enable a richer, more accurate picture of the issues. Data from interviews may also identify lacunae arising from the case studies. As outlined by Kiger (1990) media reports can be problematic sources of information, as they m ay s ensationalise r eports f or c ommercial be nefit. These s emi-structured interviews are aimed at providing a more accurate, realistic and inclusive overview of the investigative process. By comparing the investigator’s opinion of such practices and the recording of such practices with the accounts by a third person, the accuracy, depth and applicability of such opinions can be determined. Interviews were chosen as the preferred method over surveys, although surveys would have been easier to administer and analyse, particularly as s ome of t he c ases occurred in other A ustralian states. Had this research ex tended into other s tates of Australia then a survey method would have been considered. According to Fontana and Frey (2003), one of the advantages to using semi-structured interview formats is that a greater depth of data can be obtained. This was considered to be due to the qualitative nature of such interviews in contrast to structured interviews which are considered by Fontana and F rey t o be qua ntitative i n na ture. One of t he f actors c onsidered w hen choosing an appropriate methodology for this research was the higher response rate to interviews in comparison to surveys (Neuman, 2003), which had been initially expected - 150- in this research. The choice however, of semi-structured interviews was mainly due to the varying nature of investigations- inherently no two investigations are identical. As the c ases f ocused upo n oc curred ov er a pe riod of a pproximately fifteen years, investigative techniques may have changed, as too may have the technology available to investigators. Additionally, some issues of interest such as missing persons, were not apparent in all of the considered cases. Interview Method The i nterviews w ere s emi-structured i n n ature, i ncorporating bot h c losed a nd op en questions. T his f orm of i nterview s tructure allows f or f urther p robing of i ssues t hat arise during the interview process, for example when response is unclear and/or needs further e xplanation. As s tated by G ray (2004) semi-structured i nterviews pr ovide flexibility in the int erview pr ocess a llowing: q uestions to be e xcluded if ir relevant; questions to be reordered to reflect the flow of the interview; additional questions to be asked t hat ha d not be en a nticipated pr ior t o t he i nterview. The i nterviews w ere ea ch conducted ov er an a pproximate one -hour pe riod, and w ere f ace-to-face i n a ll but one interview. W hen a f ace-to-face i nterview w as not f easible, a t elephone interview w as conducted. This was at the request of the participant. The face-to-face interviews were recorded us ing an audio recorder and t ranscribed at a l ater date. The f acility t o audio record was not available at the time of the telephone interview, and notes were made of the participant’s responses. - 151- Construction of Interview Schedule The construction of the interview schedule was informed by the research questions and the pr eliminary ana lysis of t he cas e s tudy da ta. The com mon themes t hat w ere identified f rom t he c ase s tudies w ere i ncorporated i nto t he i nterview s chedule ( see attachments t o Appendix A), which w as ba sed o n the f ramework of t hematic cl usters developed for t he case s tudy an alysis. Due t o the va rying n ature o f s erial cr imes and their inv estigation, an i nflexible i nterview s chedule w ould not ha ve b een pr acticable nor effective in extracting the desired information relating to the issues of interest. This is c onsistent w ith Gray’s (2004) contention t hat s emi-structured int erviews a llow f or greater flexibility in both the way in which questions are asked and which questions are deemed relevant to the interviewee. As can be seen in the interview schedule, themes were loosely addressed by the questions t o a llow f or the i nvestigators experience and pe rception t o e merge. For example, if the case study data analysis identified a particular investigative technique as making a c entral c ontribution t o the i nvestigation, t he que stion w ould a sk a bout techniques i n general. S uch a format also a llowed f or m e t o follow up and i dentify specific issues that were raised in the case study analysis but not by the interviewee. - 152- Participants Recruitment Non-probability s ampling p rocedures were em ployed f or t his qua litative as pect o f t he research, e mploying t he s nowballing a nd pur posive s ampling t echniques (Sarantakos, 2005). T his m ethod of s ampling was chosen as t his r esearch did not ai m to g ain a representative out come. Indeed, a s pr eviously discussed, t he na ture of vi olent s erial crimes precludes the pretence of ‘representativeness’ due to its varying and rare nature. Participants were drawn from the research population employed by NSW Police currently or previously, or who were identified by NSW Police to have contributed to the investigation of s erial c rime. It ha d been originally pl anned t hat pa rticipants c ould be identified from the s pecific cl osed case f iles; however, due t o the ex tensive de lays i n accessing such files, this was not possible. Two methods were employed to recruit participants. The first method identified participants through ‘snowballing’. Snowballing is a process whereby one key individual is us ed t o l ocate ot her individuals w ho m ay b e s uitable r esearch pa rticipants i n t he research (Bernard, 2000). According to Bernard, this method of sampling can often gain access to and ‘open doors’ to ‘elites’, who are those whose opinion ‘really counts’. This process proved to be successful where one contact led the researcher to four participants, - 153- one of whom also provided another participant. There w ere a num ber of ot her persons recommended by participants in this sample; however these persons were unavailable or did not meet the criteria for participation in this research- such as t heir role i n a s erial crime investigation preceded the study timeframe. This method of sampling also enabled contact with former members of NSW Police who otherwise would not have been easily identifiable or c ontactable. However, w hile a num ber of ‘ elite’ i ndividuals w ere identified t hrough t his m ethod of s ampling ve ry f ew a greed to participate. T his w as i n contrast t o B ernard’s (2000) belief t hat t he us e of t his s ampling pr ocedure ‘ opened doors’ to the elite. The s econd m ethod w as t o di rectly t arget t he research popul ation of current NSW P olice t hrough a memorandum w hich w as c omposed a nd di stributed a cross t he network vi a E mail t hrough t he N SW P olice i nternal c ommunications pr ocesses. T he memorandum out lined t he r esearch a nd requested t hat a nyone w ho ha d w orked i n a ny capacity on a case of serial rape or serial murder within the given timeframe, to contact the researcher. Although this was considered to be the most direct method of identifying compliant participants, this resulted in very few responses. Only one of these was from someone who had worked on a case of serial crime. The other responses were from those in a n e ducation r ole w ithin N SW P olice w ho w ere i nterested i n r eceiving n ews o f t he outcome of t his r esearch. T his method w as considered as a pur posive s ampling technique a s out lined b y S arantakos (2005), whereby the di stribution of t he memorandum targeted a s pecific gr oup, w ith the unde rstanding t hat t heir ex periences would be relevant to the research. - 154- Sample Six interviews were conducted which represented the investigations of five incidents of serial murder and serial rape. Of these, five were face-to-face interviews while the other was c onducted ove r t he t elephone. T he f ive f ace-to-face i nterviews w ere w ith NSW Police detectives who had been the lead investigator of their respective cases. The sixth participant w as a f orensic a nthropologist w ho w orked i n a c onsultative capa city w ith NSW Police, providing services consistent with a criminal profiler. As t his r esearch’s aim was t o collate d ata t hat would generate a s aturation of qualitative da ta r ather th an attempting to create a ny s ort of s tatistical or qua ntitative outcomes, there was not the intention to interview large numbers of individuals. It was anticipated that int erviews w ould be obt ained from a s ma ny s uitable p articipants a s possible ( for e xample c hief i nvestigator, s enior f orensic t echnician). A lthough t he detectives int erviewed were all c onsidered the chief inve stigator/senior inve stigating officer, t here w as no r esponse f rom ot her m embers of t he i nvestigative t eam. In particular, participation from f orensic t echnicians w as s pecifically r equested in the memorandum to NSW Police members, however none was forthcoming. This may have been pa rtly du e t o t he general r esponse f rom t he F orensic S ervices G roup of N SW Police, w ho w ould not a cknowledge t he research a pproval g iven b y NSW P olice. Attempts to gain additional a pproval me t w ith resistance during the r esearch process and due to time constraints further attempts were not made. - 155- Confidentiality and Consent Prior to the interviews, participants received a consent form and Information Statement detailing the aims of the research (Appendix A). Participants were informed that the y could withdraw from the s tudy at any time. Participants were given sufficient time to read the Information Statement, ask the r esearcher an y qu estions t hey m ay ha ve ha d and c onsider t he c onsent f orm. Interviews w ere conduc ted at a t ime and place convenient to the participant and suitable for the purpose. The forensic anthropologist received verbal advice of the Information Statement and gave verbal consent. Participants w ere given the opt ion t o r emain anonymous i f t hey p referred. T his was outlined in the Information Statement and the consent form (Appendix A). It ha d initially be en envisaged that s ince s erial c rime inve stigations invol ved large te ams of personnel, a l arger num ber of i nterviews w ould be obt ained. U nder t hose c onditions, anonymity could have been maintained with greater ease as several members of the one task f orce could be i nterviewed. None of t he participants how ever, requested anonymity. Had a ny of the pa rticipants r equested a nonymity, their pos itions a s le ad investigators of t heir r espective c ases w ould ha ve m eant t hat a nonymity could not be maintained. Additionally, a s th e pa rticipants predominately inve stigated different incidents of s erial crime f rom one another, anonymity would ha ve be en di fficult t o maintain as the case itself would have identified them. - 156- Most participants did make a num ber of comments that were “off the record” or could onl y be us ed i f i t w as not a ttributed t o t hem. As s omeone not a ssociated w ith NSW Police or the investigation, it demonstrated that whilst they were comfortable in making these statements to me with the confidence that I could be trusted in how that information was used, their concern related more to possible repercussions they could face f or airing s uch views. Due t o t he l imited number of i nterviews o btained, t hese statements were not included in the an alysis as t hey could be di rectly associated with the case and thus the participant. In hindsight, some of these statements may have been available for inclusion in the research had anonymity been a universal given from the outset and in the context of a larger research sample. Comments ma de b y ot her me mbers of te ams inve stigating inc idents of serial murder and serial rape have also been utilised. Although not participants in this study, their c omments w ere a lready pa rt of th e publ ic record thr ough media in terviews a nd documentaries. Interview Transcription Consistent w ith a g rounded t heory a pproach as i dentified b y C harmaz (2003), t he interviews w ere t ranscribed by m e as i t w as de emed more s uitable t o allow f or t he saturation of material and increasing my familiarity with the material. This also ensured that confidentiality of material was maintained as there were often comments made “off - 157- the r ecord” w hich pa rticipants w ould not ha ve be en c omfortable w ith a t hird pa rty being privy. Analysis of Case Study and Interview Data As w ith the inf ormation extracted from the c ase s tudies, the int erviews were subject t o a t hematic a nalysis. A c omputer pr ogram, Q SR N Vivo 2, s pecifically designed t o analyse qu alitative da ta w as us ed t o aid the ana lysis of the i nterview materials. This program allows the researcher to clarify concepts that emerge from the data, t hrough or ganising t he ke y t hemes o r i ssues i n c ategories or nod es (Bazeley & Richards, 2000) . T hese themes a re elicited f rom t he da ta t hrough t he recognition of patterns of information from otherwise disorganised or random information (Boyatzis, 1998). Due to the nature of the sources of information for the case studies, NVivo was not a suitable option for a large proportion of the case study materials, therefore manual coding of the various sources was required. This was done manually using the coding framework used in NVivo. Consistent w ith e mploying a gr ounded t heory a pproach t o t his qua litative research, the analysis was not conducted separately. During the analysis process of the interviews, issues were raised in relation to investigative factors such as the formation of task forces. This led to a return to the case study data to ascertain if this was evident - 158- from these sources for b oth the cases identified by the participants, but also the cases not covered her such as the Snowtown case. This narrowed the scope of the research as using methods to triangulate the data identified which of the themes recognised by the research literature were evident in the data. Methodological Issues: Interviews There were s everal difficulties experienced during this stage of the methodology. The identification of participants proved difficult as one of the contacts at NSW Police was not forthcoming with information that they were meant to provide. This led to a change in the methodology as a planned pilot study could no longer be conducted as the closed case files were not available. As participants were to be drawn from these closed case files al ternative r ecruitment m easures w ere r equired. T his pr ocess w as ve ry t ime consuming, pa rticularly as m uch time was s pent a waiting pr omised i nformation f rom the NSW Police. Ethical Considerations My intention to ascertain from NSW Police whether access to closed/completed police case files could be made available for the purpose of this study led to issues relating to confidentiality and pr ivacy. In accordance w ith t he U niversity of W estern S ydney Ethics Approval which was obtained, assurances had to be given to NSW Police in the - 159- research p roposal s ubmitted t o t hem, that the inf ormation that w as s ought f rom the se closed case files w ould be s pecific t o investigative pr actices onl y and conf idential information or i nformation of c urrent ope rational i nterest w ould be t reated w ith t he greatest respect and confidentiality (See Appendix C). As t he i nclusion o f confidential i nformation or i nformation of current operational i nterest w ould not be nefit o r c ontribute t o t he r esearch i n any w ay, t hese assurances were readily given. The recording of information of this nature or relating to the victims of these crimes was onl y recorded if it was already public knowledge. For example, a vi ctim’s na me or the e mployment of a vi ctim a s a p rostitute w as onl y recorded i f publ ic s ources s uch a s books , ne wspaper a rticles or m edia releases, a nd publicly a vailable c ourt t ranscripts ha d p reviously m ade s uch i dentifications. O nly information pertaining to the use of investigative practices, techniques and technologies were o f int erest to this r esearch. All doc uments containing confidential i nformation related to this research have been stored in a secure locked location, to which only the researcher has access. Research Limitations One of the limitations to this research is the small number of interviews conducted with current and former members of NSW Police who had investigated serial crimes. It had been aimed to conduct interviews w ith more i nvestigators how ever t here w ere a number of di fficulties e xperienced i n r ealising t his g oal. T he di fficulty experienced - 160- generally stemmed from t he t imeframe of t he cas es cons idered in this r esearch. As cases o f s erial m urder a nd serial rape t hat ha d b een investigated as far b ack as 1990 were of interest, it was found that a number of key people involved in the investigative process o f va rious c ases ha d retired f rom t he f orce and were una ble t o be cont acted. Whilst t his pr oblem ha d be en pr edicted, t he br oad t imeframe t hat ha d be en a imed a t reducing t his problem b y increasing t he num ber of cases t hat f ell within t he scope o f the study, failed to adequately address this issue. Another issue was the willingness of investigators to participate in the research. On a few oc casions, t he na mes o f i nvestigators who ha d worked t he cases s pecified were given to the r esearch contact at N SW Police. After cont act w as made i t w as determined that they either did not wish to participate in the research or were unable to make t hemselves ava ilable f or i nterviewing du e t o their cur rent cas e l oad. I r eceived explanations that such interviews about the way in which the crimes were investigated were not a ppropriate. I f ound i t di sconcerting w hen I would s ometimes s ee t he s ame people w ho us ed t hat e xcuse, g iving i nterviews with t he m edia f or doc umentaries on these cases. As pr eviously outlined, a m emorandum di stributed t hrough t he N SW P olice email communications network resulted in the id entification of one p articipant. There were no other ex pressions of i nterest t o pa rticipate i n this r esearch w hich was considered surprising as numerous detectives that were experienced in the investigation of serial crimes had appeared in television documentaries about the incidents of serial - 161- crime that they investigated. This was also reflected with those detectives approached directly t hrough the research cont act at NSW Police. D ue t o t he large num ber o f investigators who had played a role in serial crime cases, it was expected that a greater number of responses would have been received. Again, it was also hoped that forensic technicians w ho had a role i n a s erial cr ime i nvestigation w ould r espond t o t his memorandum but no such response was received. This response could be largely attributed to the researcher’s status as an outsider to t he pol ice or ganisation, a nd pol icing i n ge neral. T he di fficulties i n c onducting research within pol ice o rganisations a s an ‘ outsider’ w ere examined by R einer and Newburn (2008) who de scribe how suspicions a bout outsiders c an affect t he information that the researcher receives as restrictions are imposed. This could be due to c oncerns t hat t he pu blication of c ertain i nformation c ould “ endanger or di scourage witnesses or make specific detection methods widely known and hence less effective” (Reiner & Newburn, 20 08, p.358) . T hese c oncerns w ere r aised dur ing t his c urrent research, w here t hose i n the o rganisational hi erarchy expressed concern that t he information may be used for purposes other than research, or that information that was still of i nvestigative va lue t o a ctive c ases could be di vulged t hrough publ ication. Although assurances w ere given t hat c hecks w ould be put i n pl ace t o p revent t hese problems, the concerns raised were considered to be too great for some in NSW Police. According to Pogrebin (2002) having an inside contact in a police organisation can be helpful f or r educing s uspicions a nd t he a ssociated c oncerns, how ever t his d id not provide much benefit in this research with those that held strong concerns. - 162- Reiner a nd N ewburn (2008) contend t hat pa rticipation in initiation rituals can lead to some l evel of acceptance. These r ituals i nclude going for a d rink af ter a s hift and ot her s ocial a ctivities. H owever, a s t his c urrent r esearch di d not r equire t he presence of the researcher within a particular pol ice unit or department for any period of t ime, a cceptance c ould not be a chieved t hrough t hese m eans. T he m ajority o f t he detectives who were i nvolved i n t he i nvestigation of s erial m urder or s erial r ape incidents were part of different units or located at different police stations within NSW. Although the int erview s ample w as s maller t han originally intended, the interviews were designed to build on a nd enrich the data obtained from the case study analyses. This function was not affected. As there was a sample timeframe of over fifteen years, changes in investigative practice, the availability of forensic techniques and technologies have occurred and will occur in the future. While this research does not pretend to provide an insight into all cases of serial crime in the past or seek to predict the investigation of serial crimes in the future, there are a number of considerations stemming from the research that could aid in the future construction of serial crime investigation. As will be discussed in the following chapters, this research has i dentified areas t hat have i mpacted on the investigation. These issues will be di scussed further in relation to its i mpact on the investigation of serial crime. - 163- CHAPTER 4: CASE STUDIES – OVERVIEW AND TIMELINE OF EVENTS This cha pter pr esents the cas es selected for t he cas e s tudy m ethod. These cas es are introduced through t he presentation of an ove rview of t he case a nd a v ictimological timeline for each. This timeline is used to consider the development and progress of the investigation within the cont ext of t he cas e as a whole. The use of th e victimological events as the framework for the timeline was necessary to illustrate the investigation, or lack of, as a whole as s ome cas es w ere not i nvestigated at t he t ime t he of fences occurred. This victimological time line would differ f rom the time line of the a ctual investigation a s i n a nu mber of cases s uch as Milat, S nowtown a nd Trindall, t he investigation into these offences did not occur at the time of victimisation but up to ten years after the event. - 164- Introductions…Serial Murderers Paul Charles Denyer In 1993, P aul C harles D enyer s talked t he V ictorian s uburb of F rankston killing t hree young women and a ttempting t o a bduct a f ourth, ove r a r elatively s hort s even w eek period. T he t hree m urder vi ctims w ere vi olently stabbed t o d eath, how ever non e was subjected to a s exual as sault. Denyer w as i dentified after a pos tman made a report t o police of a suspicious person in a vehicle, which matched the description of a vehicle a police of ficer ha d pr eviously i dentified as be ing ne ar t he l ast m urder scene. As t he vehicle did not display any number plates, the police officer had recorded the vehicle’s details from the registration sticker, enabling identification of Denyer. - 165- Table 6: Denyer- Timeline of Victimological Events 14 Date Event 11 June 1993 Elizabeth Stevens, aged 18, is reported missing after failing to return home 12 June 1993 The body of Elizabeth Stevens found in Lloyd Park 8 July Roszsa Toth, 41, is violently attacked at gunpoint while walking home from work. The attacker attempts to drag her into bushland. Roszsa escapes and seeks assistance from a passing motorist. Police search the area but cannot locate her attacker. 13 July Debbie Fream, aged 22, goes missing after leaving her twelve-day-old son with a friend while she drives to the local shop for milk. Debbie Fream’s remains are found in a paddock by the farmer. 30 July Natalie Russell, aged 17, goes missing after riding her bike home from a local College. 31 July Natalie’s remains are found eight hours later in bushes near the bike track. A piece of skin, believed to be from the killer is found on her body. Paul Denyer is identified after his vehicle was noted by police as being in the area where the murder of Natalie Russell took place. Denyer is questioned by detectives at his home before being taken to the police station for formal questioning. 1 August Denyer confesses to the three murders after requests for DNA samples are made. 15 December Denyer pleads guilty to the murder of the three women and abduction of a fourth. 14 Derived from Petraitis (1995) and Swanborough (2004) - 166- Kathleen Folbigg In 1999 after the death of her fourth child, and the deaths of her three previous children, the deaths of Kathleen Folbigg’s children were investigated by NSW Police. Although the de aths of he r pr ior children C aleb, P atrick and S arah w ere attributed t o S udden Infant D eath S yndrome or na tural caus es, the c oincidence of l osing f our c hildren t o similar caus es r aised questions. F ollowing t he d iscovery of Folbigg’s di aries b y he r husband w ho p assed t hem ont o pol ice du e t o t heir r evealing n ature, f urther investigations into the d eaths as s uspicious were i nstigated. Folbigg was convicted of the murder of three of her children and the manslaughter of her fourth child. - 167- Table 7: Folbigg- Timeline of Victimological Events 15 Date Event 1 February 1989 Caleb is born at the Western Suburbs Maternity Hospital, Waratah Newcastle. Folbigg is living in Mayfield, Newcastle. 20 February 1989 Caleb, 19 days old, dies. Post-mortem is conducted and the cause of death attributed to SIDS Police officer from Mayfield attends residence to extend sympathies. Attends 3 June 1990 hospital with Craig Folbigg (husband) for formal identification Patrick is born at the Western Suburbs Maternity Hospital 18 October 1990 21 November 1990 Patrick is taken, semiconscious, to hospital. Later suffers seizures. Doctors believe that Patrick had suffered a catastrophic asphyxiating event, possibly causing the damage to occipital lobes of the brain, causing blindness and possibly seizures. 13 February 1991 Patrick, eight months old, dies. Doctor states death due to asphyxia caused by an epileptic fit on death certificate. 14 October 1992 Kathleen Folbigg decides to move house. Kathleen and Craig move to Thornton, Maitland Sarah is born at the John Hunter Hospital in Newcastle 30 August 1993 Sarah, ten months old, dies. Remains taken to Maitland Hospital. Detectives from Maitland Police Station attend scene. Do not take further action although discrepancies exist in timeframes given by Kathleen and Craig. 31 August 1993 Post-mortem conducted on Sarah. Although abnormalities noted, cause of death attributed to SIDS. Folbiggs move to Singleton or Glendalea. 7 August 1997 Laura is born in Singleton Hospital 1 March 1999 Laura dies, aged 19 months. Detective Senior Constable Bernie Ryan is assigned to review death of Laura. 19 April 2001 Kathleen Folbigg arrested and charged with the murders of her four children 21 May 2003 Kathleen Folbigg found guilty of the murders of Patrick, Sarah and Laura and the manslaughter of Caleb and the infliction of grievous bodily harm. a 15 There is some discrepancy between sources as to the exact location at this time. Derived from Benns (2003), Marks (2003) - 168- John Wayne Glover Between January 1989 and March 1990, six women, mostly elderly, were murdered and several others were sexually assaulted, molested or assaulted. These crimes occurred in the North S hore a rea of Sydney. These crimes o ccurred i n a range of areas i ncluding nursing hom es, retirement vi llages and publ ic s treets. In 1990 a s uspect, J ohn W ayne Glover w as i dentified: however he was not a rrested until a fter c ommitting hi s s ixth murder whilst under police surveillance. Glover was convicted in 1991 o f the murders of the six women. - 169- Table 8: Glover- Timeline of Victimological Events 16 Date 11 January 1989 Event Margaret Todhunter, 84 assaulted in Mosman. Offence not reported to police until December 1989 as she was a visitor to the region. Report made to Queensland police after her return to Queensland and following the media coverage of the crimes. 1 March 9 May Gwedoline Mitchelhill, 82, is murdered in Mosman Lady Winifred Ashton, 84, is murdered in Mosman. Strangled with own pantyhose 6 June Marjorie Mosely is sexually molested in a nursing home in Belrose 24 June Lillian Travey and Jean Mc Gilchrist are sexually molested in a nursing home in Lane Cove 8 August Euphemia (Effie) Carnie is assaulted in Lindfield 6 October Phylis McNeil is sexually molested in a nursing home in Neutral Bay 18 October Doris Cox, 86, is assaulted in her retirement village in Mosman 2 November Margaret Pahud, 85, is murdered in Lane Cove 3 November Olive Cleveland, 81, is murdered in the same nursing home in Belrose where 4 November Mosely was attacked North Shore Murders Task Force is established 23 November 11 January 1990 Muriel Falconer, 92, is murdered in Mosman Daisy Roberts is sexually molested in Greenwich Hospital Chatswood detectives identify John Glover as a suspect in the last assault. After being contacted by police Glover attempts suicide. Staff at the hospital give detectives the suicide note with ‘no more grannies’ written on it. 18 January Chatswood detectives interview Glover and photograph him for identification purposes End January Suicide note and photograph passed on to Detective Inspector Mike Hagan, head of the task force investigating the murders and assaults Task force detectives interview Glover. Glover put under 24-hour surveillance 19 March 1990 Joan Sinclair, 60, is murdered in her home while police had the property under surveillance for over seven hours. Glover is found in the bathtub after attempting suicide November 1991 John Glover pleads not guilty to six counts of murder on the grounds of diminished responsibility. The jury find him guilty of all counts. 16 Derived from NSW Police closed case file, Kidd (2000, 2008), Writer (1992) - 170- Ivan Marko Milat Ivan M arko M ilat w as c onvicted i n 1996 w ith the m urders of s even pe ople a nd t he attempted murder o f an other. Milat’s m urder v ictims di sappeared over a t wo year period f rom l ate 1989 t o 1992, w hile reportedly backpacking f rom S ydney t o t he southern regions of NSW. All of these individuals were reported missing to authorities within an average of one month from the date of the last known contact. Of the seven murder vi ctims, only o ne, S imone S chmidl, was t ravelling a lone. T he r emaining victims w ere tr avelling in the f ollowing pairs: Deborah E verist and J ames G ibson; Gabor Neugebauer and Anja Habsheid; and Caroline Clarke and J oanne Walters. The first r emains were unc overed i n t he B elanglo S tate F orest s outh of S ydney i n 1992. However it was over 12 months before the remains of the other victims were uncovered in the same region. - 171- Table 9: Milat- Timeline of Victimological Events 17 Date Event December 1989 Deborah Everist and James Gibson, both 19, contact family/friends for the last time 15 January 1990 25 January Deborah and James are reported missing to police Paul Onions is the victim of an attempted robbery and assault near Bowral after accepting a lift while hitchhiking DecemberFebruary 1990 A camera belonging to James Gibson is found near Galston Gorge, on the north18 January 1991 western outskirts of Sydney Simone Schmidl, 19, contacts friends for the last time, and is reported missing four days later 25 December 1991 30 January 1992 Anja Habschied, 20, and Gabor Neugebauer, 21, contact family for the last time Anja and Gabor are reported missing to the German police and Australian Embassy April 1992 September 1992 Caroline Clarke, 21, and Joanne Walters, 22, last seen in Sydney heading south The remains of Caroline Clarke, and Joanne Walters, are discovered in the Belanglo State Forest October 1993 The remains of Deborah Everist and James Gibson are discovered 12 October 1993 Task Force ‘Air’ is formed to investigate these murders 1 November 1993 The remains of Simone Schmidl are discovered. Two days later the remains of Anja Habschied and Gabor Neugebauer are uncovered nearby November 1993 Paul Onions contacts NSW Police and recounts January 1990 incident December 1993 Forensic psychiatrist Dr Rod Milton and social anthropologist Dr Richard Basham provide a profile of a suspect February 1994 Onions’ report recognised as significant to investigation Surveillance on Ivan Milat commences 22 May 1994 12 November 1994 20 July 1996 Ivan Robert Marko Milat is arrested Ivan Milat is ordered to stand trial Ivan Milat is convicted of seven counts of murder and one count of attempted murder 17 Derived from Bellamy (2004), Brown (2002), Hoysted and Kidd (2002) and Whittaker and Kennedy (1998) 18 Reported dates of the discovery of the camera vary across the various sources - 172- Snowtown Murders The four year period from 1994 t o 1998 w as dotted with the disappearances of twelve people who lived in or around the area of Snowtown in South Australia. However the absence of some of these individuals went unnoticed and unreported. In 1999 during a lengthy investigation into the disappearance of three of the missing individuals, police discovered t he r emains of e ight pe ople i n a di sused ba nk va ult. S hortly t hereafter, another three bodies were uncovered on a property associated with one of the offenders. One of the victims had been discovered several years prior but was not associated with this c ase unt il a l ater s tage. J ohn B unting, c onsidered t he pr imary m otivator, R obert Wagner, Mark Haydon and James Vlassakis were convicted of these murders. - 173- Table 10: Snowtown- Timeline of Victimological Events 19 Date Event 16 August 1994 Skeletal remains of an unidentified young male adult were discovered in a field in Lower Light 1995 Clinton Trezise is reported missing 1997 Barry Lane disappears Late 1998 Elizabeth Haydon disappears May 1999 Task Force ‘Chart’ is formed 20 May 1999 Remains of eight individuals are discovered in disused bank vault after a year long missing persons investigation 21 May 1999 John Bunting, Robert Wagner and Mark Haydon are charged with murder 23 May 1999 Remains of a person are recovered from Bunting’s former residence 26 May 1999 The remains of an additional person are recovered from Bunting’s former residence 2 June 1999 June 1999 Vlassakis is arrested and charged with murder Unidentified body discovered in 1994 is identified as Clinton Trezise from dental records 2001 Vlassakis confessed to four murders 16 October 2002 Trial of Bunting and Wagner begins 9 September 2003 Bunting found guilty of eleven counts of murder Wagner found guilty of seven counts of murder. Wagner had pleaded guilty to an additional 3 murders prior to the trials commencement The jury could not reach a verdict on a 12th charge of murder 19 Derived from Boston (2004), Kidd (2000), McGarry (2005), and Pudney (2005) - 174- Introductions…Serial Rapists Anthony James Daley Between 1998 a nd 2000 , A nthony J ames D aley attacked at l east s even women, aged between 21 a nd 40 years of a ge. T hese of fences w ere c ommitted i n t he s uburbs of Parramatta, west of Sydney, and Leichhardt in Sydney’s inner west. Daley targeted his victims after dark when they were alone, often when they were making their way home from w ork. Daley threatened his vi ctims w ith e ither a kni fe or a s crewdriver be fore sexually assaulting them. - 175- Table 11: Daley- Timeline of Victimological Events 20 Date Event 5 September 1998 A 26-year-old woman is attacked as she gets out of her vehicle at a golf club in Parramatta. Victim is threatened with a knife, dragged into nearby bushes and sexually assaulted. 31 October A 22-year-old victim is walking from a bus stop in Leichhardt, when she is threatened with a screwdriver and sexually assaulted. 5 November While walking from a bus stop, a 40-year-old victim is attacked as she reaches her house in Leichhardt. Offender produces a knife. 20 April 2000 A 30-year-old woman parks her car in the garage of her Parramatta home, the offender approaches threatening her with a screwdriver and sexually assaulting her. 12 May A 35-year-old woman drives into her garage in Westmead. As she leaves her car she is attacked and sexually assaulted by Daley who has a knife. 23 June A 25-year-old woman attempts to leave a car park in her vehicle when the offender enters the vehicle. Victim is sexually assaulted at knife point. 26 June 7 July Strike Force Sapelo is formed A 21-year-old victim is walking to her vehicle from the Westmead Railway station when she is attacked at knife point. She is dragged into the front garden of a house, where the offender attempts to sexual assault her. 14 July Daley is arrested on the Great Western Highway at Parramatta 20 Derived from NSW Police closed case file, Cowper (2000), Toy (2000) and court documents ("R v Daley," 2002) - 176- David Glen Johnson Twenty nine women were the victims of David Johnson. From 1995 t o 1997 J ohnson was convicted of crimes against 18 of these victims. Of these victims, eight related to charges of sexual assault. A majority of these women were the victims of sexual assault or attempted s exual assault whilst ot hers m anaged t o di srupt t he offender before s uch acts c ould be a ttempted. J ohnson, know n publ icly as t he ‘ Western S uburbs R apist’, would break into the victim’s residence in the early hours of the morning, usually with the i ntention of sexually assaulting hi s vi ctim. He us ually stole ite ms f rom the residences, of ten m oney a nd/or j ewellery, be fore a pproaching hi s vi ctim w hile they were as leep. Frequently, J ohnson w ould m ake s exually s uggestive c omments t o hi s victim. In response to these attacks, Task Force Ardmore was set up in early May 1997. Johnson was arrested later that month after fingerprint evidence linked him to some of the reported incidents. - 177- Table 12: Johnson- Timeline of Victimological Events 21 Date 13 November 1995 Event A 60-year-old woman sexually assaulted in home. Property is stolen from premises. Fingerprints left at scene 15 November A 56-year-old woman is indecently assaulted. Victim is spoken to. Property is stolen. 26 November Attempted assault 45-year-old. Victim is spoken to. Property is stolen from premises. 4 December Attempted assault 68-year-old. Victim is spoken to. Property is stolen from premises. 5 January 1996 Attempted assault 46-year-old. Victim is spoken to. Property is stolen from premises. 8 January Attempted assault 46-year-old. Victim is spoken to. Property is stolen from premises. 11 January Attempted assault 23-year-old. Victim is spoken to. Property is stolen from premises. 23 January Attempted assault 35-year-old. Victim is spoken to. Attempted assault 36-year-old later that same night. Victim is spoken to. 7 February Property is stolen. A 46-year-old woman sexually assaulted. Victim is spoken to. Property is 9 February stolen from premises. 17 February Attempted assault (non-complainant). Victim is spoken to. 19 February Attempted assault 61-year-old. Victim is spoken to. Attempted break and enter. Attempted break and enter at another location that same night. 21 February 1 March Attempted break and enter. Victim is spoken to. Break enter and steal Attempted sexual assault at another location. Victim is spoken to. Items from the other location are found at this scene. It is established that the offender had attempted to assault the victim at this location previously. Task Force Ardmore is set up to investigate these crimes. 12 March Indecent assault of 27-year-old. Property is stolen from premises. A torch is found at the scene. 24 April 21 Sexual assault of non-complainant victim Derived from NSW Police closed case file, Keogh (1997), Riley (1997) and Warneminde (1997). - 178- Event (continued…) Date 23 May Indecent assault of 31-year-old. Victim is spoken to. 9 June Break and enter. Offender disturbed. 31 July Break and enter. Victim is spoken to. 10 August Sexual assault of 24-year-old victim. Victim is spoken to. Property is stolen from premises. 27 August Attempted sexual assault. Victim is spoken to. Property is stolen from premises. 2 September Attempted break and enter. Victim is spoken to. 16 October Break and enter. Victim is spoken to. 2 January 1997 Break and enter. Victim is spoken to. 26 January Repeated sexual assault of 36-year-old. Victim is spoken to. Property is stolen from premises. 4 February Attempted break and enter. Victim is spoken to. 20 March Johnson is identified from a fingerprint match from the first case 21 March Search warrant is executed on Johnson’s property. Johnson evaded police. 28 March Johnson is arrested. - 179- Bruno Presta Over an eight month period in 1997, Bruno Presta targeted young males aged 15-21 in the g eneral a rea of Liverpool i n S ydney’s w est. In e ach c ase, P resta us ed a gun t o threaten his vi ctims int o complying w ith his ins tructions. His vi ctims w ere f orced or coerced into his vehicle, from where they were taken to his home and drugged, bound and sexually assaulted for s everal hour s. These as saults w ere vi deotaped by P resta. These t apes w ere di scovered by pol ice and used later i n court. After P resta ha d completed his assaults he would dump his victims at a different location to where they were kidnapped. - 180- Table 13: Presta- Timeline of Victimological Events 22 Date Event 28 March 1997 At 10.30pm a 15-year-old boy walking home from a party is forced into vehicle at gunpoint. He is taken to Presta’s home where he is drugged and sadistically sexually assaulted. 21 April A 17-year-old boy is on his way home late at night when offered a lift by the offender. Presta produces the gun and takes the boy to his house where he drugs and sexually assaults him. 12 May A 21-year-old male is jogging in the early hours when grabbed from behind and threatened with a gun. He is taken to the house where he is sexually assaulted. 29 October While waiting at a bus stop late at night in Claymore, an 18-year-old is forced into a vehicle at gun point., taken to offender’s house and sexually assaulted. Report is received by Detective Faucett who initiates investigation. 2 November A 15-year-old boy walking home in the early hours is forced into a vehicle at gunpoint. He is taken to the home where he is violently sexually assaulted. Investigation into all five crimes commences. November Identification of Presta via phone records leads to the execution of a search warrant and arrest. August 1998 22 Presta convicted is 26 charges relating to the assaults. Derived from NSW Police records, Rowlands (2000) and court documents ("R v Presta," 2000) - 181- Wayne Anthony Trindall Wayne T rindall w as convicted of t he vi olent s exual a ssaults of s ix w omen a nd t he attempted sexual as sault of a s eventh between 1995 and 2004. A m ajority of t hese offences o ccurred i n t he D arlinghurst a nd Kings C ross a reas in Sydney. T he vi ctims were p ersuaded by t he offender t o leave w ith him of ten on the pr etence of t aking photographs of t he vi ctims or pr oviding dr ugs. A m ajority of t hese v ictims w ere described a s pr ostitutes a lthough not ne cessarily w orking at t he t ime t hese of fences took place. A Task Force was initiated in 2004 after initial DNA analysis linked three offences. - 182- Table 14: Trindall- Timeline of Victimological Events 23 Date Event 22 September 1995 Victim is approached for calendar photo shoot. Taken into bushes at Queens Park Road where she is sexually assaulted. Crime is reported to Kings Cross Police. 31 May 1997 Trindall approaches victim as a customer, making an agreement to take the victim ‘home’ for services. Victim is driven to Royal National Park, Audly, where she is sexually assaulted. Crime is reported to Sutherland Police. 30 January 2000 Woman is approached for swimsuit photos. Taken to Centennial Park where she is sexually assaulted. Crime is reported to Kings Cross Police. 12 October 2003 Offender poses as a customer. Victim taken to Centennial Park where she is sexually assaulted. Crime is reported to Maroubra Police. 26 January 2004 Victim is approached and assaulted in Centennial Park. Crime is reported to Maroubra Police 17 May 2004 Victim is offered cannabis and leaves area with Trindall. Goes to Centennial Park where she is sexually assaulted. Crime is reported to Surry Hills Police. June 2004 September 2004 6 August 2004 Task Force Grahame is set up to investigate three of the offences linked by DNA Link is made to a fourth victim Attempted sexual assault is reported to Redfern Police. Victim identifies Trindall as offender. 31 August 2004 April 2006 23 Trindall is charged with first offences relating to the sexual assaults. Trindall is sentenced to a minimum of 22 years imprisonment. Derived from NSW Police Brief of Evidence and Begg and Norris (2006) - 183- Victimological Representations As discussed in Chapter 4, it has been suggested that the victimology of serial crimes can h ave an impact on h ow the offences are investigated and thus the construction of the i nvestigation. As will be demonstrated i n t he following chapters, di fferent aspects of the victimology of serial rape and serial murder offences is reflected in the cultural practices of policing organisations. Below i s an overview o f t he de mographic cha racteristics of t he c ase s tudies. The s tatistics f or s erial murder s ince 1990 i n A ustralia a re p rovided i n some c ases, providing a cont ext for the case study data. As each serial-rape case study used in the sample occurred after 1995, these statistics are also compared to all incidents of serial rape that occurred after 1995. - 184- CHAPTER 5: RESULTS- EMERGENT THEMES Thematic Analysis of Case Studies and Interview Data Several t hemes w ere i dentified from t he r esearch literature as k ey i ssues i n the investigation of serial murder and serial rape cases. A subsequent thematic analysis of the da ta de rived f rom c ase s tudies and i nterviews e xamined t he pr esence o f t hese themes i n t he i nvestigations of t he c ases of consideration. These the mes w ill be introduced in this chapter with further discussion in Chapters 6, 7 a nd 8. These themes did not ne cessarily apply across a ll c ases i dentified b y e ither of t he m ethodologies employed, but were considered significant regardless. These themes can be categorised into three different perspectives on the process of inve stigating s erial crimes in Australia. The f irst is the me thods employed to investigate t hese cr imes that will be examined in C hapter 6. T he i nfluence of po lice cultural c onstructions on t he inve stigation will b e e xamined i n C hapter 7. C hapter 8 will di scuss the the mes that e merged i n t he us e of i ntelligence-led policing p ractices and criminal profiling. The f ollowing c hapter c onsiders t he s ocial c ontext w ithin which these cr imes occurred and w ithin w hich t he pol ice i nvestigations w ere c onstructed. Crimes of a serial nature receive attention from the media which informs the public’s understanding - 185- of bot h t he of fenders a nd t he pol ice i nvestigation i nto t hese c rimes. T his c an a lso be seen to influence t he understanding of pol ice and t heir response t o t hese c rimes. Additionally, t he vi ctimology of t hese of fences can be s een t o be i nfluenced b y t he general community understanding of victims. Construction of an Investigation There are s everal com mon themes t hat em erged from t he cas e s tudies an d interviews when c onsidering t he w ay i n w hich a n i nvestigation i nto s erial c rime w as de veloped and constructed. 1. Serial Crime Recognition and Establishment of the Task Force. One of t he strongest of these themes was the concept of the ‘task force’ which appeared across the majority of serial crime investigations. The task force was identified as the key strategy to obt aining r esources. In m ost of t he c ases w here a t ask force w as d eployed, t he amount of resources at the disposal of the task force was considerably greater than that apparent for other serial crime investigations. As will be discussed, many of these task forces were set up following the recognition that a number of offences had taken place, or had possibly taken place. An i nteresting poi nt t hat emerged from t he t heme of t he t ask force w as t he question of what constituted a task force, as there was some variance in the number of - 186- personnel assigned to the individual cases. Another salient feature of the task force that transpired was how they achieved effective and efficient information management. 2. Media as an Investigative Tool. Another strong theme was how the media was used and managed in a s erial cr ime i nvestigation. A s out lined i n C hapter 2 , i ncidents of serial murder and serial rape tend to receive disproportionate media coverage, which is often sensationalist i n na ture. T he s ensationalist r eporting is f requently restricted to incidents of serial murder and although not directly of interest to this research, this was apparent here. Of greater interest were the calculated attempts by investigators to limit such reporting in some cases. A common feature across serial crime investigations was the use of the media to request i nformation f rom t he publ ic a bout t he crimes. T his w as achieved t hrough a variety of m edia out lets, a lthough t his de pended on t he s ize of t he i nvestigation. T he media was also used as a means for reassuring the community and as a means of crime prevention. 3. Contribution of forensic techniques. The extent t o w hich t echnologies s uch a s DNA testing and other forensic analyses could be seen to contribute to an investigation was al so of i nterest. It emerged that t here w ere quite va ried ex periences, pa rticularly with the extent t o w hich D NA t echnology contributed to t he out comes. A t heme t hat arose from this was one of mismanagement of evidence, of both DNA and other forms. - 187- This mis management r eflected directly on t he progression of s ome i nvestigations considered. In a ddition t o t his, a t heme t hat e merged f rom m ost i ncidents w as one where blame was attributed to one or more aspects of an investigation. This is often applied by the media through their reporting or publishing of biographical texts of the cases. From the interviews, there was a general dismissal of such allegations, often emphasised by the context in which any ‘error’ was made. Cultural Constructions of Policing This research aimed to consider the investigation of serial crimes within the context of police c ultural kno wledge a nd p ractices, a s i dentified b y C han (1999b). There w ere several significant t hemes t hat em erged from t he cas e s tudies and the i nterviews relating to how police cultural knowledge was reflected in the investigative practices. 1. Policing subcultures. The f irst of t hese w as t he ex istence of s ubcultures. These subcultures could be seen to conflict and compete during the investigation of both serial rape and serial murder offences of focus. These competing subcultures however, were not as s trongly divided a s s uggested by C han (1999b, 2003b), as will be discussed i n Chapter 7. These subcultures were identified as the ‘detective’ and the ‘specialised unit’ whereby some conflict or lack of understanding about the roles and responsibilities of each was apparent. - 188- This di sparity also existed a cross t he va rious l evels i n t he o rganisation. T his research was ch allenged b y di fferent l evels of t he or ganisation, m ainly m iddle management, w hile r eceiving s upport f rom t hose a t t he g rass-roots l evel a nd uppe r management. 2. Experience of lead detectives. Another issue that arose from the case study analysis and the interviews was the amount of experience that some of the lead detectives had at the time of the investigation. This could be seen to vary with the size of the task force or i nvestigation, w hereby the d etectives s econded t o l arge t ask forces t ended t o ha ve greater experience i n bot h t he pos ition of de tective a nd t he i nvestigation of s erial crimes. Consistent w ith this, the de tectives inve stigating inc idents of s erial r ape or in one cas e s erial m urder without t he r esources o f a t ask force t ended generally not t o have this experience. 3. Case Theories. A t heme t hat pervaded the ana lysis w as t he i dea o f f ollowing a ‘hunch’ or a case theory. Again, similar to detectives’ levels of experience, it was found that t here were t wo s trands of t hought emerging on t his i ssue. O ne cons idered following i nstinct a s a n a cceptable s trategy. T he ot her c onsidered t he ne gative consequences t hat s uch a l imited a pproach t o a n i nvestigation c an b ring. This i s presented within the context of good management of resources. - 189- 4. Knowledge. A significant the me tha t e merged from thi s r esearch was the w ay in which pol ice di ctionary knowledge i nforms police pr actice and reflects on t he investigations of serial crimes. As outlined in Chapter 1, police dictionary knowledge is shaped t hrough c ultural understandings a nd i nforms t he w ay i n w hich po lice w ork i s conducted, kno wn a s d irectory kno wledge. One a spect of t his know ledge i s how stereotypes are given meaning through cultural expectations. These stereotypes relate to specific groups of people and affect how these groups are responded to. Such stereotypes can be seen to be reflective of some myths which are common throughout t he general community, pa rticularly t hose r elating t o s exual a ssault and homosexuality. The concept of the worthless victim as proposed by Fattah (1997) was also seen to be reflected in the cultural understandings of the investigations into serial crimes at different points. An interesting f inding was the la ck of di rectory knowledge relating to s erial murder or s erial rape i nvestigations e vident i n t he i nterviews and c ase s tudies. W hile the r esearch literature s uggests t hat s ome i nvestigative t ools, such as t he m edia and crime linkag e s ystems, are us ed within a s pecific f ramework f or t he i nvestigation of serial crimes with expected outcomes, this was not evident. Thus, the knowledge gained by pr evious i nvestigations i nto serial m urder a nd s erial r ape w ere not c onsidered t o have actively informed subsequent investigations. - 190- In the discussion of dictionary knowledge of police in Chapter 1, the concept of a s pecialised language u tilised by pol ice t o express a cultural und erstanding given t o particular terms was discussed (Devery, 2003). This was not seen as a significant theme emerging from the case studies or interviews although there were some inferences that may be considered reflective of such language. The absence of specialised language in the research environment is however also considered in the discussion in Chapter 7. Intelligent Investigations and Culture There ha s be en significant r esearch in the ar ea of cr ime l inkage. As discussed in Chapter 2, the ability of pol ice to recognise a c ase of serial murder or serial rape can impact on t he pr ogress a nd out come of t he s ubsequent i nvestigation (Egger, 1998) . Every case that was considered in this research experienced a delay in the recognition of the offences as serial in nature. Some of these delays were extensive, up to ten years. All invol ved more inc idents tha n the tw o victims tha t a re required to identify s imilar crimes as serial. The reason for these delays can be seen to reflect some of the cultural understandings discussed in Chapter 7. 1. Linking Crimes Using Intelligent Tools. In C hapter 2, t he i ntroduction of intelligence-led policing practices i nto many as pects of pol ice pr actice w as di scussed. One of the focuses of this approach is to aid in the investigation of series of crimes, as identified by Ratcliffe (2003) in his discussion of the UK National Intelligence Model. In Australia the ViCLAS was introduced to provide a national intelligence-led approach - 191- to monitoring missing persons and a range of serious offences, among other things. One of the purposes of this system was to enable greater assistance in recognising a series of crimes that may be serial in nature. While this system has since been abolished, it was in place during the investigations of many of the cases studied. From the case study and interview analysis it w as clear that ViCLAS w as not being us ed as a n investigative tool , thus c ontributing to the di smissal of thi s s ystem. Although some of the s erial c rimes studied occurred p rior t o t he i ntroduction of t his system, those c rimes tha t oc curred within the op erational time frame did not us e this system in their inve stigations. Again, this c an be s een to reflect the c ultural pr actices evident in police organisations. 2. Contribution of Criminal Profiling. Another t ool t hat ha s be en used in several investigations was criminal profiling. The use of a criminal profile was seen to be as a tool to aid other investigative techniques and was seen in cases where a suspect was not readily app arent. There were how ever, mixed successes u sing thi s as a n investigative tool in the cases of interest. In addition to the use of a criminal profile, there was also some evidence that typologies were used in particular cases. Again, the use of these in an i nvestigation c ould n ot be s een t o i mpact on the inve stigative out come, as w ill be discussed in Chapter 8. These t hemes pr ovide a pi cture o f an investigative p rocess tha t i s s till individualistic. There is no evidence that once crimes have been recognised as serial in - 192- nature t hat t here ha s be en or i s any s pecific r esponse t o these crimes. It ha d been expected that there would be some change over time due to the timeframe chos en for this study but this was not evident. From the se the mes, a ty pology o f the inve stigation of s erial c rime ha s be en developed, as presented in Chapter 9. This typology reflects the ability of investigators to r ecognise a nd l ink r elated of fences, and t he i nfluence t hat s uch r ecognition ha s on the investigative practices employed. - 193- CHAPTER 6: ISSUES IN THE INVESTIGATION OF SERIAL CRIME This cha pter w ill pr ovide a pi cture o f t he i nvestigative p rocesses em ployed i n t he investigation of s erial r ape a nd serial mur der in Australia. The t ypes o f inve stigative techniques and technologies employed will also be identified. As serial rape and serial murder c annot be a ssumed t o be c onsidered a s t he s ame f rom a n i nvestigative standpoint 24, these two offence types will usually be discussed separately. As outlined in Chapter 2 one aspect of investigative practice that is of interest to this r esearch is int elligence-led policing. This c hapter w ill ide ntify emergent the mes relating to the incorporation of intelligence-led policing practices into the investigation of serial crimes as reflected in the case studies and interviews conducted. The most detailed information relating to the processes involved in investigating a serial crime emerged from the closed case file of the Glover case. This was the only record obt ained t hat w as ne ar c omplete i n c ontent. Unlike B riefs of E vidence, which only included evidentiary-related information about the investigation, the complete case file also records the processes that do not directly lead to the arrest of the offender. This is particularly pertinent in the information relating to the investigation of other suspects which i s absent from t he B rief o f Evidence. Other i nformation i ncludes t he s trategies 24 As described in previous chapters, the crime of murder is considered as more heinous than the crime of rape thus the importance attached to each can influence the construction of an investigation. - 194- employed s uch as c anvasses, pro-active pa trols, a pproaches t o i nstitutions s uch a s schools and foot patrols in the areas where the offender was thought to have identified his vi ctims. Unfortunately the G lover case i s the ol dest in terms of the t imeframe of events. Therefore it cannot reflect completely on the use of techniques introduced since the investigation of 1990. Pictorial of the Investigation A Resourceful Investigation: The Task Force An i mportant aspect of t he i nvestigation examined, i s t he w ay i t i s r esourced and t he availability of such resources. In all but one of the cases of serial rape or serial murder, a t ask force w as f ormed t o m anage t he r esources a nd s tructure t he i nvestigation. A s recognised previously in Chapter 2, a task force is considered as an approach which can effectively and efficiently handle the range of challenges arising from the serial murder or s erial r ape i nvestigation, t hrough pr oviding a c entralisation of r esources. T he following task forces were examined: • • • • Task Force Air- Ivan Milat Task Force Pulsar- Paul Denyer North Shore Murders Task Force- John Glover Task Force Chart- Snowtown Murders - 195- • • • Task Force Ardmore- David Johnson Strike Force Grahame- Wayne Trindall Strike Force Sapelo- Anthony Daley Two of the case studies did not have formalised task forces established. These were the cases of serial rapist Bruno Presta and serial murderer Kathleen Folbigg. Task Force Air- Milat An issue with the initial Milat investigation which followed the discovery o f the fi rst victims w as the w ay in w hich it w as c onstructed. According to Bellamy (2004) the investigation was set u p as individual g roups of de tectives which were w orking independently until T ask F orce A ir w as s et up a nd he aded b y S uperintendent C live Small. This task force was not formed until after the discovery of the remains of Everist and G ibson i n O ctober 1993, ove r 12 months a fter the f irst tw o victims w ere discovered. Initially, this task force was said to be 20 members strong with a mandate to i nvestigate t he de aths of t he f our vi ctims w ho ha d be en di scovered at t hat poi nt (Walsh, 1993). Walsh (1993) also suggested that the task force’s role was to consider links w ith ot her m issing b ackpackers. The di scovery o f t he r emains of S chmidl, Habscheid and Neugebauer soon after led the task force to announce publicly that they were treating these crimes as serial murder (Bellamy, 2004). - 196- It i s unkno wn w hether investigators ha d been treating these cases as s erial i n nature pr ior t o t his a nnouncement. If t hey h ad not , t his m ay ha ve ha mpered t heir investigation. T his point was emphasised b y E gger (1998), w hich w ill be a ddressed further in the discussion of crime linkage in Chapter 8. Of immediate interest is how the decision to form a task force was made and the factors to be considered. Superintendent Small w as a sked a bout w hat c onsiderations w ere m ade i n t he f ormation of t he t ask force t o i nvestigate t he m urders, which o ccurred a fter t he s econd di scovery i n t he Belanglo State Forest. He explained: There were a number of things that played on the decisions that were made. On the—one of the matters was that 12 months earlier there had been bodies found i n t he f orest area. There ha d been a police search. No f urther bodies were found but here we were 12 months later finding another body…. That’s not a criticism of the original police investigation, it’s simply saying that the size of t he ar ea t o be se arched was v ery, very l arge and we r eally ha d to make some decisions about how we were going to search the forest. And that was a big issue. This response suggests that there were clear con siderations during this time that these crimes c ould be s erial i n na ture a nd t hat t here c ould have be en m ore victims. The recognition of the limitations placed on the initial investigation, and the impact that this had on that investigation, contributed to the decision to form a task force to ensure that the r esources for t his would be a vailable. T his how ever, i n i tself, also pr esented problems as evident as Small continued: - 197- … The second thing is that the forest itself, and that in part the forest where the bodies w ere f ound w ere quite i solated a nd rough terrain. There w eren’t any s hops or c anteens or anything l ike t hat w here y ou c ould g et d rinks o f water, food, refreshments. Even toilet facilities weren’t there. So that in itself was a significant l ogistics issue…. The other thing, the t hird thing was that the local police station was, I think, around 20km or so away… so even the local resources were quite distant from where the crime scene was. And then the- another part that was relevant was the fact that whilst the inquiry was at this t ime w as esse ntially i n the B elanglo Forest t he reality i s i t l ooked as though it would be a l ong a nd protracted i nquiry a nd t hat probably a better place f or t he command c entre after the forest h ad been searched would be Sydney so we had to set up a Sydney command post as well to take over the running of the inquiry for its duration…. The other one was how many police did w e ne ed dow n t here, and w here w ould w e g et them f rom and w here would they be accommodated. The lo gistical problems of m anaging s o m any personnel w ere a l eading factor i n the decision to form a t ask force, as specified by Small. The problem with having no l ocal police r esources a nd t hat a s econd c ommand c entre w ould be ne eded means t hat t he task force could not be as centralised as would normally be expected. Small concluded his explanation by saying: - 198- …also the f act that th ere were a lready r umours f loating pa rticularly in the media but also a mong pol ice about ot her di sappearances t hat cou ld have occurred in the area. It was within the context of all those sorts of things that it w as de cided t o s et up a t askforce a nd de cisions w ere m ade a bout t he structure and the sources for that taskforce. This investigation took place over a pe riod of at least 13 months up until the arrest of Milat, although t he t ask f orce s till ha d a role in t he processes i nvolved i n t he prosecution of M ilat. According t o t he case review, t he c ost of T ask F orce Air exceeded $4 million dollars (Small, 1998). Task Force Pulsar- Denyer According to one source (Swanborough, 2004) the task force investigating the Denyer murders, T ask F orce P ulsar w as f ormed a fter t he s econd m urder. This however, may have been a strategic formality as there were no apparent changes in the way in which the cas e w as i nvestigated after t his t ask force was es tablished. S eparate t o this w as Operation R eassurance, which w as a he lp c entre s et up t o pr ovide i nformation on personal safety for women (Kidd, 2008). This crime prevention initiative was formed in response to the increased fear in the community after strong media coverage (Petraitis, 1995). This w as t he onl y example w here resources w ere s pecifically allocated as a crime prevention initiative. - 199- North Shore Murders Task Force- Glover The a ctual t ask f orce for t he G lover i nvestigation, T he N orth S hore Murders T ask Force, was comprised of a similar number of police to the Milat investigation, although initially it started with 35 detectives (Sunday Mail, 1989b). There were however, fewer supplementary p ersonnel a s w ill be di scussed i n relation t o i nvestigative pe rsonnel i n the next section. The closed case files included the reporting of the resources utilised by the taskforce at di fferent stages, recording ch anges and the reasons for these ch anges. The r esources al located to the M ilat and Glover t ask forces w ere i n stark contrast t o those dur ing t he T rindall i nvestigation. T ask F orce Grahame c onsisted of t wo detectives f rom t he S ex C rimes U nit and access to their r esources i n t he f orm of a n intelligence analyst, f or e xample. A s a cknowledged pr eviously, t his t ask f orce w as initiated after Trindall had been identified as a suspect by DNA linkages to some of the crimes. As an offender h ad already been identified, this may have limited the need to dedicate greater resources to the investigation. This may however, have been due to the nature of the victims. As suggested by Marché (1994) the allocation of resources may be reflective of the community interest in relation to the crimes. M arché identifies the race of t he vi ctim a s on e of t he de fining i ssues, but t his m ay e xtend t o those f rom marginalised groups (Marché, 1994). - 200- Task Force Chart- Snowtown Murders Task Force C hart ha d an initial br ief t o investigate t he conn ection between three missing pe rsons (Boston, 2004) in South A ustralia. While the initial membership is uncertain, B oston (2004) states t hat at i ts pe ak, the t ask force cons isted of 3 3 pol ice officers from the Major Crime, Crime Scene Examination and Missing Persons Squads of S outh A ustralia P olice. T hese of ficers w ere s upported b y administrators t o a ssist with the management of information. Similar to Task Force Air, specialists in the form of anthropologists and pathologists were consulted on this task force. Boston (2004) describes T ask Force C hart as a year-long m issing p ersons investigation. T his di stinguishes t his t ask f orce f rom t he ot her c ases of s erial m urder considered by this research, in that it did not begin as a murder investigation. It was this investigation into missing persons that led to the discovery of the r emains of victims. This is unlike the other serial murder cases where at least some of the victims had been located prior to the establishment of the task force. Task Force Ardmore- Johnson Following the attacks on 28 women in Sydney’s western suburbs, Task Force Ardmore was s et up. A ccording t o documents from t he closed case file and m edia reports, t his task f orce w as e stablished f ollowing a n i nitial i nvestigation, a nd w as di rected to re- - 201- investigate t hese of fences (Goodsir, 1997) . T his m ove f ollowed t he l oss of ke y evidence by investigators which led to some of it being destroyed (Riley, 1997). It was also s uggested t hat f orensic pr ocedures w ere n ot c arried out , w ith pol ice failing t o attempt fingerprint analysis on a torch that the offender had dropped at a crime scene (Warneminde, 1997). The allocation of these resources, however, did not occur at the time that a number of incidents were linked. According to one media report, similarities between t he f irst e ight of fences ha d be en i dentified, l eading t o t he c ompilation of a report detailing these cases which was then submitted to senior NSW Police members and the then Minister for Police (Riley, 1997). It was some 13 m onths later, following an internal investigation into the first investigation of these crimes, that a task force was dedicated to the offences (Warneminde, 1997). Warneminde (1997) describes t he t ask f orce a s consisting of 24 ‘hand-picked detectives’. This followed public assurances by the then Assistant Commissioner, Clive Small, that the crimes were “serious enough for resources to be upgraded as required” (Keogh, 1997) in the form of police and the use of technology. The closed case file also describes a further f our i nvestigators be ing a llocated t o T ask F orce A rdmore, a long with an analyst and clerical officer. Strike Force Grahame- Trindall Strike Force Grahame w as initiated following th e linkage of thr ee incidents of sexual assault t hat oc curred over a pe riod of s even years, a nd s ubsequently l ed t o t he - 202- identification of further offences that occurred over the previous nine year period (Begg & N orris, 2006) . Personal com munications w ith the l ead detective on this cas e, Detective G raham N orris, s how t hat t he t ask f orce ha d be en e stablished onl y t wo months before the suspect was identified, as the most recent victim had been known to the offender and had named him in her report to police (Norris, 2006). As outlined in the Brief of Evidence, it was the forensic evidence from this incident that was used to link it to other crimes through DNA analysis. According to Norris, it was only after the development of the task force that these linked crimes were fully investigated (Norris, 2006). A ccording t o N orris t he t ask f orce c onsisted of t wo de tectives a nd one intelligence analyst (Norris, 2006). As this task force was established after the suspect had been identified through DNA analysis, the inv estigative functions pe rformed were mor e limited than those o f other task forces considered. The main investigative function of this task force was to identify if other similar cases existed. Once similar crimes had been identified, the task force had to establish what physical evidence was available to be analysed in order to link these offences (Norris, 2006). Strike Force Sapelo- Daley Following a number of cases of sexual assault in western Sydney suburbs over a period of approximately 18 months, Strike Force Sapelo was set up (Rowlands, 2000). There were eight m embers o f t his t ask f orce, w ho h ad come f rom a s pecialised uni t (AAP, - 203- 2000). The closed case file identifies that the strike force was set up after six offences were committed. Initial media reports suggest this strike force was investigating sexual assaults in the suburbs of Parramatta and Westmead (AAP, 2000; Cowper, 2000; Toy, 2000). W hile i t w as a cknowledged b y pol ice t hat t his s cope w as l ocalised a nd t hat related crimes m ay ha ve oc curred i n other ar eas (AAP, 2000) , it was not unt il D aley was a rrested t hat o ffences f rom t he s uburb of Leichhardt were pub licly l inked (Rowlands, 2000). Critique of Task Forces Although t here i s general s upport for t he us e of t ask f orces i n a s erial c rime investigation, there ha ve be en s ome c riticisms levelled at th em, such as t hose b y Fleming (1994). Fleming identifies and discusses several serial crime cases in America, where the task forces were considered to have functioned poorly. In the main this was not a pparent i n t he i nvestigations c onsidered b y this current research. T here were however, i nitial pr oblems e vident w ith t he i nvestigation i nto t he s eries of s exual assaults committed b y J ohnson. The c ase review i ndicated t hat t he problems with t he investigation previously mentioned were to be addressed by a thorough reinvestigation. This included re-interviewing victims and witnesses. Task F orce Ardmore faced ot her c hallenges, a s i t w as w orking a round a n Internal A ffairs inve stigation into the inve stigative pr actices of th e ini tial de tectives working the case (Riley, 1997). The closed case file also acknowledges that following a - 204- public a ppeal for i nformation t hat i ncluded a vi sual de scription of t he of fender, t hree serving pol ice of ficers were nom inated as pos sible persons o f i nterest by t he publ ic. One of these was the Commander of Task Force Ardmore; this led to an investigation into those allegations and the involvement of Internal Affairs. Another i ssue r aised i n the literature b y Innes (1999) was t hat t here w as pressure t o reduce t he r esources available t o prolonged i nvestigations. T his di d no t emerge as an issue in any of the investigations from the case studies or the interviews. However i n the F olbigg case, Detective R yan did describe s ome i nitial r esistance t o instigating the investigation: Detective- I had difficulty getting a full time strike force on it. It was put in the corner of an office and you picked it up every now and then. Interviewer- So there were problems getting resources put on to it? Detective- Yeah…. I suppose there was always that- flogging a de ad horse, you'll never prove it. How much do w e throw at this you do a ll those things but everything I asked for I make sure I had the evidence to back it up. So I’d say- listen I want to go to America and England to do this, this and this, right and this is the reason why. And say no, and we'll go to the coroner and you'll have to answer questions as to why you didn't… - 205- This resistance w as ex plained by the f act at that s tage of t he i nvestigation, t here w as little evidence to substantiate that the deaths were homicides. In order to obtain expert evidence t o determine whether the de aths of t he F olbigg c hildren w ere s uspicious, Detective R yan ne eded t o c onsult e xperts i n England a nd A merica. When a sked whether there were problems obtaining resources to travel to England and America to consult with experts, Detective Ryan explains: No. Back in those da ys the w ay the pol ice service was st ructured overseas investigative t rips g ot f unded by t he D eputy C ommissioner. T hey ha d a budget a nd it w ent to the Minister t o s ign of f on. We do a n itinerary a s t o reasons to get permission. But I found absolutely no problems at all in going overseas. This variance in the attitude to the provision of resources from different levels may be an indication of the value placed on some forms of information, such as the obtainment of expert evidence. It may also reflect the priority to what was being proposed as a case of serial murder. As a task force had not been initiated in this case, it is more likely that there were not sufficient resources at a local level to support an intensive investigation based on circumstantial evidence. - 206- Investigative Personnel The number of investigative personnel on each case varied significantly. On most of the cases, accurate numbers w ere not av ailable as the r ange o f s ources available of ten recorded different numbers of personnel or gave personnel different titles that were not comparable. T his w as also due t o t he una vailability o f s ome c losed case f iles or t he incompleteness of t hose f iles t hat w ere ava ilable. As pr eviously stated some of t he records available f rom N SW P olice w ere limite d to Briefs of E vidence and therefore did not contain such information. The case with the greatest available detail was the Milat case, where accounts of the number of personnel required for different parts of the investigation were available. The r esources av ailable to the i nvestigation team, Task Force A ir, varied at di fferent stages of the investigation. It is reported by Bellamy (2004) that by the end of 1992 the team numbers had been either reduced or redirected to other cases and, just prior to the discovery of the bodies of Gibson and Everist, were practically nonexistent. From the Milat case review, Small had identified that the task force consisted of 40 personnel with additional support staff on hand to deal with physical evidence and intelligence. A pproximately 500 police were r equired to search the Belanglo State Forest and another region north of Sydney where belongings of one of the victims were found. The search of the properties associated with Milat and the subsequent arrest of Milat required 280 police personnel. These personnel were required at different times - 207- during the investigation depending on their role. Also included in the resources for this investigation were two cadaver dogs, used to aid the search of the Belanglo State Forest during the s econd search of t he ar ea (Small, 1998) . This is c onsistent w ith media reports that emerged at the time of Milat’s arrest in 1994, which identifies the task force of consisting of “some 500 pol ice, i ncluding a nucleus of 38 i nvestigators, uni formed police, forensic s cientists, criminologists a nd f resh Police A cademy graduates” (Sheather, 1994, p.26). This indicates a strong level of communication with the media about the Milat investigation. Such information was not commonly reported on i n the other serial crime investigations of focus in this current study, at a time pertinent to the investigation. H owever some bi ographical a ccounts of ten a ttempted t o de scribe t he investigative te am in hi ndsight, probably in an attempt to illustrate the investigative procedures. From t he c ase s tudy analysis, there w ere s uggestions t hat em erged relating t o the resources initially available in the Milat investigation. According to Bellamy (2004) some of the personnel who conducted the search in the Belanglo State Forest were from the Tactical Support Unit. Prior to their involvement the forest had been searched but not s ystematically o r w ith a f ormalised system. Although i t i s not pos sible t o s econd guess w hether t he i nitial s earch was ex tensive enoug h, it can be s aid that t his de lay reduced the ability of investigators to link these victims at an earlier stage. In the G lover case, the inve stigation into the murders w as de emed a m ajor investigation a nd t he N orth S hore M urders T ask Force w as s et up e ight months a fter - 208- the first murder and after the deaths of at least three (possibly four 25) victims. From a Briefing i n the cl osed case f ile, the pe rsonnel f igures are available. Similar to Task Force Air, t his t ask force was 40 members strong, with four analysts and intelligence officers a nd e ight s urveillance pe rsonnel. T here w ere a lso a n undi sclosed num ber of uniformed and plain clothes police performing patrols at targeted times. Non-traditional forms of resources were also apparent in the Milat investigation. After Superintendent Small was appointed head of Task Force Air, he formed a ‘Think Tank’ that included psychiatrist Dr. Rod Milton; Dr. Richard Basham a p sychological anthropologist; and B ob Y oung a c omputer a nalyst (Bellamy, 2004) . This w as in addition to a pr obability specialist, members from t he t ask force and the S tate Intelligence Group (Small, 1998). The Think Tank with its specialists from a range of different b ackgrounds was t here i n a s upportive c apacity t o t he t ask f orce. Superintendent Small explains that: What the pur pose of the g roup was t o discuss t he i ssues, to discuss t he options, to float ideas just to see what people thought. And that was the role of that group. In the majority of the serial murder cases considerable resources, particularly in the form of personnel, appear to have been readily available. For example the task force 25 The exact date of establishment varies between sources and may have occurred the day before or after the death of Glover’s fourth victim. - 209- during t he D enyer i nvestigation i s r eported t o ha ve be en 150 people s trong (Swanborough, 2004). The resources available to serial rape investigations are in stark contrast to those available f or s erial m urder cas es. In many of t he serial r ape cas es considered i n t his current research, it was found that such resources were either absent or their availability only came after a considerable length of time. For example, the Trindall investigation was c omprised o f onl y t wo d etectives w ho made up the ‘task force’. T ask Force Ardmore, investigating the series of sexual assaults committed by Johnson, was set up 20 months after the first three assaults. Similarly the task force into the investigation of the crimes by Daley, Strike Force Sapelo, was formed after a s imilar time period after the f irst of fence. A lthough t he num ber o f pe rsonnel on T ask Force A rdmore i s not known, one article describes an eight detective strong Strike Force Sapelo, all from the then Homicide and Serious Violent Crime Unit (AAP, 2000). Another a spect of t he t ask f orce di mension i s t hat a lthough s erial c rimes a re usually considered major investigations, this is not always reflected by the formation of a task force. The formation of the investigative team in the Snowtown investigation is not as clear as those for the other investigations discussed. From the biographical texts and articles, there does not appear to have been a task force as such in place. There are few r eferences t o t he s trength of t he i nvestigative t eam, a lthough McGarry describes that there were “…50 people from the CIB, major crime and forensic services” (2005, p.137) who attended a b riefing a fter the ini tial d iscovery of the r emains at the ba nk - 210- location. T his w ould i ndicate t hat a lthough the inve stigative te am w as n ot la belled a ‘task force’ the resources available were comparable. There w ere i ssues r elating t o the ava ilability of r esources dur ing t he P resta investigation, for which there was no task force formed. This was identified by the lead detective as a pr oblem with t he i nvestigation o f t his c ase, a s c an be s een f rom t he following exchange with Detective Faucett: Interviewer- Were there any other problems that you experienced with the case? Detective - Just th e m atter of th e r esources- we j ust di dn’t g et he lp with them. It’s the same with everything. Myself and the other… fellow, we were leading the i nvestigation. C ompared to today we’re not t hat i nexperienced but at the time were very inexperienced in plain clothes. We had both done plain clothes work but we were the most junior in the office at the time. We happened t o be on c all a t t he t ime. S ex C rimes s quad d idn’t he lp. Behavioural S ciences he lped a fter w ith s ome s tuff. Even t he bos ses di dn’t really…weren’t that interested in it…. We were supported really well by the Macquarie Fields staff as soon as we needed to but apart from that it was just another job come into office. Interviewer- Even when it was recognised that it was serial and there were a number of other offences was it still hard getting resources? - 211- Detective- Yeah- we j ust di dn’t ha ve it. D idn’t ha ve t he a dded l uxury because as it turned out it went so quickly we got the SPG in quickly which it’s not hard to get them in when there’s a firearm involved. But they’re only talking a bout one d ay. S urveillance w as good. They c ame onboa rd straight away- as soon as he was recognised. Our own staff did the warrant which we did anyway. It just wasn’t treated as anything major really. From these comments, it is evident that there was inconsistent support by various units within N SW P olice. Emerging from thi s is a pervasive attitude t owards s ome cas es even when the series of sexual assaults have been linked. Even once these crimes were linked, it did not alter the way in which the crimes were investigated at the local level as i t w as not cons idered as “anything ma jor” (Detective F aucett). T he u ncertainty of such resources was also evident from Faucett’s interview when she described the use of surveillance: …And we kept surveillance on h im the night before we did the warrant and arrest so w e di dn’t w ant him t o ha ve a nyone w ith hi m w hen w e w ent i n there. But unfortunately the bikies starting shooting each other at the Hellfire Club the night before and they pulled all our surveillance so we went in blind not knowing if he had someone in there or not. The uncertainty of the availability of resources was also raised in the Glover case, where some briefings included the movement of detectives between the task force and their regional positions. This was attributed to reasons such as their prior involvement - 212- in ot her m urder i nvestigations, a nd in one c ase, w here the d etective w as an Italian interpreter, that his skills were required on another investigation. This movement of key personnel c ould b e a rgued t o i mpact on t he s haring of kno wledge o f t he c ases b eing investigated. A s a me mber of the te am l eaves, all the inf ormation that t he indi vidual received c annot be recorded s ufficiently. For example, t he w ay i n which information has be en understood or connected l oosely t o ot her e vidence m ay not b e a dequately expressed. As stated above, there w ere v arying levels of resources available t o the different investigations at any one time. The identification of offences as serial did not necessarily result in the allocation of large quantities of resources. Information Management One of the problems that emerged from many of the serial crime investigations was the way i n w hich l arge a mounts of i nformation w ere ha ndled. This w as pa rticularly th e case in the serial murder investigations studied. In his interview, Superintendent Small describes how Task Force Air had received, “a couple of million pieces of information in a very short time”. According to Mouzos (2001), the holdings of information were not initially that large but had “increased from around seven to ten thousand pieces of information to around 1.5 million in just 12 weeks” (Mouzos, 2001, p.5). This influx of information came when investigators had what they considered to be much circumstantial evi dence and in an attempt to gain some ‘real evidence’ a public hotline w as s et up (Bellamy, 2004) . T his c reated f urther di fficulties f or pol ice, a s t he - 213- computer s ystem that t hey w ere us ing a t the time , the T ask Force Information Management S ystem, w as not a ble t o de al w ith t he qua ntity of i nformation t hat w as produced f rom t his i nvestigation (Bellamy, 2004) . T he ne ed f or an e ffective information management system became paramount. According to Small, It was clear that our e xisting system could neither handle the am ount of information involved nor was it sufficiently flexible to allow for adjustments in a nd the realignment of i nformation groupings a s c ircumstances c hanged, nor the retrieval of information in desired forms. This s tatement is in part s upported by K idd who claims tha t the ta sk force was considered t o have m ade l ittle pr ogress due t o t he l arge num ber of l eads t hat w ere received daily (Kidd, 20 00). It w as ide ntified by B asham in an interview t hat he w as aware of problems experienced with the incompatible computer systems. This led to the decision i n D ecember 1 993 t o i ntroduce a ne w program t o a nalyse t his information, which c reated further d elays a s all ne w i nformation now then had t o be pr ocessed manually (Bellamy, 2004). According to Small, this process required that, “12 different software pa ckages w ere t aken i nto us e a nd l inked” ( p. 7), r equiring t hat m ost of t he information a lready r eceived ha d t o b e m anually t ransferred i nto t he r eplacement system. S mall c onsidered t his pr ocess a s l abour i ntensive but it ensured t hat no important information was overlooked. The need to formulate a ‘new’ system for managing information for Task Force Air came only two years after the Glover investigation, where a s imilar need had been - 214- identified a nd a ‘ solution’ pr ovided. F rom t he c losed c ase f ile, i t i s a pparent t hat a recently i ntroduced s ystem, t he M ajor Incident Investigation S ystem, ha d be en implemented during the Glover investigation. It was described in a document from the closed case f ile as ha ving, “achieved remarkable r esults i n speed of c ollation and analysis o f i ntelligence”. T he ne ed f or even more efficient system than t he M ajor Incident Investigation S ystem m ay ha ve be en due t o t he l arge a mount of i nformation being received through the public hotline during the Milat investigation. Task Force Air would h ave be en r eceiving t housands of l eads e ach da y, w hereas s ome of t he documents f rom t he G lover case m ention s maller num bers. O ne doc ument s tates t hat six hundred and eight calls had been received by the task force, and details the general purpose of each call. Similar issues have not emerged from the case studies of serial rape. However, as pr eviously di scussed, t here w ere i ssues r elating t o t he publ ic hot line i nformation received in the Johnson case, which identified three serving police officers as persons of int erest. There w as l ittle da ta a vailable r egarding the inf ormation received from public appe als i n the ot her s erial rape cases. From t he ava ilable Briefs o f E vidence i t does not a ppear t hat m uch i nformation of i nvestigative us e w as obt ained f rom t hese appeals. - 215- Using the Media in Serial Investigations Separate t o the l argely sensational r eportage o f s erial c rimes i n the m ass m edia, as outlined in previous chapters the media has also been used as an investigative tool. This may b e in the f orm of c ommunity r eassurances, a ppeals f or i nformation f rom t he general public, or in the form of a preventative, proactive activity. Again, this coverage varied according t o the t ypes of s erial crime, where t he m edia w as mor e like ly to be used as an investigative tool for the incidents of serial murder. Media releases in any investigation are carefully controlled. In NSW the Police Media U nit ha s the pr imary responsibility f or the inf ormation that is r eleased to the media. This was evident from t he Glover closed case file, which contained num erous media r eleases as w ell as t he doc uments out lining what was d escribed as t he m edia strategy at different points during the investigation. Much of the information present in the m edia r eleases coul d later be s een in the n ewspaper a rticles and ot her f orms of media output. Often these were incorporated into feature stories, such as one published in the Sunday Mail, which made connections to Jack the Ripper (Sunday Mail, 1989a). The article also includes several ‘profiles’ provided by experts who were not working on the cas e. One de scribes the of fender as a p sychopath getting a t hrill out of t he murders, while another suggests that the offender may actually be a w oman. Similarly, the media reporting into the D enyer cas es has b een described as s ensationalist, where “Not onl y was t he c ommunity unde r s iege w ith a ki ller i n its m idst, b ut it w as a lso under siege from the media” (Petraitis, 1995, p.80). - 216- This s ensationalist r eporting w as c learly not w hat w as i ntended to be encouraged by N SW Police w hen releasing me dia s tatements. From the draft me dia strategies ava ilable f rom t he cl osed case f ile of t he G lover i nvestigation, the a im of such initiatives em erges as doe s t he audi ence that di fferent f orms of media w ere targeting. One s trategy t hat ha d be en pl anned w as t o i nclude t en pol ice officers w ho would be pa trolling t he M osman a rea on m otorcycles, a c lear attempt t o pr omote community safety. The strategy identifies that this should be reported to Sydney-wide media bulletins and thereafter the general public. Present in the file was a running sheet detailing a r equest t o l ocal s tore ow ners a nd r esidents t o r eport s uspicious c onduct, particularly in the presence of the elderly, promoting both community safety and crime prevention. Another a spect o f t he s trategy i ncluded i nterviews on S ydney radio s tations 2GB a nd 2U E. A gain, t he s trategy d etails t he t arget a udience, i n t his c ase s mall businessmen. The aim of this strategy though was not to reassure the community, but a direct appeal for information. In the same draft media strategy document, it notes that a media r elease s hould be di stributed t o out lets requesting “ media a ssistance i n not publishing or broadcasting material likely to incite further attacks”. This approach was also a key feature in the media strategy of the Milat investigation. According t o S uperintendent S mall, a s trong r elationship w ith t he m edia w as necessary in balancing the dissemination of ample information to satisfy the public and media whilst not creating undue public unrest. Superintendent Small explains: - 217- The way that you look at the media I think depends on the different stages and circumstances o f an investigation. I n the f irst instance dow n at B owral an d Belanglo when the searches were on, the risk of the media, the risk for us then was that the media would start making all these speculations, producing highly emotive he adlines that m ay w ell ha ve be en quite w rong. But be cause of the publicity that they would draw, because of the media, because of the political concerns that they w ould generate, because o f t he p olice e xecutive po litical concerns t hat t hey would g enerate, my lament- we would be dr awn of f a nd forced to go down wrong tracks simply to cater to the excitement generated by the media Therefore the management of the media reportage of these crimes can be considered as an i nvestigative s trategy, r educing t he n eed t o pursue unf ruitful a venues and waste resources. Additionally, as outlined in his review of the investigation, Small identifies the global na ture of the m edia a ttention. According to Small, the int ernational me dia attention was due to the nationalities of some of the victims, with media from Britain and Germany present. In his interview, Small also describes the presence of a US CNN reporter which was considered “a first for US coverage of Australian news events”. The coverage of t hese c rimes w as not l imited t o t he m urders t hemselves or to the pol ice investigation, but S mall describes how t he i mpact on A ustralian t ourism w as a lso a n issue. Such i ssues m ay have a lso be en of c oncern dur ing t he G lover i nvestigation, a s one s ource s tates t hat, “The at tacks ha ve al ready attracted interest f rom t he London - 218- press and J apanese t elevision” (Sunday M ail, 19 89a). At th at time , and still today , a large proportion of visitors to Australia originated from the United Kingdom and Japan (Australian Bureau of Statistics, 1989, 2002). During the Milat investigation, it was with these additional pressures that Small decided t hat he ha d t o de velop a m edia s trategy that carefully controlled t he information flowing through to the local and international audiences. This control was gained through a form of cooperation with the media. As outlined in the investigative review, Small explains how time was taken to ensure that the needs of the media were met through the release of strictly controlled release of information through interviews at pr earranged times. T his how ever, w as not a one -sided r elationship a nd t he m edia were expected t o honour t his a ttentiveness. D uring hi s i nterview, Small describes t he relationship i n r elation t o t he publ ic, pol itical a nd m edia p ressures t he police were encountering: And i n t hose s orts of e nvironments m y- and a lso i f y ou di dn’t c ater t o t he media they were going to get photographs and leaks and stories one way or the other- so in that sort of environment my view was the best thing you can do is try a nd c ontrol t he media by pr oviding t hem with t he f acts as be st you c ould give them and in a sense controlling those facts. And also getting them to understand that anyone that was caught, if you like, cheating or going out of bounds would be banned from attending but a lso gathering your trust that- or having the trust that we would not give a story to - 219- one j ournalist but not t he w hole l ot. So all be treated fairly a nd this was not about- bear in mind that at the time of the searches we had around 300 police, so anyone could be in a pub having a drink, anything could happen, having a cup of coffee, whatever, and just make a comment that could send you off on a tangent. So what we did we got the media and we set some ground rules…. Being perceptive of the needs of the media played an important part in this relationship. Small describes the way in which such needs were understood and subsequently met: …okay, we know when you have to meet your deadlines and- in the morning and in the afternoon- and also we know that you want a bit of colour and you need some background, so in the morning we will do an interview each day in the forest. That gives you the scene of the investigation. But some of you can’t message [sic] from t he f orest be cause there w as no sat ellite t o beam t o the television screens so in the afternoon, and we can do that at a time that allows you to get b ack i nto Bowral and beam y our m essages a nd do w hatever y ou want to meet your deadlines, and in the afternoon because things are tighter, we can do an interview at the police station in Bowral each afternoon and those of you that need to can make it a direct feedback where you don’t have a delay. The r easons f or ens uring m edia com pliance w ere t hree-fold. A s ignificant i ssue a s identified by S mall w as the ne ed to reinforce public c onfidence i n t he p olice a nd t he investigation. T his w as particularly i mportant f ollowing t he m edia r eportage of t he discovery of t he f irst r emains and subsequent s earch of t he Belanglo State F orest, which did not yield any further remains. Twelve months after this search, more remains - 220- were found, and Small states in the investigative review that it was important to restore confidence in the investigation. In his interview, Small states that: And a lso w e had t o convey that w e w ere doing e verything pr actical so t here was always a confidence in the investigation itself. And that was one of several themes that continually came through. And that worked quite well. This was later reiterated when he states that: One of the things that came out of the media was, some of the themes we kept giving the media, and which they took up w as, here’s a group of people doing everything t hey c an t o s olve t hese c rimes. The c ommunity s hould ha ve confidence in them. T he c ommunity s hould n ot g o into a s tate of hy steria or panic. A nd t hat w orked i n our f avour. A nd t he s econd t hing w e w ere do ing with the m edia w as, by d oing a ll o f that, if w e g ot to t he s tage w here w e couldn’t solve the crime, what we had was a media that had given the message that t hese pe ople w ere doi ng e verything t hey c an. A nd w e w ould be i n the position to say, look we have done everything we can. We can’t do anymore. If anyone has any suggestions, we’re happy to take it up. So you wouldn’t have had a media provides a negative attack on police fail in this. You know, here is a g roup t hat’s don e e verything t hey c an, they’ve be en una ble t o s olve t hese murders. I t wasn’t about doing a PR job for ourselves. It was more about making sure the community retained the confidence in the police and that they do the job properly. - 221- This supports the ideas proposed by Chavez and Gibson (Chavez & Gibson, 2004) that such public r elations act ivities ar e a m ethod of cr isis ma nagement. From S mall’s discussion it i s cl ear t hat t he m edia w as us ed as a m eans of r eassuring t he publ ic following the discovery of additional bodies in the Belanglo State Forest. However, the insistence on portraying the investigation as one doing everything possible could be an indication that thi s w as part of a publ ic r elations e xercise, attempting to gain public confidence. T he ne ed t o bui ld this c onfidence c an be s een t o s tem f rom t he i nitial response t o t his c ase. T his i s s uggested b y Bell (2002) when he cl aims t hat as pol ice had i nitially “…downplayed t he e arlier di scoveries…they’d i nvited a m uch m ore rigorous media vigilation of what they did now” (Bell, 2002, p.237) . The multifaceted nature of t he pol ice-media r elationship during t he M ilat inve stigation is s ummed up best b y S mall w ho states tha t w hen looking at the pr iorities a t the s tart of hi s involvement with the investigation: The second t hing w as how w e w ere going to handle the media, because they were b ecoming qui te s peculative, and t hat creates t he po ssibility of unnecessary concern in the community, and can also side-track investigations. And another thing was, “How do we stop the fear level increasing in the small local communities?” (quoted in Bell, 2002, pp.239-240). Further to this was an important theme emerging from the interview and can be seen a s a m otivator f or di fferent a spects of t he i nvestigation. It w as evident tha t investigative t echniques were em ployed in a w ay t hat s upported the m edia s trategy. - 222- When di scussing t he r easoning b ehind t he f ormation of t he T hink T ank a s a n investigative strategy, Small validated this step with the following explanation: What you were envisaging was I guess a coupl e of scenarios. One was that we would never solve the crimes. But we had to make sure that we had done everything we c ould. E verything t hat w as humanly p ossible t o solve t hem. And w e a lso ha d to make s ure t hat t he public w as aware that w e ha d done everything t hat w as hum anely prac tical t o solve t hem so that t hey ha d confidence in t he i nvestigation. “ Yeah, l ook. T hey di dn’t s olve i t bu t t hey gave t heir be st sho t. You can’t c riticise them”. I’m not t alking a bout criticising m e pe rsonally, but t hey ne eded t o ha ve c onfidence i n t he investigation. This reinforces the argument that such media strategies were an attempt to boost public confidence in the investigation. Additionally, in relation to use of the media as a di rect investigative tool, Small concedes that there had been a diversion from the usual media relationship. In what can be seen as a direct use of the media as an investigative tool, this had been done in order to elicit information from their suspect, Milat and members of his family. Small explains: One thing when we did use the media was Chris Masters on the Four Corners program had actually done their own homework. And they came across two women from the ABC who had told this story about being picked up w hile hitchhiking a nd escaped and C hris Masters p roduced t hat m aterial. B ut he also wanted to do a Four Corners story and he was going to use the girls in - 223- it. And the girls in fact they identified Richard Milat and Ivan Milat, I think it was- but I’m not s ure o f that- but t hey were s hown phot ographs and they picked two men from the Milat f amily, one of whom was Ivan. But t he evidence w asn’t s trong en ough to refer cha rges a nd t hey ha d escaped. So even that report was on the r ight track. And Chris Masters in fact t hen, we cooperated with them doing a Four Corners program. One of the reasons we cooperated was, because Masters had done a lot of his own work too, but we were interested in seeing what reaction if any there might have been among several of t he Milat families. And we monitored those r eactions in various ways. So yes we used the media to do that. This strategy is identified by Small as being aimed at putting pressure on the offenders. Finally, the media strategy also aimed to gain information from the public, as was seen in the Glover case. As stated in the case review, “and perhaps our most obvious strategy was to jog the memory of people who might have relevant information and to get them to come forward” (Small, 1998, p.6). Many i nvestigations of s erial crime utilised the me dia a s a more ge neral investigative tool , with the m ost c ommon forms be ing calls f or publ ic assistance or information. This w as evi denced in the m edia r eleases f rom cl osed case f iles of t he Glover c ase as di scussed a bove and dur ing t he M ilat i nvestigation, a mongst ot hers. This approach however was not always beneficial to the progress of an investigation. In his di scussion of t he M ilat c ase, Bellamy (2004) states t hat publ ic as sistance w as no t contributing t o pr ogress a s m any s ightings of t he t wo vi ctims w ere r eported t o oc cur - 224- after they were believed to be murdered. This problem of misinformation gathered from the publ ic a nd its lack of pos itive c ontribution to the inve stigation highlights th e argument tha t s erial crime inve stigations a re c onfronted with more di fficulties tha n traditional i nvestigations. T his w ould ha ve be en f urther i mpeded due t o t he pe riod of time between the victims’ disappearance and the discovery of their remains. The use of the media as a crime prevention tool during an investigation was also apparent in some of the serial murder and serial rape case studies. One example detailed previously occured during the Glover investigation whereby community members were asked t o b e a ware of t hose a cting s uspiciously around t he e lderly. D uring t he P resta investigation, there w as a poi nt w hen it w as de emed necessary t o use t he m edia i n a similar capacity as illustrated by Faucett who states: Detective- We were in the process of doing that [going to the media]- when we identified him. At the time things weren’t quick like they are now. If you put in a phone trace now you can get it back in 24 hours if you need it. At the time it w as taking weeks. And it got to the point where I’d approached t he boss as I felt we had to go public with it because of the danger involved in it. They’d just agreed they would do that when we got the return to identify the person a nd w e c ould p ut him straight unde r s urveillance w e k new w hat h e was doing. - 225- Interviewer- If you hadn’t identified him, the reason before for going public would be a s afety c oncern f or ot her vi ctims. So i t would be a pr evention measure? Detective- Yep. Yep, he was getting more violent. And that difference occurred, he w as g etting m ore violent, t hey w ere c oming out w ith m ore injuries and they w ere g etting m ore f requent. There was a v ery sho rt time between the last two. Until that time, the media reportage had been limited to individual incidents in the l ocal m edia, but it i s unc lear how m uch attention these r eceived from t he l arger media outlets. The extent of the media reporting of the other cases can be considered as limited. The cases of Folbigg and Presta were reported on only sparingly prior to their arrests. T he F olbigg c ase ha d had limited coverage i n the N ewcastle ar ea and surrounding regions where the investigation was taking place, although this was largely after she had been charged with the crimes. The use of the media as an investigative tool was not seen to be significant in any of the cas e studies of s erial r ape. There w as no m edia c overage of t he T rindall crimes prior to his court appearance. The nature of the investigation precluded the need to engage t he m edia in the investigation, as the task force was only initiated after t he suspect had been identified through DNA. - 226- The influence of a task force on the level of media coverage relates to the police informing the media o f t he task forces establishment. This was evident i n the cas e o f Daley. Once T ask Force S apelo had been initiated, there w ere regular ne wspaper articles t hat covered the general features of t he s exual as saults and the d escription of the pos sible of fender. As w ith t he c ases o f Milat a nd G lover, there w ere s pecific requests f or i nformation, of ten i ncluding t he c ontact phone num bers (Cowper, 2000; Toy, 2000). The task force however, did not make any connections to the incidents in Leichhardt, indicating that there had been no linkage between these crimes at that stage. The media reporting also aimed to increase community awareness and promote the s afety of w omen. Inspector M cKay, who led Strike F orce S apelo, is quot ed a s saying, “We w ant t o prevent an y further attacks f rom oc curring” and “We s trongly recommend women who are on their own take particular care especially while parking their c ars at ni ght a nd, i f pos sible, r emain i n t he company of ot hers” (Parramatta Advertiser, 2000) . T his had followed a n a ssurance b y Inspector M cKay t hat t he investigation “…was working relentlessly to bring the man responsible for these attacks to j ustice a nd a ll a vailable r esources a re be ing de voted t o t his i nvestigation” (Parramatta A dvertiser, 2000) , echoing th e s entiment e vident dur ing the M ilat investigation t hat a t horough i nvestigation w as be ing c onducted, t hus de emed as fulfilling a public relations role. During t he J ohnson i nvestigation, no m edia strategy per se is evident. Y et t he media coverage of the investigation into the serial rapes can be seen to have impacted - 227- on the i nvestigative pr ocess. T he i nvestigative r eview do cument, states t hat t he investigation received a great deal of media attention, some of which covered the loss or m ishandling of e vidence collected f rom t he c rime s cenes a nd allegations of negligence during the initial investigations. This media attention is identified as having led to the direction to reinvestigate all incidents associated with these serial of fences, including t he r einterviewing of w itnesses a nd vi ctims, but a lso t he l ocating of a ll evidence related to the cases. The then Minister for Police, Mr Whelan, is reported to admit that further victims may have been spared had the investigation been conducted properly ini tially (Riley, 1997) . Therefore t he media cove rage w as a m easure of accountability for this investigation. Another way in which the media had been used was to help identify information that had been previously reported on in media sources. During the Presta investigation, a number of diaries had been obtained from search warrants on the offender’s property. These diaries had recorded details about other apparent assaults. Faucett describes her use of a media organisation to gain more information on t he cases that she was unable to identify at the time of the Presta investigation: I used the media after we had charged him to locate old offences that he had …we found 7 years of diaries but unfortunately he had stopped writing at the time of ou r o ffences. They da ted back qui te a few y ears. A nd when I w as reading through them I found… sort of confessions to other assaults. And I used Fairfax Research Group to find some old comfits that had appeared in - 228- the paper. So they were brilliant. It was one phone call and they got straight back pretty quickly. It w as unc lear a s t o w hy t his i nformation c ould not ha ve be en a ccessed t hrough t he records of NSW Police, although Faucett had stated earlier that cases were difficult to find using their systems unless one knew what they were looking for. In three of the cases studied in this research there were no media reports evident prior to the arrest of the offenders. In both the Trindall case and the Snowtown murders there were no media reports found relating to the cases or the victims at the time of the criminal e vents. T his l ack of m edia reporting i n bot h c ases can b e a ttributed t o t he failure of pol ice t o l ink of fences t o ot her s imilar c rimes. T hus i n bot h c ases, t he investigations were not subjected to public scrutiny in the way that the Milat, Glover or Daley c ases w ere due t o t he l ack of publ ic a wareness. D uring t he S nowtown investigation this public awareness may have been beneficial to investigators, allowing them to conduct the in vestigation without a larming the of fenders. H owever i n t he Trindall c ase, the la ck of me dia attention is a ttributed to the f act th at the re was n o active i nvestigation unt il a fter t he of fender ha d been i dentified t hrough a c old D NA match, where a match was made from an old case. In the Folbigg case, there was minimal media coverage of the crimes until after the arrest of the offender. This is largely attributable to the nature of Folbigg’s offences and the f act t hat t he crimes w ere not i nitially r ecognised as s uch. It i s not s tandard - 229- procedure to issue press releases when children are deemed to have died from natural causes, such as SIDS. Investigative Techniques and Technologies Utilised A range o f i nvestigative t echniques a nd forensic t echnologies w ere u sed i n t he investigations of both serial murder and serial rape cases concerned here. It was cl ear that there were several issues emerging from the case study materials that were critical of s ome a spects of i nvestigative pr actice, i ncluding t he f orensic analysis of evi dence. The case s tudies demonstrate t hat most c ases u sed, t o s ome extent, DNA t echnology. Other fo rensic t echnologies were al so used and can be considered t o have h ad an impact on the outcome of these investigations into the serial crimes. Several f orms of f orensic t echniques w ere ut ilised i n t he i nvestigation of t he Milat victims. Forensic odontology was used to formally identify some of the victims including G ibson , E verist a nd S chmidl (Bellamy, 2004 ). Ballistics technology w as used to aid in linking these crimes as the weapons used in separate cases were found to be identical (Bellamy, 2004). However problems reportedly arose in the collection and analysis o f D NA evi dence. Bellamy (2004) suggests t hat du e t o de composition, evidence of sexual assault was difficult to determine in the cases of Clarke and Walters. During Milat’s trial, the court was advised that evidence of sexual assault after a period greater than three months was near impossible. Although this may have been the case, problems with contamination of this evidence were raised. - 230- It was found that the scientist who conducted the DNA testing on the Clarke and Walters samples had found foreign DNA present; however, retesting produced different and inconsistent results on each occasion (Shears, 1996). It was found that the samples had be en c ontaminated on m ultiple oc casions b y t he s cientist a nd t he laboratory equipment in the used in the DNA testing (Shears, 1996). Further to this, the scientist admitted under questioning by the defence lawyer, that he was not qualified to conduct such testing. In addition to this, hairs found in Walters’ hand when her body was discovered were d etermined not t o be he rs. The f orensic scientist s tated that co mparison to a possible of fender w ould ne ed t o be c onducted w hen a s uspect w as l ocated (Shears, 1996). I n a nother alleged act o f i ncompetency, it w as r evealed that s ome of t he ha ir samples ha d b een ha ndled b y a s cientist w ho di d not ha ve t he relevant experience i n DNA testing and whose prior work in this field was questionable (Shears, 1996). Had these s amples be en the mos t s ubstantial e vidence against M ilat, in all probability he could ha ve be en acquitted. T he contamination of e vidence m ade any results unreliable and unusable, thus possibly affecting the investigators’ ability to limit their f ield i n r elation t o suspects or g ain a n und erstanding of how t hese c rimes w ere committed. One of the main issues that are highlighted here is the lack of control police had over the samples. The police reportedly had no knowledge of which individual was conducting the testing, what their qualifications were, and the way in which the samples were handled (Shears, 1996). - 231- When a sked a bout t he r eported c ontamination of t he f orensic e vidence Superintendent Small explained in his interview that his recollection was that the hair samples had been degraded due to exposure and therefore had been deduced to be not belonging to the victim. With the subsequent investigation following the discovery of the other victims, retesting of evidence occurred and “the hair and other samples were taken to the United Kingdom which was then regarded as the foremost body and they found that the testing of the hair at Glebe was incorrect.” These latter tests concluded that the hair samples had belonged to the victim. In relation to the contamination of the samples ta ken to determine if D NA evidence in dicating whether a s exual as sault h ad occurred, Superintendent Small states that he was not aware of this problem. Therefore it was not possible to gauge what impact this may have had on the investigation. Problems with the processing of DNA evidence were not exclusive to the Milat investigation. A s out lined i n ne ws r eports, a n inquiry i nto t he i nvestigation of t he Johnson s eries of r apes found t hat pol ice ha d lost s emen samples i n what ha s be en described as a ‘botched’ investigation (Riley, 1997; Warneminde, 1997). According to Warneminde (1997) these s amples were not found for 14 months b y which t ime they were deemed useless for the purposes of DNA testing. This reflects on t he knowledge of procedures relating to the collection and storage of DNA evidence. As the Trindall case de monstrated t hrough t he s uccessful a nalysis of e vidence up t o t en years a fter collection, properly s tored DNA evi dence c an be ef fectively an alysed years af ter t he date of c ollection. Therefore t he p roblem e xperienced i n t he J ohnson c ase m ay be - 232- attributed t o t he e xperience and/or t raining of t he pol ice of ficers w ho collected and handled the DNA evidence. Similar pr oblems w ere a lso evident in the T rindall s erial r ape inve stigation. From the Prosecutorial Brief of Evidence it was evident that the DNA evidence in the form of Sexual Assault Investigation Kits (SAIK), collected at the time that the assaults had occurred w ere not processed unt il 2004, n ine years a fter t he first a ttack w as reported t o pol ice. Initially, onl y t hree cases d ating ba ck t o 1997 w ere linked by t he Diagnostic a nd Analytical Laboratory (D AL). Following t his a m anual r eview f or similar cases was conducted by Strike Force Grahame, which then had to identify the location of the evidence collected from each case, with one dating as far back as 1995, in or der t o obt ain a D NA a nalysis and s ubsequent c omparison t o t he o ther m atched samples. From the r ecords available in relation to thi s c ase, it is c lear tha t the re w as minimal, a nd of ten no, formal i nvestigation i nto t he s exual a ssaults w hen they w ere reported to police. This lack of formal investigation meant that there was no follow-up on the forensic evidence submitted. From the Brief of Evidence, forensic evidence logs show tha t s ome ite ms c ollected from the vi ctims or the c rime s cenes were de stroyed prior to the analysis of the other DNA evidence. From the log it was also evident that some of this evidence was never forwarded for scientific or investigative analysis. - 233- There may be a number of reasons for this, in addition to the lack of a follow-up investigation at the time of the offences. The Trindall crimes can be seen as an example of linkage blindness. Without the knowledge that these cases were linked, the resources needed to conduct a co mprehensive serial crime investigation were not available. The task force to investigate these crimes was only set up once the DNA evidence had made the initial link to the offender. Failure to follow-up on the testing of these samples may again be attributable to detective experience. In this case, however, the issue of the victimology emerged. In one news article, Detective Norris who l ed t he i nvestigation, i s qu oted as s tating t hat t he i nvestigation, “…involved a number of girls” (Ravens, 2006) although all victims were adult women, thus r eminiscent of t he term ‘ call g irl’ of ten us ed t o r efer t o pr ostitutes. In a nother article, the detective is quoted as making the point that, “Everybody in society is at risk of this happening to them, and if somebody is sexually assaulted we encourage them to come to police a nd tell us a bout it, a nd we'll inve stigate it” (ABC P remium N ews, 2006). This can be seen as a reassuring statement to the public, creating the image of an impartial police force. Absent from this was the acknowledgement that these cases were not labeled as a priority when they occurred, which led to the delays in the analysis of the DNA evidence. Unusually, t he i nvestigation i nto t he s erial r apes committed b y P resta di d not rely on forensic evidence. The offender had videotaped his crimes and these recordings - 234- were found during the execution of the search warrant. In relation to the presence or use of DNA evidence, Detective Faucett states: No. t here m ay have be en but w e di dn’t use i t. The DNA w ould onl y ha ve been from P resta’s h ouse. We di dn’t g et any D NA f rom the sexua l ass ault kits. This went so quickly and since it didn’t go to trial we didn’t need to get it. The young bloke from Emu Plains- the 15 year old- the sexual assault kit said he wasn’t sexually assaulted. The videos showed that he was. So there would have been no phy sical evidence to back up a ny claim he would have made of sexual assault. Without the video there would have been nothing but a common assault on him. This highlights the importance of looking at other evidence and not relying on DNA or other forensic technologies as the limitations of such technologies are often overlooked. Other investigative techniques Problems w ith the an alysis of ot her evi dence were al so present i n ca ses s uch as Johnson’s, in addition to those outlined above. One of the main issues raised in news reports was the loss of a balaclava belonging to the offender which was not processed for f orensic clues as i t had be en forwarded t o l ost pr operty (Warneminde, 1997) . Additionally a t orch t hat a lso be longed t o t he of fender ha d a pparently not be en fingerprinted and thus was overlooked as evidence (Warneminde, 1997). - 235- The use of forensic techniques was not always associated with problems. In the Milat case, m any of t he victims could onl y b e i dentified t hrough forensic odont ology due to extensive decomposition. From the news reports at the time it was evident that this pr ocess w as a n e fficient m ethod of i dentification: J oanne Walters was ide ntified within 48 hours of the discovery of her remains (Herald Sun, 1992). Richard Basham, who was a consultant on t he M ilat i nvestigation, i dentified t hat t here ha d be en a tendency of pol ice to vi deo tape t he cr ime s cenes. During t he i nterview, he s aid that there ha d be en v ery few g ood s till phot ographs t aken and that h e be lieved that t he recordings w ere not of good qu ality or clarity. This c ould not be c onfirmed a s t he closed case file that would contain these photos was not available. The use of forensic odontology was not limited to the Milat investigation. The discovery of the first Snowtown victim had been made some years before he could be identified. As the r emains w ere s keletal, other f orms of i dentification w ere r equired. When one of t he vi ctims, Trezise, was reported missing a year after hi s de ath, S outh Australian police had requested the Forensic Science Centre in Adelaide to compare the remains f ound t o t he recent m issing pe rsons r eport. T he e xpert w ho conducted t his comparison negated the identification, significantly delaying identification and thus the investigation of Trezise’s death by a further four years (McGarry, 2005; Pudney, 2005). This can be seen to have impacted on t he investigation significantly as t he linkage of the murder of Trezise to other missing person cases ultimately led to the identification of the offenders (McGarry, 2005; Pudney, 2005). - 236- Traditional investigative techniques such as surveillance were a feature of many of t he investigations studied. This t echnique w as m ore f requently u sed after t he identification of a suspect, although with serial investigations producing large numbers of suspects, this meant that surveillance was not limited to the actual offender as it was also used to eliminate s uspects. This w as particularly evident i n t he G lover investigation, where documents from the investigation identify the use of surveillance on a num ber of s uspects. S urveillance was also us ed i n t he i nvestigation of P resta although t his was l imited t o t he d ay be fore t he execution of a s earch warrant on hi s premises. During the Folbigg investigation, a technique called Scientific Content Analysis (SCAN) w as us ed t o pr ovide a doc ument analysis of t he pe rsonal di aries of t he t hen suspect, Kathleen Folbigg. The aim of this tool in the context of this investigation was to give the detective information that would aid detectives during the interview of the suspect (Smith, 2001) . Detective R yan de scribes t he us e of t his t ool a nd i ts lack of contribution to the investigation when he stated that: He looked at them (the diaries) and it’s a technique that analyses the way that people write- I don’ t know- whether t hey put a pr onoun or a n a dverb I don’ t know- to try and give the investigator a perspective about the personality. The reason we w anted that w as t o start to develop an i nterview pl an. She w as- I wanted to be able to try and gain a confession or an admission…. She made a couple of a dmissions bu t t he r eport really di dn’t give assi stance a t al l i n relation to her personality it sort of gave us the opposite. It said that one of my - 237- techniques should be to develop a rapport get a detective that could empathise with her - a female detective that had had children. As it turned out every time she asked a question K athleen would direct her answer to me. She w as more comfortable with me. Sort of enjoyed the interaction she was having with me and that’s.... While this tool did not produce the information to assist in the investigative process as had been hoped, Detective Ryan acknowledged the importance of using a range of tools to inform the investigation. Although this technique was unsuccessful in contributing to an effective interview plan, this detective had not dismissed using this technique again in the future. It is interesting to note that this technique was not used in the Presta case, where a num ber of pe rsonal di aries w ere o btained f rom t he o ffender. W hen a sked i f there had been any forensic analysis of the diaries, Detective Faucett stated: No one has actually seen them other than me. SY (Behavioural Crime Unit) has them at the moment but no one has ever really looked at them. We had some blokes come out from Canada last year, who only looked at one as an example. From this statement it is evident that the diaries were not analysed forensically for over eight years, as P resta was ar rested in 1997. This i s de spite D etective F aucett’s contention t hat t he di ary e ntries i ncluded r eferences t o ot her c rimes t hat the of fender had committed. T his may have been be cause t he di aries were not obt ained b y pol ice until t he s earch warrant ha d been executed and the s uspect ar rested. When this occurred, t he s ubstantial e vidence obt ained f rom t he pr operty ne gated t he ne ed for - 238- further investigative tools to be used as the police may have been satisfied that they had enough evidence to obtain a conviction. Due t o the r elatively r are na ture of s erial m urder and serial r ape, many pol ice investigators did not ha ve any ex perience in investigating s erial crimes. The l ack of familiarity and training in collecting f orensic evidence may b e a ttributable to this inexperience. Moreover, the value of some evidence may not be realised at the time of the offence. Due to the prolonged timeframe over which these crimes often occur, this realisation may onl y c ome af ter crimes h ave be en linked and a s erial of fender suspected. A s de monstrated, t his w as e vident i n t he T rindall c ase w ith t he de lay i n DNA evidence analysis. Pressures on Police A common t heme running t hrough m any of t he biographical and newspaper a ccounts was one of bl ame. T his w as di rected t owards t he pol ice who w ere of ten bl amed f or reacting poorly to the circumstances and failing to recognise that a s erial offender was at work, which was compounded by the absence of formal linkage systems. This blame was directed at police during both serial murder and serial rape investigations. The key concern was t hat i nadequate i nvestigation of va rious c rimes l ed t o t he continued offending behaviour of the serial offender and thus further victimisation. - 239- The cas e s tudies identify tha t a n allegation of i ncompetence was first l evelled during the 1989-1990 serial murder investigation into John Glover. Kidd (2000, p. 237) suggests that towards the end of Glover’s crimes he was “leaving clues all over town” which m ay ha ve s aved a c ouple of l ives ha d d etectives recognised the v alue of t hese clues. The f ailure of pol ice t o l ink a s eries of sexual as saults ag ainst el derly w omen which oc curred c oncurrent t o t he m urders c an also be c onsidered a s h ampering t he outcome of the investigation as the incidents later had a significant role in identifying Glover (Kidd, 2000). Such claims emerged later on in the investigation, after it became clear that the case theory leading the investigation was at fault. Similarly, it was also alleged that if police had responded “a lot more efficiently on Onions’ report of the 1990 i ncident with Ivan Milat, the killer could possibly have been apprehended a l ot s ooner” (Kidd, 1999, p. 140) . The r eport o f t he at tempted murder and robbery was t aken b y pol ice at t he t ime but from t he s ources, i t does not appear that there was any investigation into this attempt on Onions’ life. It was reported in some sources that police put a description of the vehicle over the police radio (Kidd, 2000) but there was no evidence of further investigation of the report at the time. The suggestion is that lives may have been saved had the Onions incident been investigated t horoughly as t his e vent occurred p rior t o t he di sappearance of f ive of Milat’s victims. Indeed Kidd (2000) claims t hat many l ives coul d have b een saved if police had responded adequately to the report filed by Paul Onions stemming from the attack on him by Milat (who at that time was unidentified), which occurred during the - 240- same month that Everist and Gibson disappeared. Kidd (2000, p. 260) describes this as a “ monumental bl under by country pol ice”, s uggesting the inadequate e xperience of police in ‘country’ areas was to blame. As identified by a number of sources including Deans (2006), Shears (1996) and Whittaker a nd K ennedy (1998) amongst ot hers, O nions’ r eport i n November 1993 made to the police hotline in response to the media attention following the discovery of the other bodies in the Belanglo State Forest, still remained unlinked to the other crimes for a further four months. This criticism could be carried further, in that even after the discovery of the bodies in Belanglo, no one from the police station that took the report, thought it s ignificant or possibly r elated. One s ource s uggests that t he r ecord of t his attack was m isplaced a nd that the onl y r ecord remaining w as f rom an officer’s notebook (Shears, 1996). According to Shears the original statements taken by police at Bowral P olice S tation w ere di sposed of w hen t he c onstable w ho t ook t he r eport w as transferred t o a nother s tation. T his i s s upported t o s ome e xtent b y D etective S mall when he explained that: It wasn’t recorded as it should have been. And in the first instance we didn’t do the phy sical ch eck, we w ere told i t ha d be en c hecked. S o w e t hen ha d t he report of the incident, Onions and the woman from Canberra. And at that point it became a priority. So there was a delay. I’m not suggesting it happened in 24 hours. But the delays we a re t alking about a re d ays and weeks, not anything longer. A nd it i s no t un reasonable g iven t he amount of i nformation that w e actually had. - 241- These delays were not seen to have had an impact on t he investigation as such delays were considered reasonable in an investigation that spanned over two years. Detective Small also responded to the criticism that some victims died due to the lack of police action to the Onions report. He stated that: It w ouldn’t ha ve c hanged a nything. I n f act, probably w ouldn’t ha ve ev en shortened t he enquiry. W hat i t m ight ha ve meant w as that t he enquiry ha d a slightly di fferent skew t o it. And the subject w ould have come f irst but even that was not enough to simply race out and arrest Ivan…. We h ave sa id that this m atter should have be en pr operly f ollowed up a t the time. W hether it w ould ha ve made a ny di fference or not i s a nother i ssue because at the time of the Onion incident they hadn’t found any of the bodies. So w hat they w ould ha ve ha d was a one -off i ncident t o pursue…. They (the victims) hadn’t all disappeared and I guess you can get all emotive and say that you could have prevented that murder. The fact of the matter is that you had a one-off incident and that’s where you would have stayed. This suggests that there was little about the Onions case to suggest that there would be any links to the disappearance of backpackers in the general area. As previously stated, there was no f ormal l inkage s ystem i n pl ace at t he t ime o f t hese of fences. Had one existed, it is possible that police may have been more aware of the number of victims who had disappeared under s imilar ci rcumstances, and t herefore t heir i nvestigative capabilities may have been improved. - 242- Similarly, the s ignificance of the d elay in linki ng th e O nions assault to other information received from the public was also questioned. In responding to this claim, Small puts the information in the context it was received, as follows: Interviewer- Another t hing t hat c omes up a f ew t imes- Paul O nions, t he ‘victim that got away’, it’s been r eported t hat he r eturns t o England, t he bodies are found, hotline is set up, he rings the hotline, but nothing was done. Detective- Well, I don’ t know a bout t hat. C ertainly I t hink i t’s f air, I can’t remember t he number, bu t w e ha d h undreds o f thousands of c alls ov er a couple of weeks, so I think its fair to say that no one would have seen, no one on the other end of the phone would have jumped up a nd said, “We’ve got him! Q uick. J ump on a pl ane a nd c ome out t o A ustralia,” or a nything l ike that. But w hat ha ppened was abou t the time P aul Onions rang w e a lso received a pho ne call- sorry three s eparate things occurred. About the time Paul Onions rang, a woman rang from Canberra and she was the person who had pi cked up P aul O nions a nd dr iven hi m t o Bowral pol ice s tation. Then when Paul Onions rang, we were able to put those two things together. We had also, prior to that, been told by l ocals about this abduction that h ad occurred and in fact they were talking about the Paul Onions matter. And that had be en reported to Bowral police station. We ha d checked Bowral police station for the records but couldn’t find anything. When Paul Onions and the woman r ang and were p ut t ogether w e s aid “H ang on, t his is the B owral incident”. We need to go back and check everything again. And in doing that we then found a record of it. - 243- Similar accusations of further victimisation resulting from police incompetence were levelled during the Johnson investigation. In 1997 the then Minister for Police, Mr Whelan, w as r eported in t he m edia as ha ving claimed that “ some of the women may have be en s pared t heir ordeal i f i t w ere not f or t he ‘ gross i ncompetence’ of pol ice” (Riley, 1997). This accusation was seen to have partially stemmed from the revelation that it took NSW Police 13 months to instigate a task force to investigate this series of sexual attacks against women. Chapter Summary This c hapter e xamined the w ay in which the inv estigations int o serial mur der and serial rape are constructed. From the case study and interview data, the importance of t he a llocation of r esources i n the f orm of a t ask force em erged as a k ey i ssue t hat emerged. This chapter has also demonstrated the role of the media in the investigation of violent serial crimes, both as an investigative tool and as a source of pressure on the investigation. - 244- CHAPTER 7: THE POLICING PERSPECTIVE- CULTURE & PRACTICE Strong themes emerged from the data relating to the use of police cultural knowledge to inform the investigation of serial rape and serial murder cases. Supporting claims made in the r esearch literature a nd t he i nterview da ta s howed t hat s ubcultural di fferences were evident within the police organisations. The strongest issue that emerged was the use of dictionary knowledge, and thus stereotypes, in informing the responses of police in the sample. As suggested by the research literature, victim stereotypes were evident; these w ere f ound t o ha ve i nfluenced t he i naction of pol ice a t t he t ime t hat s ome incidents were reported, thus suggesting that this knowledge delayed the investigative outcomes. A s s uch, di ctionary kno wledge m ay be c onsidered t o f ail i n i ts de sign t o contribute efficiently to policing outcomes (Chan, 1999b) in the investigation of serial murder and serial rape. Although there have been several incidents of serial murder and serial r ape experienced A ustralia-wide, there appears t o be l ittle in t he w ay of dictionary knowledge s pecific t o t hese c rimes, or t he di rectory know ledge t hat i s informed by it. - 245- Reflecting Subcultures Competing Subcultures In C hapter 1, the ex istence of va rying and sometimes com peting o ccupational pol ice subcultures was di scussed. O ne s et of competing s ubcultures ha d be en identified b y Ericson (1993) as b eing between general d etectives and detectives from s pecialised units. T his sense of competing s ubcultural i dentification e merged from one of t he interviews pa rticularly when t he i nput of s pecialised uni ts to the inve stigation was raised. There was however a difference in opinion depending on the duties performed by the particular unit. What is pe rhaps mor e te lling is tha t s ome pa rticipants di d not r ecognise the input of other units at all. When forensic evidence is a key feature of a case, one would expect t hat i n N SW t he F orensic S ervices G roup w ould ha ve h ad s ome r ole i n t he collection of e vidence. T his group however w as ne ver m entioned. T his was pe rhaps due t o an assumption about t hat unit’s responsibility which pr ecluding it from having an ‘investigative’ role similar to that of detectives. In the Folbigg case, forensic experts were sourced from around the world but there was no reference to NSW Police forensic experts or the Forensic Services Group having had any role in the identification of these experts. One de tective interviewed described his ex perience i n investigating s erial rapist, Luke Slater. During the investigation of this case, the Forensic Services Group - 246- was seen to offer a criminal profile of a suspected offender although there had been no request made of them (Detective Pryde). The B ehavioural S ciences U nit of N SW P olice pointedly did not r eceive an y positive comments from those interviewed. In a serial sexual assault case the BSU was called upon t o pr ovide a pr ofile of a pos sible of fender. T he geographical pr ofile w as produced which identified the offender as “…living somewhere over this Penrith way, Penrith-mountains s ort of w ay.” ( Detective Faucett). T his pr ofile w as c onsidered inaccurate, as the detective also stated that, I assumed he would have been from the Liverpool area as that’s where the first offence occurred so I thought that’s where he’d be familiar. This was the area in which the offender was later located. Similarly the criminal profile provided by the BSU was: Very, very general. B ut i t w asn’t ba sed on a nything ne w… t he kids ha d given us , pr etty m uch out lined w hat ou r profile w as g oing t o be s o t here wasn’t a l ot m ore to add t o it bu t w e w eren’t g iven anything el se any way (Detective Faucett). While Detective F aucett conceded that t he BSU ha d been responsible f or some assistance, it was made clear that this assistance was not constructive. - 247- However, dur ing t he S later i nvestigation, Detective P ryde ha d a di fferent experience with one of the specialised units of NSW Police. When asked about support he readily acknowledged that: …a d etective i nspector f rom t he st ate c rime com mand serious and violent crime … took a n i nterest i n the m atter. H e a ssisted i n conducting s ome enquiries interstate using the resources of the state crime command. Consistent with this opinion, other specialised units were perceived in a similar light. Detective F aucett had pr eviously c ommended t he S tate P rotection G roup a nd Surveillance t eams for r esponding qui ckly t o her requests for assistance on t he P resta case. These units were not directly involved in the investigation and were brought in to support the lead detectives, and therefore would not have the same claim to ownership of aspects of the case. Detective Experience From t he i nterviews c onducted, t here appears to be a s trong p attern emerging i n relation to the issue of investigator experience and the way in which an investigation is managed or constructed. It w as evi dent t hat s ome of t he de tectives i nterviewed w ho were r esponsible f or he ading a s erial r ape or serial m urder i nvestigation ha d no pr ior experience i n investigating s erial cr imes of a similar na ture. The ex ceptions w ere former Superintendent Clive Small who had been involved in a number of serial crime - 248- investigations a t di fferent poi nts dur ing hi s c areer a nd D etective P ryde w ho h ad previously worked in a peripheral role on the investigation led by Detective Faucett. In Detective Faucett’s case, she had been in the position of detective for only a few m onths. Detective Faucett s tates that she and t he ot her de tective w orking t he investigation: …at the ti me w ere v ery i nexperienced in p lain c lothes. W e ha d done plain clothes work but we were the most junior in the office at the time” The decision to put inexperienced detectives in a lead investigative role on a serial case was attributed to the belief that the job was considered difficult whereby: It just seemed that-looking on the surface of it when we first got it there wasthey didn’t believe there was much chance we’d lock someone up. So it was kind of ‘a lot of work, no result job’ to begin with (Detective Faucett). Although some may have been inexperienced in investigating serial crimes, all of t he de tectives i nterviewed had been with NSW P olice f or m any years, the l east experienced having 14 years of s ervice at the time of int erview. This is not to be unexpected as t hey each held the r ole of le ad de tective. Similarly, the de tectives identified in the case studies as leading the investigations into the other serial crimes of interest to this r esearch, ha d all be en w ith their r espective pol ice or ganisations f or a - 249- lengthy period of time and were considered experienced in the investigation of murder and rape cases, although not necessarily serial crimes. The r ole of ex perience i n the i nvestigative t eam or t ask force w as also a significant issue emerging from the data. In relation to the Snowtown investigation, the Acting Police C ommissioner of S outh A ustralia P olice, Neil M cKenzie stated that, “…Never before, in the history of South Australia has the challenge been so great… to investigate a s eries of crimes as a s ingle event” (Boston, 2004). This was despite some very hi gh p rofile s erial murder i nvestigations ha ving t aken pl ace i n S outh A ustralia, such a s t he T ruro m urders of s even w omen i n the l ate 1970’ s (Kidd, 2 000) and t he ‘Family M urders’ i n t he 1970s a nd 1980s , a lthough a conviction w as no t obt ained i n the latter case (Pinto & Wilson, 1990). Playing the Hunch Contrary t o existing literature regarding t he n egative us e of c ase t heories i n t he investigation of c rimes, the int erviews showed t hat ‘following a hunc h’ was u sed f or reasons ot her t han reading t he criminal el ement, as is proposed b y Innes (2003). Following a hunch proved fruitful to one investigation where it was the forensic experts that were i n question. In relation t o t he us e of forensic experts i n t he i nvestigation of Kathleen Folbigg, Detective Ryan states that; - 250- I interviewed those doctors over the phone, and you know how you get that gut f eeling? I j ust di dn’t like th em. I di dn’t l ike th e f act t hat they w ere defensive…. A s a n i nvestigator I don’ t think w e trust ou r g ut instincts enough and our ability to be able to assess people quickly. Detective R yan’s ‘ gut i nstinct’ was proven correct w hen the ex pert evi dence given in relation to another case by the two English forensic experts in question, was overturned in the Court of Criminal Appeal in the United Kingdom. This idea that instinct has a role in the investigation of serious crimes s uch as serial m urder w as further e vident i n t he Folbigg case. W hen di scussing t he c ase, Detective R yan explained t hat w hen he he ard t hat t here ha d be en t hree prior de aths when investigating a f ourth death in the f amily of K athleen Folbigg h e s tated that, “that’s got to sound alarms”. There w as s ome evidence to s upport t he pr actice o f u sing a c ase t heory to develop a n i nvestigative s trategy. U nlike t he example of Harry Blackburn p reviously discussed in Chapter 1, these case theories were formulated based on what appeared to be accurate information received at the time. Small described t he ne ed t o ha ve a ba lanced a pproach t o your c ase t heory, going be yond a reliance on a feeling or a hun ch. Here he describes a s ituation where one of the investigators was focused on proving a case theory against Milat: - 251- …when a particular member of the investigation team kept coming up to me and saying. “Ivan did it. Ivan did it. Look, I’ve got this, this, and this” which shows me there certainly was a bui ld up of evidence that strongly suggested he did it, and my comment was something along the lines of, “That’s terrific. Now go and tell my why he can’t have done it.” Small explained that his response had been received with some officers questioning his commitment to the proposed case theory and subsequent evidence: And there was a bit of comment by a few people, saying why does he keep saying tell me why he can’t have done it, doesn’t he believe he did it? That wasn’t the point. Small explains his reasoning for these responses, using an example where the use of a case t heory i n an investigation into a s eries of s exual as saults l ed to an inaccurate investigative outcome. He continues: What I was trying to say was we needed a very tight case and it isn’t about finding t he i nculpatory evidence, it’s about f inding all of t he evidence whether i t’s i nculpatory or e xculpatory…. A nd that means be ing t horough about the w hole of t he e vidence, and the H arry B lackburn case i s an example. They had made their mind up t hat this person did it, so they found all this e vidence t hat t hey t hought s howed t hat he d id i t a nd pus hed a side evidence that exculpated him. At the end of the day, what happened was that - 252- they t otally destroyed all the i nculpatory evidence against any offender and they had the wrong person. Detective Small had previously asserted that the objectivity of a case theory as being an important approach to an investigation. One of the problems you have with an investigation like this is that you can often get a lead which looks very good. What you have to be careful of is that you don’ t pu t a ll y our e ggs i n one b asket a nd t hen f ind that the lead w as completely w rong a nd y ou’ve w asted the m ajority of y our r esources e ffort chasing down false leads. Even when Milat was increasingly e merging a s a prime suspect we still had another million leads to follow and you couldn’t discard them. And you c ouldn’t discard t hem on t he ba sis of: one , it might have i dentified the r eal of fender; t wo, i t m ight ha ve pr ovided e xtra information inculpating Milat, or further providing information exculpating. So there’s a number of things. There is I guess, some people who take a risk and say that’s the lead that’s where it’s going and it works out and that’s fine. But it can also be very disastrous for you. The importance of an impartial investigation was also echoed by Detective Ryan. In describing the investigation into the deaths of the Folbigg children, he explains the approach investigators should take in investigating such deaths: We pr obably think of i t m ore as a su spicious de ath. We'd have t o start t o narrow our investigation to look at it as murder, but murder is the intentional - 253- killing o r reckless murder. What we base ou r investigation on i s the inquisitorial search for the t ruth and t hen if you' re l ooking a t it a nd establishing t hat truth. I f I ’m r unning a m urder investigation now , I w ould instruct no t t o put i n t hings l ike t hat ' murder'- it a ffects the in tegrity of yourself. O ur job i s to g et a ll the f acts w hether t hey’re for the accused or against the accused; you put it all in there. The l esson not t o direct r esources ba sed entirely on a cas e t heory w as learnt during t he G lover i nvestigation w here a significant pr oportion of resources ha d be en allocated to what w as later di scovered to be an inaccurate l ead. Early on i n t he investigation a witness s tated that the y ha d seen a young ma le in the vi cinity of the crime. This l ead was picked up on, and influenced the r esponse t o ot her i nformation received during t he i nvestigation. The cl osed cas e f ile indi cates th at s ignificant resources were di rected based on the case theory that the offender w as a young male, for example the investigation of schools situated in the area. This case t heory also i nfluenced t he way i n which ot her i nformation had been classified. Priority was given to information relating to young people. A comfit of the young pe rson of i nterest w as di stributed t hrough m edia out lets, r equesting a ny information about the case or that person. This subsequently resulted in an investigation that was misdirected which impacted both on the way in which the resources were used but al so the publ ic aw areness of t hese crimes. Once i t w as r ecognised that a m ature aged of fender m ay be r esponsible, c onnections were m ade be tween r eported s exual assaults at nursing homes in the area. - 254- From the Glover closed police case file it was also evident once this case theory had be en que stioned a nd a ne w t heory f ormed, t hat t he of fender may be ol der, investigators were forced to return to previous witnesses and question them about this new suspect. This illustrates Detective Small’s point about the dangers of allocating all resources to the one case theory. While this t heme ar ose in relation to all of the N SW-based cases of serial murder, t he us e of a c ase t heory i n s erial r ape cases w as e vident i n onl y two of t he cases. Unlike the serial murder cases, where the case theory related to the way in which the investigation proceeded, case theories of serial rape investigations often appeared to have a negative impact on the investigative process. Prior t o t he 2004 i nstigation o f T ask Force G rahame, t he s eries of s exual assaults committed by Trindall was not subject to investigation. The records available indicate that although some of the forensic evidence was forwarded on to be processed, no formal investigation into any of the individual incidents took place and no follow up on t he e vidence oc curred. T his i naction r eflects the theory t hat t he i ndividual c ases were not seen t o be a profitable us e of r esources. S imilar ina ction during the early stages of the investigation into the crimes of the serial rapist, Presta, reflected on a case theory that the complainants were not real victims. Instead they were considered by one detective to be mischief makers. The underlying basis for the development of such case theories could perhaps be better understood in the context of the dictionary knowledge that investigators use. - 255- Dictionary Knowledge and the Conceptualisation of Victims As discussed in previous chapters, police dictionary knowledge is a dimension of police cultural pr actices (Chan, 1999b) . A ccording t o C han t his di ctionary k nowledge i s usually used by police to form judgements on the criminality of people or groups. This use of di ctionary know ledge l eads t o s tereotypical r esponses t o pa rticular groups o r people. One of t he s trongest t hemes t hat emerged from t he r esearch was t he w ay i n which the use of stereotypes could be seen to reflect on the investigative practices. This was evident with the investigation of both serial murder and serial rape incidents. Although Chan (1999b) considers dictionary knowledge to usually relate to the labels a pplied t o of fenders, t his w as f ound t o a pply t o t he vi ctims a s w ell. F rom the interviews, it w as e vident tha t in some cases this s tereotypical knowledge m ay h ave been beneficial t o the vict ims of s ome of these cr imes. W hen di scussing t he vi ctims Detective Faucett commented that “ …These were j ust r egular ki ds. They weren’t crooks, t hey w eren’t out us ing [ drugs].” a nd l ater, “ …these ki ds w ere on t heir w ay home from parties- decent kids just on their way home…”. It is clear from the detective’s comments that these victims were considered as vulnerable. T hese w ere not s treet w ise youths w ho ha d be en i nvolved i n i llegal activities. One vi ctim was even sympathetically described as not being a “particularly bright kid” who had thought he was simply getting a ride home. This more positive use of s tereotypes m ay be s een t o i nform t he di rectory know ledge of t his detective. In - 256- response to the claims made by some media reports that some of the victims were street kids, t he de tective ha d f elt t he ne ed to emphasise t hat t hese w ere ‘normal, i nnocent’ rather t han ‘street w ise’ victims pe rhaps va lidating he r i nvestigative r esponse t o t he cases. Detective F aucett l ater ex plained that there w ere di fficulties i n tracing ear lier victims of Presta as: He w as charged but i t w as withdrawn ‘cause t he offender- being st reet kids most of them- some of them were turned away. We did find out that some of them were turned away when they originally went to report it. Most of them are drug addicts now. Some of them you’ll never ever be able to identify. Dictionary knowledge as constructed by different investigators, results in a varied set of investigative s cripts de riving from t he s tereotypes be ing u sed t o i nform t he w ay i n which they analysed a situation. In the Presta case, this resulted in stereotyping which negatively impacted the victims and also extended to the offender. A different opinion of these victims was held by at least one of the detectives involved in the investigation. When discussing the reportage of one of the incidents in the series of crimes Detective Faucett said about one of the detectives, …he [the victim] was in the hospital one of the police officers that had went to t ake t he or iginal report had a n a rgument w ith h im a nd t old him he w as lying a bout it. B ecause h is s tory k ept c hanging. T urns out h is story di dn’t change at all it was just a long story and he was telling parts of it and he had been given Rohypnol and so he was having [sic] in and out of memory. That - 257- report was nearly not made because they didn’t believe him. He had cigarette burns all over him and whip marks everywhere but they still hadn’t believed a word he said. And every one of the kids interviewed had said if they hadn’t been picked up by an ambulance they wouldn’t have reported it at all. This w as not s imply a matter of reasonably qu estioning a c omplainant’s ve rsion o f events. In this case, one of the victims was perceived as an offender. Detective Faucett continues: The one from South Penrith was interviewed [unclear] and they were going to charge him with public mischief ‘cause they didn’t believe him either. So there was a major problem with policemen believing that this could happen to m ales I t hink. A nd t hat w as a m ajor p roblem dur ing t he w hole investigation I f ound it really di fficult t o talk to him a nd really di fficult during the search warrant and everything. It was really quite hard. This perception may be considered to be the product of the dictionary knowledge that males are the offenders of sexual assault rather than the victims. Such an attitude can be s een t o reflect s ome of t he common myths and stereotypes s urrounding m ale s exual a ssault, n ot j ust i n pol icing but i n t he general community, a s out lined b y M cAdam a nd F itts ( 1999) i n t heir r esearch i nto t he responses to the reporting of male sexual assault. Common myths include the belief that males c annot b e vi ctims of s exual a ssault or tha t hom osexuality is a f eature of o r a - 258- result of such acts. Although the behaviour described may not be entirely attributable to the dictionary knowledge that males are rarely victims of sexual abuse, in the situations described this g eneral at titude is cl early reflected i n t he di rectory knowledge; that is , how the investigation has been carried out based on t his information. As evidenced in the ea rlier quot e, at l east one o f t he vi ctims w as threatened w ith criminal cha rges for wasting pol ice t ime. In another c ase t he r eport nearly w ent unr ecorded be cause t he victim had not been believed. Detective Faucett later commented on the behaviour of the detective in question during t he execution of the search warrant and arrest o f t he o ffender i n t his cas e. In describing t he de tective, F aucett de scribes how “ he w ouldn’t g o a nywhere ne ar t he bloke during the search warrant. You could feel the hostility.” As this was spoken in the context of the detective’s response to the of fender one may surmise that the detective also had issues arising from me n committing the se c rimes a gainst ot her ma les or perceived homosexuality i n general. T his w ould be c onsistent w ith t he b elief i n r ape myths discussed by M cAdam and Fitts (1999). S uch ope n hos tility could c ause problems when a ttempting t o bui ld a r apport w ith t he of fender dur ing t he i nterview process. This l ack of an appropriate r esponse towards t he ve racity of t he vi ctims w as also evident in the investigation into the brutal sexual assaults of six female prostitutes over a nine year period. From the available records, it is clear that there was minimal investigation of these sexual assaults against these women at the time of the incidents. - 259- In t hree cas es a cr ime s cene examination was c onducted and evidence was col lected, (Begg & Norris, 2006) although the Brief of Evidence shows t hat some of t his w as destroyed after a p eriod of t ime and w as not a nalysed. Even though all cas es w ere reported t o pol ice, f ew vi ctims ha d their formal s tatements r ecorded at t he t ime of reporting. This i nconsistent pr actice c an, i n p art, again b e a ttributed t o t he di ctionary knowledge of c rimes a gainst pr ostitutes. A ccording t o B ridgett a nd R obinson (1999) prostitutes are one group in society that are marginalised and subsequently vul nerable to s uch a cts of vi olence. A New S outh W ales Standing C ommittee of Social Issues report (Legislative C ouncil, 1996) found t hat t here i s a n extremely hi gh incidence o f violence committed against pr ostitutes although the r eporting of th ese crimes is low . Many of these offences involve the use of a weapon and result in significant physical and ps ychological i njury t o t he vi ctims. T he r eport a lso f ound t hat t he offenders of these c rimes of ten h ave previous c riminal hi story of vi olence or s exual of fences a nd thus target prostitutes as they feel they are less likely to be convicted of these offences (Legislative Council, 1996). Bridgett and R obinson (1999) discuss t he l egal restrictions pl aced on the s ex industry i n ge neral, t hat f orce s ome pr ostitutes i nto uns afe w orking e nvironments, which increases t he r isks t hey face. There i s f requently an i nadequate o utcome i n t he criminal jus tice s ystem. Bridgett a nd R obinson s tate t hat vi ctims w ho ha ve r eported incidents of s exual assault t o pol ice have felt t hat t hey w ere not b elieved and di d not - 260- receive a ny s upport t o assist i n f ollowing thr ough with their allegations. Police ar e inherently in the position to be aware of this issue and to some degree at least, of the extent of vi olence e xperienced. T herefore t heir di ctionary know ledge i s i nformed b y this, and t hus s o i s t heir s ubsequent di rectory know ledge about how t hey s hould respond to these crimes. There i s an awareness t hat m ost of t hese cases do not achi eve a s uccessful outcome i n c ourt. Police are a lso a ware of a num ber of pr oblems a rising f rom t he mobility of s uch victims, often m aking i t ha rd to f ollow t hrough on aspects of t he investigation a nd pr osecution. T his pr oblem w as r ealised i n t he c urrent c ase, w here initial investigators had trouble tracing the victims to get formal statements which were not made at the time of the report. As indicated by Begg and Norris (2006), this led to the i nvestigations s talling. According t o S tewart’s ( 1999) r eport i nto t he r esponse of NSW Police t o cases of s exual as sault, t his t endency not to take s tatements w as i n accordance with police protocols at the time of some of the offences. Detective McGann has detailed the problems he experienced in the investigation into t he di sappearance and de ath of J odie Larcombe i n 1988 . 26 In t his c ase, t he problems e xperienced w ith vi ctim pe rception w ere not w ith t he N SW Police F orce, who ha d s upported h is i nvestigation, but w ith t he D irector of P ublic P rosecutions (DPP) who had initially proceeded to ‘no bill’ the charges against the offender, Daryl Suckling. M cGann de scribes i n bot h hi s i nterview a nd a n a rticle w ritten detailing t he 26 Although this case was outside the research time frame, the interviews were not limited by this. There were significant delays in the prosecution of this case which went on into the 1990’s - 261- case, that a s olicitor with the DPP had told him, in regards to the victim, that it was “a shame that she (Larcombe) was not a better class of victim”, which McGann attributed to the knowledge that Larcombe was a prostitute. Further s tereotyping of vi ctims a lso oc curred i n t he investigation of t he Snowtown serial murders. Acting Police Commissioner of South Australia Police, Neil McKenzie was quoted in the media as describing both the victims and offenders as “a group t hat pr eyed upon i tself” (Boston, 2004) . C ommissioner M cKenzie ha d been alluding to the homosexuality of many of the victims and the homosexuality of some of the offenders. This is a clear example of what Fattah (1997) refers to as the ‘worthless victim’. A r eview o f a ll t he cas e s tudy m aterial shows t he abs ence of a pol ice investigation into many of the victims once they had been reported missing. According to P udney (2005) and as illus trated in Table 10 i n Chapter 4 , t he investigation into the disappearances did not commence until after Suzanne Allen was identified as missing. She did not belong to this ‘group’ in that she was not homosexual but the wife of one of the offenders, and therefore a more traditional victim. It was only after he r di sappearance w as l inked t o t hat of t wo of t he ot her vi ctims t hat f urther investigation into the other persons reported missing was initiated. This discrepancy in the initiation of an investigation can be considered to demonstrate the extent to which police cul tural pr actices ar e pr evelent in the inve stigation of s erial c rime. This discrepancy i s a lso a r eflection of t he c ultural know ledge a nd shows how s uch - 262- knowledge f orms a n un derstanding of di fferent groups i n s ociety, thereby ins tructing the directory knowledge from such understandings. The Missing A theme that emerged from the case study an alysis was that victims reported missing are al so typecast, which is c onsidered t o ha ve i nfluenced t he w ay i n w hich t heir disappearances are ini tially inve stigated. M any of the vi ctims’ families ha ve b een reported a s commenting t hat t hey w ere t old b y police t hings s uch a s ‘ young pe ople often f orget t o c ontact hom e’ or ‘ backpackers of ten g o “ missing”’. Many of t he relatives overseas spoke of being brushed off by allegedly unconcerned police in NSW. According to Whittaker and Kennedy (1998), when the mothers of Everist and Gibson went t o r eport t heir c hildren m issing t hey were not t aken s eriously a nd were quot ed missing pe rson statistics. According to Whittaker a nd Kennedy, “It w as n ot a s erious case in the eyes of the police- two nineteen-year-olds off on a bit of a skylark. If they investigated every such case, they’d be doing nothing else, especially in mid-summer” (p. 189) . When it is c onsidered how ever, t hat of t he 35 000 pe ople w ho a re r eported missing e ach year in Australia, a ma jority do return home w ithin a f ew w eeks, t he attitude of police is both substantiated by fact and thus understandable. This response to reports of missing victims is further supported by a statement by Superintendent Small. When asked about the investigations into the disappearances of the M ilat vi ctims w hen they were ini tially reported as mis sing, D etective S mall - 263- admits that the re ha d been a limite d investigation however jus tified this inve stigative approach. He states that: There w ould have been v ery l imited i nvestigations u ndertaken at that time because they w ere si mply…they w ere backpackers. They had no friends or relatives he re, their con tact ha d been… the r eporting by t hem of their movements had been quite intermittent, so you really had very little to go on. This s tatement is r eflective of the kno wledge o f mis sing pe rson incidents, where in Australia the majority are located within a few weeks. This statement also supports the proposition by Fattah (1997) and Fox et al. (2005) that marginal groups are targeted by offenders due to the lack of reporting of such disappearances to authorities. Detective Small’s c omment tha t t he vi ctims’ c ontact w ith family was int ermittent pr ovides support for this aspect being a marginalising factor for this group. This in turn impacts on the a bility of pol ice t o i nvestigate t he di sappearances. T he di ctionary know ledge internalised b y pol ice w ould ha ve i dentified t he l ikely out comes of s uch a n investigation. T his poi nt is el aborated on by Detective S mall when di scussing t he probable investigative o utcomes a fter the di scovery of the vi ctims b y the de tective when he stated: …the victims were backpackers which meant they had no close connections in the i mmediate Sydney ar ea o r t he B owral-Belanglo area, s o y our opportunity for leads were less than might be the case. You also had the fact that there h ad been a degrading of t he crime scene simply be cause o f the - 264- weather over several years, and trying to track the immediate movements of the v ictims w as al so very difficult. Because t hey w ere si mply ba ckpackers travelling alone. The nature of these victims and the features which characterise these victims were seen to c ontribute n egatively to t he pos sible out comes of t he investigation. T he t raditional sources of i nformation w ere not a vailable, a s Small de scribes t he op portunities t o receive leads and information from the crime scene had been minimised due to both the nature of the victims and the timeframes of the crimes. This problem of gaining information related to missing persons was apparent in the Snowtown case. As outlined in Table 7 in Chapter 4, one of the victims had been discovered in August 1994. Police, however, were hampered in their investigations as they were unable to determine the identity of the victim. The victim, Trezise, would not be r eported m issing unt il 1995 (Pudney, 2005) . Thi s delay i n r eporting Trezise as missing was due to his volatile relationship with his family, who had assumed that he had s imply l eft (Pudney, 2005) . P udney de scribes how e ven a fter t he m issing pe rson report was m ade, Trezise’s r emains w ere not i dentified due t o an er ror made b y t he expert at the Forensic Science Centre leading to a four year delay in identification. - 265- Dictionary Knowledge and Serial Crime Although i t i s e vident t hat di ctionary know ledge i s e mployed when investigating the serial murder and serial rape incidents, this knowledge was not specific to these types of crimes. This may relate to the lack of experience as the large proportion of detectives interviewed had little e xperience w ith investigating vi olent s erial c rimes as ide ntified previously. It w as not ed t hat i n t heir de scriptions of t he of fenders a nd t heir c rimes, none made any connections to other cases of serial murder or serial rape, nor did they justify the use of particular investigative tools or techniques based on p rior experience with t hese i nvestigations. T his w as de spite t wo of t he d etectives i nterviewed h aving prior experience, one of whom had extensive experience i n t he i nvestigation of s erial rape. This l ack of kno wledge a bout inve stigating serial c rimes w as r aised by K idd who describes how during the Glover investigation police: …had never experienced such a case, investigators had nothing to go on. There were no g uidelines to s teer t hose trying t o find t he elusive m urderer” (Kidd, 2000, p.237). Therefore, as a l ack of ex perience m eant t hat t here w as i nsufficient di ctionary knowledge to aid investigators in understanding these crimes, this subsequently meant that t he cul turally appropriate r esponses t o these cr imes, and the act ions ne eded to - 266- achieve organisationally acceptable outcomes were absent. Thus, the requisite directory knowledge which is usually informed by the dictionary knowledge could not exist. Use of Language to Understand Investigations As discussed in Chapter 1, dictionary knowledge of police can be expressed through the use of specialised terms and references as a w ay of ex pressing the shared understanding, as proposed by Devery (2003). There were few examples of ‘specialised language’ present in any of the case studies or the interviews. There was one example of w here t his s pecialised know ledge extended t o m ore m undane t asks s uch as paperwork. From t he cl osed police c ase file of G lover, running s heets w here bundl ed into f our set cat egories- “1. V ery i nteresting, 2. Interesting, 3. A verage, 4 . H o-hum”. From the content of these running sheets it was evident that the information contained within these cat egories reflected not onl y the i mportance of s uch i nformation, but presumably the level of attention each received. This however, was the only example of the us e of pol ice di ctionary kno wledge i n t he i nvestigative pr ocess e vident f rom t he closed case files and Prosecutorial Briefs of Evidence. Less spe cifically ‘ specialised’ l anguage emerged in a few i nstances during t he interviews. One ex ample w as w hen Detective Ryan was de scribing the pr ocess of initiating the investigation into the deaths of the Folbigg children, and commented that the previous deaths of the other children, “ought to be enough to start ringing the bells. Sound a n a larm.” T his i s a phr ase that doe s not r equire s pecialised know ledge of t he - 267- cultural me anings to be unde rstood. A nother s imilar statement to this came f rom Superintendent S mall w ho, i n de scribing t he p ossibility of approaching s chools for information during the Milat investigation said: And so writing to all the schools saying, look any teachers who were at these schools, did you remember a kid who was a bit of a bastard (laughs), kind of thing, would have stood about because of this, this and this, cause he would have been a bit of a control freak in those early days. The us e of t he t erms ‘ bastard’ a nd ‘control f reak’ a re colloquial t erms w hich ha ve a shared und erstanding among the general c ommunity, rather t han a s pecific understanding based upon the dictionary knowledge of police. The a bsence of ot her examples of t his s pecialised l anguage f rom bot h t he records and the interviews can be explained in two ways. First, it would not have been professional on t he pa rt of pol ice to i nclude s uch t erms i n doc uments t hat w ould or could later form part of the Brief of Evidence. The specialised language is a m eans of informally communicating information in a way that is meaningful to the receiver, so that b y using a particular term, i t would express both t he di ctionary and t he di rectory knowledge. In t he above e xample f rom t he G lover c ase f ile, t he di rectory knowledge can be considered intrinsic to the labelling of the information, thus the label applied to each ( dictionary kno wledge) pr ioritised t he r esponse t o t he c ategories o f i nformation (directory knowledge). - 268- The l ack of a pparent i nclusion of s pecialised l anguage i n t he i nterviews w ith detectives may be attributable to the fact that the researcher is not with NSW Police or any other police organisation and therefore does not share the cultural understandings of s uch l anguage. T his may also e xplain t he us e of m ore c olloquial t erms w hich t he interviewees may h ave gauged w ould obt ain a greater l evel o f unde rstanding on t his researcher’s part. Chapter Summary The use of c ase theories to inform the investigation of serial crimes w as seen to have varied consequences. It was found that the use of intuition, in the form of following a hunch, w as found t o ha ve be en us ed i n s ome cases, how ever t here was i nsufficient evidence to support that this experience informs the way in which detectives investigate these crimes. The conceptualisation of the victim as worthless, or not credible, has led to problems with the investigation of serial crime in a number ways. This can be seen most pa rticularly i n t he l ack of di ctionary kno wledge a nd t he i nfluence t his ha s on directory knowledge s pecific t o bot h s erial c rime i nvestigations, a nd t he vi ctims of these c rimes. T his l ack of s pecific know ledge w as de monstrated t o i mpact on t he investigative pr ocesses t hat w ere e mployed and t herefore, t he pr ogress of t he investigation. - 269- CHAPTER 8: THE CULTURE OF INTELLIGENT INVESTIGATIONS As di scussed i n t he p revious c hapter, t he i nvestigations of s erial c rimes r eflect the police cultural knowledge that pervades these organisations. Police cultural knowledge and practices can be s een to shape t he i nvestigations. C hapter 8 will di scuss the emergent themes relating to the incorporation of intelligence-led policing practices into the investigation of serial crimes in Australia. As will be demonstrated, police cultural knowledge c an a lso be seen t o s hape the w ay i n which intelligence-led pol icing ha s been integrated as an investigative m odel. This c oupled w ith t he us e of ot her investigative tools will be examined. Crime Linkage Timeframes of Recognition One of the key areas of interest to this research is the ability of police to link a series of related crimes. This was determined by establishing the timeframe between the second crime 27 and the f ormal r ecognition of t he cr imes as s erial b y pol ice, as ascertained from either police closed case files, interviews, newspaper articles or other 27 Since at least two crimes must occur before they can be considered serial, police would be unable to ‘link’ incidents until after the second crime had been committed. - 270- biographical accounts. As the nature of the offences varied, 28 the time since the second contact or r eport t o pol ice w as used as the s ignpost to establish the tim eframe. The impact of thi s ide ntification of a s eries of c rimes a s s erial on the out come of the investigation, measured as t he achievement of an arrest o f a s uspect, i s a lso of consideration. A factor that will affect both timelines is the frequency of offending. For example, in the Glover case, a number of crimes were committed over a relatively short period of time, while in the Milat case there were several months, and at times a year, between some offences. Time from Second Offence to Recognition of Crimes as Serial From examining the i ncidents of s erial m urder and s erial rape, there ap pears t o be a delay in the f ormal recognition of t he crimes in the cas e s tudies as s erial. The recognition of a pos sible l ink be tween of fences usually m eant t he e stablishment of a task force t o investigate t he s eries of cr imes, w ith the ex ception of t he G lover investigation. Although from the c losed case f ile it is c lear tha t the m urders w ere considered linked, it was only after the fifth murder that the North Shore Murders Task Force was formally set up. Only two incidents of serial murder were linked soon after the second offence; being t he Glover a nd D enyer i ncidents. In bot h c ases, t he pol ice r ecognised t he l ink 28 Some victims were reported missing several months before their remains were found and others, as victims of sexual assault reported crimes as they occurred. - 271- between t he s econd a nd f irst i ncidents r elatively qui ckly. In t he D enyer c ase, t he victims w ere f ound i n busy, u rban a reas and were young women i n a nd around Frankston in Victoria (Petraitis, 1995). The nature of the offences, the way in which the victims were murdered, also enabled recognition of the possibility of these crimes being linked. That they occurred over a two month period also reduced the linkage blindness often apparent in serial crime investigations. Similarly, in the G lover c ase the vi ctims w ere all f ound within the r egion of Sydney’s North Shore, over a period of months, in similar circumstances. Although an initial a ssault w as not r eported f or ne arly 12 months, onc e t he s econd m urder ha d occurred pol ice r ecords indicate t hat bot h m urders w ere t reated as t hough t hey were related. Following the discovery of the remains of Clarke and Walters during the Milat investigation, police released a statement to the media stating that they did not believe that further bodies would be discovered (Bellamy, 2004). This could be seen to indicate that although they had recognised that several backpackers were missing, they were not preparing to investigate them as related crimes. The investigation was reported to have stalled within a f ew w eeks of t he di scovery o f C larke and Walters, as the ph ysical evidence that was col lected did not provide information that assisted the investigation (Bellamy, 2004). Other similar missing person cases were not considered at that stage of the investigation. It was only after the second set of remains were discovered some - 272- 12 m onths l ater that t he pos sibility of l inks t o ot her m issing pe rson c ases w as considered. The Snowtown murders involved more complex processes linking the victims, as some were reported missing, while others were not. Although some victims had been linked during the investigation, this could not be confirmed until the remains of all 11 victims w ere f ound ov er a pe riod of days and t he pr ocess o f i dentification was completed. D ue t o t he n ature of t hese vi ctims’ l ifestyles, a num ber of t hese vi ctims were not for mally reported missing to police in South Australia making it difficult to ascertain an accurate t imeline f rom t he s econd r eport of a m issing person t o t he establishment of the task f orce t o i nvestigate t he l arge num ber of m issing pe rsons within the small community. Timelines from the second offence to the recognition by police that crimes were linked t ended t o be l onger f or t he i ncidents o f s erial r ape c onsidered b y t his c urrent research. In the Trindall case, there was a significant delay between the second offence and the r ecognition that i t w as pa rt of a s eries. This s even year de lay w as due t o an inappropriate d elay i n t he a nalysis of t he forensic e vidence of a num ber of s exual assault cas es (Norris, 2 006). Extensive de lays were also r eported i n t he J ohnson investigation. Although similarities w ere r ecognised after the f irst e ight vi ctims, the resources a llocated i n l ine w ith a n i nvestigation i nto a s eries of of fences w as not forthcoming until some 13 months later (Riley, 1997). This followed a “massive botchup” on t he part of NSW Police, as described by the then Acting Police Commissioner - 273- Clive Small which included the failure of four area commanders to coordinate a joint inquiry into these offences (Riley, 1997). In the Presta case, the crimes were only recognised as serial in nature following the fifth and final offence in this series. This was eight months after the second offence took place. This l ink was es tablished when Detective F aucett de scribes t hat af ter t he fourth offence, “It was only a number of days later that we got another call in relation to a younger man” who described an assault similar to the fourth offence. It was then that ot her c ases e merged t hrough v arious i nformal m easures, as w ill be di scussed below. It is unclear in the Daley case as to when the cases were initially identified as serial i n nature. However f rom t he s ources av ailable, it appe ars t hat this w as not formally r ecognised unt il t he s ixth of fence 29 which led to the e stablishment of S trike Force S apelo. T he s eventh a nd f inal of fence i n t his s eries oc curred 1 1 da ys l ater. Therefore, this case appears to follow a similar progression to the Presta case. Time Until Arrest from Recognition of Series of Crimes Egger (1990a, 1998 ) proposes t hat delays i n the r ecognition of a s eries of cr imes as such could impact on the achievement of a successful investigative outcome. Incidents 29 There were a total of eight offences for which Daley was investigated, however one victim was considered a non-complainant and charges were not laid in relation to this offence - 274- of serial rape considered here had shorter timeframes between the recognition of crimes as s erial and the ar rest o f t he of fender t han the i ncidents of s erial m urder cons idered. Three of the cases of serial rape had timeframes between recognition and arrest of less than one m onth. In t he Presta c ase, t he of fender w as a rrested t he w eek f ollowing t he recognition that t he cr imes w ere r elated. In the T rindall cas e, the t ime be tween recognition and arrest was approximately two and a half months. This can be attributed to t he r equirement of t he t ask f orce t o t race pr evious vi ctims a nd obt ain t he f orensic evidence t hat ha d b een stored a t di fferent l ocations de pending on w here t he a ssaults were reported. This time range of between seven days and two and a half months could not be seen to be i nfluenced significantly by t he l ength of t ime ov er w hich the cr imes ha d occurred. As such, although the Presta, Daley and Johnson cases occurred over a period of be tween a pproximately 18 t o 24 m onths, a ll i nvestigations r esulted in a n a rrest within a m onth of i nvestigators r ecognising t he serial na ture of t he crimes. A lthough the longest investigation was the Trindall case, where the recognition of the incidents as part of a s eries t ook s everal years, t his t wo and a h alf mont h timeframe c ould be explained by t he ci rcumstances of t he of fender. As he ha d been arrested and charged with t he a ssault of hi s f emale c ompanion, he was unde r a b ail or der t hat m eant t hat police c ould m onitor h im. T his a llowed f or t he c ollection o f evi dence and the verification and formalisation of the links drawn between all cases. - 275- Although the time range from the recognition of serial rape cases to the arrest of the offender was relatively narrow, a comparative range for serial murder investigations were more varied. For serial murder cases, the timeframe ranged between 17 days and two years. This can be s een to be correlate di rectly with the time between the second offence a nd t he r ecognition of t he c rimes a s s erial i n na ture. T herefore t he D enyer investigation a chieved an a rrest after s eventeen da ys f rom t he da te when i t w as recognised as serial in nature and a task force was set up to investigate these crimes. At the other end, Folbigg was arrested two years after the recognition that the death of her last child was part of a series of murders. This case had the longest timeframe between the second offence and recognition of the offences as being serial in nature, with a nine year time frame. This de monstrates the impor tance of r ecognising c rimes as s erial i n nature e arly a nd a llocating r esources a ppropriately thus supporting Egger’s (1990a, 1998) contention t hat a r eduction i n l inkage bl indness i ncreases t he l ikelihood of a successful investigative outcome. Recognition of Crimes as Serial As previously outlined in Chapter 2, the ability of investigators to link related crimes is an issue that has been identified by many academics as important to the investigative outcome. As identified, one of the most significant themes that emerged from both the interviews a nd t he c ase s tudies, w as t he i nability o f pol ice t o r eadily l ink r elated offences. In all the cases represented by the interviews conducted none was able to use any formal system to link related crimes. This was despite lengthy research timeframe - 276- that covered t he i ntroduction of t he formal n ational l inkage s ystem, ViCLAS. A s discussed in Chapter 2. the ne w i ntelligence s ystem V iCLAS w as s et up a f ew years after t he M ilat c ase, a nd w as de signed t o analyse i nformation a bout s olved a nd unsolved hom icides, s erious s exual a ssaults, a nd l ong t erm m issing pe rsons f rom around the nation. When asked about t he i nvestigative pr ocess of i dentifying r elated cases, a similar response was elicited from all participants: that difficulties were experienced in all cases as they were without an adequate tool. Of the closed case files reviewed, only one cas e app ears to ha ve c ompleted ViCLAS s ubmissions. T his was f or t he c ase o f Terry John Williamson, which was not a focus in the case study methodology. It should be noted that these submissions to ViCLAS were made after the case had been solved and therefore could not have assisted in the investigation of these crimes. When asked whether t here were an y problems i n the i dentification of s imilar offences as sociated with the s eries of r apes c ommitted by P resta, Detective F aucett responded, Major problems. The only reason we found them was because we knew they existed. We knew they had come from the Emu Plains and Penrith area so I knew to narrow it down…. The Liverpool one I knew about. I knew who the bloke was so that was no problem finding I could just give him a ring and ask about that one. B ut the two f rom E mu P lains/Penrith i t w as onl y be cause I knew they had been in the paper. I knew that they were both boys who had - 277- been dumped, that I knew where to look for it. It still took me a whole shift to a ctually f ind those e vents. I f I ha dn’t k nown a bout them pr eviously I wouldn’t have found them. Interestingly, Detective Faucett had earlier commented that, We t hen…I…when I w as c alled ou t f or the s econd one a m ember of m y family had said to me “it sounds very much l ike something t hat was i n the paper from Emu Plains” and we remembered because the Comfit in the paper reminded us of s omeone w e k new qui te w ell. S o I looked up - got on t he computer system and found that offence and that was in relation to a 16 year old w ho w as pi cked up f rom a pa rty i n E mu P lains a y ear be fore- around about a year before. While it w ould now be e xpected t hat t he r eliance on i nformal m easures and local networks t o l ink r elated c rimes w ould no l onger be ne eded du e t o t he existence of systems like V iCLAS, this is not the c ase. Although ViCLAS was i ntroduced i n t he mid 1990s, it has now been ‘decommissioned’ in NSW and therefore does not submit cases any longer. This diminishes the effectiveness of the initiative. This could also be seen to indicate th at the linkag e of s uch crimes ma y occur m ore r eadily through informal methods, relying heavily on a sharing of knowledge at a local level. The r easons f or the d ecommissioning of V iCLAS were not imme diately apparent. T hrough c ontact w ith the V iCLAS personnel of ficer a t N SW P olice it - 278- emerged that this system was, at one time, under review but with no immediate plans to continue s ubmitting c ase in formation for inclusion on this s ystem. W hile no submissions appear to be made, it was apparent that the facility to request analysis of the s ystem f or s imilar c ases is s till pos sible. A lthough as N SW P olice ar e not consistently submitting information one could conclude that any analysis would not be an accurate reflection of incidents of murder, sexual assault, long term missing persons and the other cases it was designed to monitor. The i mplementation a nd us e of V iCLAS w as a lso c onsidered pr oblematic t o some e xtent b y t hose with t he oppor tunity t o us e t his s ystem. S ome of t he cases represented by the interviews occurred prior to the establishment of ViCLAS whilst the crimes i n t he F olbigg c ase i nvestigated b y D etective R yan w ould not have m et t he criteria f or i nclusion on t he da tabase i nitially, as t he d eaths of t he children w ere attributed to natural causes. Therefore the only cases where inclusion on ViCLAS was expected were the serial rape incidents. From the interviews it w as evident that there were some other issues that arose in t he ut ilisation of t his s ystem, i n a ddition t o the de signed pur pose of t he s ystem. Detective P ryde d escribed the p roblems w ith submitting a case to ViCLAS and t he unmet expectation that followed: ViCLAS themselves did not assist us in any way in this particular matter. As a m atter of fact I got no f eedback w hatsoever from ViCLAS. I ’m not s ure whether they’ve ev en looked at t he matter o r no t. It was more a c ase of - I - 279- don’t really have a reason why- why they didn’t or why they hadn’t, perhaps it was a resource issue, I don’t know. As the offender about whom Detective Pryde was seeking information about was in the armed f orces a nd ha d be en l inked t hrough i nformal m eans t o ot her i ncidents i n ot her states in Australia there is a reasonable expectation that ViCLAS may have been able to make some contribution to the investigation, if only to actively eliminate possible links to other cases on the system. There a lso a ppeared t o be pr oblems i n t he s ubmission of i nformation to t he system, particularly in relation to whose responsibility it w as to complete the ViCLAS form. In relation to the P resta cas e, Detective F aucett’s r esponse t o whether V iCLAS submissions about the incidents had been made was: Detective- Yeah…because t he B ehavioural Sciences [ Unit] ha d g one to Liverpool to look at the original case. They’d done a backtrack on it. Because when I went over to pick up the case file there was a heap of stuff missing out of it that they’d taken from the case file over there. Interviewer- So they had taken that to submit an actual ViCLAS - ? Detective- I be lieve s o, y eah. They w ere l ooking a t t hat case. They w ere looking at profiling that case to see if they could find anything else but there was no thing a t t hat st age t o come of it. I t ha dn’t b een l inked t o t he E mu Plains one. - 280- This ex change i ndicates t hat t here may h ave b een a m isconception about both t he ‘profiling’ pr ocess a nd t he pr ocess i nvolved i n making a ViCLAS s ubmission. T he discussion of profiling i n relation to ViCLAS could indicate a lack of ap preciation of the system. It may also be seen as a distancing from the BSU and what may have been perceived as one of their responsibilities which did not result in a usable outcome. This is consistent with the previously discussed attitudes evident towards specialised units. The M ilat i nvestigation, a s pr eviously m entioned, pr edated t he i ntroduction of formal c rime l inkage s ystems a nd t herefore r elied on i nformal m easures of l inking related crimes. In t his case, it w as f ound t hat investigators ha d little di fficulty in identifying the r emains f ound in the B elanglo S tate F orest. According to Bellamy (2004) police be lieved that t he identity of t he vi ctims m ay ha ve b een know n b y colleagues i n S ydney’s King’s C ross a nd f rom t he M issing P ersons B ureau, not l ong after t he r emains of C larke and Walters w ere identified not l ong a fter t hey w ere discovered. It i s s uggested t hat t he pol ice ha d k nowledge o f t he i nvestigations being undertaken by t heir cou nterparts i n S ydney and cont acted them i mmediately. W hen asked about the quick identification of the victims, Superintendent Small stated that: …also the f act that th ere were a lready r umours f loating pa rticularly i n t he media but a lso a mong pol ice, about ot her d isappearances that c ould h ave occurred in the area. Expanding on t his i n f urther de tail l ater a nd hi ghlighting t he p ragmatic pr oblems i n substantiating this local knowledge, Small states that: - 281- What you have w as w hen the pe ople w ent missing t hey w ere r eported at a variety of d ifferent l ocations. Some w ere r eported at K ing’s C ross police station, some were reported to Missing Persons, some were reported- Gibson and Everest ... were reported to NSW police. Anja Habschied and Schmidtl were reported through the German authorities to the Australian Federal Authorities so the Australian Federal Police had reports of them. And w hat w e e nded up with a t t he e nd of the d ay w as f inding t hat- the backpacker- there w ere ab out sev en different locations t hat ha d reports of these m issing ba ckpackers t hat h ad their own recording sy stem but they didn’t l ink, weren’t ne cessarily com patible an d we ha d to draw t hem together. So t hat m eant that w e just cou ldn’t g o to one p lace a nd make a search and come up with a list. So that was one thing we had to be careful of and that came out in our inquiry. But in terms of the quick identification, was that s ome of t he people had c lothing t hat w as r eadily i dentifiable an d well described when they w ere m issing pe rsons a nd t hat facilitated their identification. This s tatement hi ghlights a p roblem tha t featured in other c ases, in particular, the reporting to different police stations or organisations. There were several indicators that aided in the identification of these crimes as serial in nature and the responsibility of the same offender/s. The remains of the victims were all cove red with debris f rom t he forest s uch as br anches and leaves (Bellamy, 2004; Whittaker & Kennedy, 1998). Another feature that indicated to police that these - 282- crimes were related was the discovery of a ba rbecue made from house bricks at two of the l ocations (Bellamy, 2004; W hittaker & K ennedy, 1998) . A dditionally, Superintendent S mall e xplains t he pr ocess w hereby k ey commonalities w ere us ed t o seek connections between other disappearances: You ha ve t o set criteria first. One of the things that we did, after searching the forest and we found the rest of the bodies one of the things that was done was that- one of the things we did was to establish a set of parameters. And not long after the bodies had been found, that’s all of them had been found, we were able to establish a number of characteristics. For example they were backpackers, t hey ha d l eft f rom t he L iverpool a rea had be come a poi nt of interest, they w ere y oungish, and there w ere a num ber o f t hose so rts o f characteristics a nd we compiled those pa rameters, we sen t then a l ist t o missing pe rsons- said, “Can y ou pr ovide a l ist of a ll t he p eople w ho disappeared ov er this p eriod of time w ho w ould fit t hese parameters?” So that was one source. We also checked reports of bodies that had been found that may not have been identified. Do any, as far as we know, do a ny of the characteristics fit these circumstances? Another c ommonly reported item w as the be lief held by inve stigators tha t M ilat ma y have b een i nvolved i n at l east t hree other murders (Kidd, 2000) . T his s uspicion ha s been upheld by a coronial inquest in which Milat was linked to the disappearances of three young women during the 1970s in the Newcastle region and at least another five murders (Stewart, 2001 ). In relation to the Newcastle cas es, a f ormer publ ican identified Milat as a guest at his hotel around the time that one of the young women, a - 283- patron t o t he hot el, ha d di sappeared (ABC, 2002) . As s tated by S mall, there w ere a number of cases where Milat had not been eliminated as a suspect (Small, 1998). The ‘ lack of ne tworking’ or l inkage bl indness c apability i dentified b y E gger (1990a) was apparent in the Wayne Trindall case where six incidents were reported to four different police stations. A factor that compounded this situation was the nine year timeframes within w hich T rindall c ommitted h is of fences. A r eview of r ecords indicates that a lthough several of t he of fences occurred dur ing t he t imeframe w hen ViCLAS w as i n us e, none of t hese of fences a ppear t o ha ve be en s ubmitted t o t he system f or in clusion or l ater c omparison. T he t imeframes over w hich these of fences occurred and the lack of viable linkage capabilities undoubtedly delayed recognition of these offences as committed by a serial offender. Although there can be no definitive explanation as to why such a step was not taken, several factors when considered together may provide a t entative answer. These factors are the vi ctim selection methods e mployed b y s ome s erial o ffenders, as identified by Egger (1990a) and Fox et al. (2005), coupled with the influence that this understanding ha s w ith c ommon di ctionary knowledge of pol ice and t he r esulting outcomes. A lthough t he Trindall crimes w ere l inked us ing D NA t echnology, N SW Police records indicate that some of these cases were not investigated until 2004. T his arguably reduced the ability of police to recognise that a serial offender was responsible for the sexual assaults, which did not occur until mid-2004. The records also indicate that there was little if any formal investigation of the offences when they were reported. - 284- One e xplanation f or t his i s c onsistent w ith t he a rgument p roposed b y Turvey (2002d) that t he cr imes m ay not ha ve be en viewed as w arranting a c omprehensive investigation. Police socialisation practices as outlined by Devery (2003) identified the existence of a “…s pecialized language of ope rational pol ice” (quoted, p. 118) that i s used t o c ategorise pe ople t hey w ould e ncounter i n t heir pr ofession, a s discussed i n Chapter 4. T his dictionary know ledge a s de scribed b y C han (2003a) has s ome influence on the actions that are taken in response to a situation. In the Trindall case, police were usually aware that many of the victims worked as prostitutes, and t hat violence i s not a r are con sequence o f t he job. This s pecialised knowledge m ay have l ed t o t he m inimum r ecording and gathering of evidence t o b e conducted, thus allowing the offender to go undetected for longer. From the inf ormation available on the D aley case, it is unc lear as t o w hen or how this series of crimes were linked. All media articles from the inception of the task force unt il jus t pr ior to the a rrest of D aley reported on c ases t hat oc curred i n t he suburbs of Parramatta and Westmead. It was only after Daley’s first court appearance for these crimes that there w as an y mention of t he sexual assault cas es i n Leichhardt. The s tatements ma de to the me dia b y the ta sk force’s Inspector M cKay ha d not previously m ade r eference t o t he Leichhardt of fences and du e t o t he una vailability of records, it i s unc lear how t his l inkage w as m ade. T here i s s ome i ndication f rom t he records t hat t he c rimes were l argely l inked b y the m odus ope randi, w hereby D aley would a ttack hi s vi ctims w hen t hey w ere i n a c ar pa rk, w ith one exception, a nd ha d - 285- produced a weapon each time. Demands made by the offender for money and oral sex were also features of his modus operandi, perhaps aiding in the linkage of these crimes. Similarly, the crimes committed by Johnson appear to have been linked due to the similarities in his modus operandi. The case review states that Task Force Ardmore was set up in response to a series of sexual assaults without describing how these cases had been linked. A newspaper article however, it describes how, “police have identified a num ber of s imilarities w ith each crime and ar e co nvinced t he attacks, ranging from sexual as saults t o rape, have b een committed by t he s ame m an” (Keogh, 1997) . T his supports the idea that these crimes were linked by their modus operandi. The issue of linkage blindness also surfaced during the Snowtown investigation. This i nvestigation be gan as s eparate m issing person enquiries. As detailed b y P udney (2005), t he na me o f one of t he of fenders, J ames B unting, w as connected t o t hese individual cas es at an early stage o f t heir di sappearances. Due t o the i nability of th e systems t o a dequately cross c heck t his i nformation adequately, i t w as not unt il a n officer working on one of the missing persons cases made a connection by chance that these incidents were linked. The Paper Trail As i dentified b y S heptycki (2004) there i s a di slike of the pa perwork involved i n intelligence-led approaches. Therefore, the avoidance of paperwork may also be argued - 286- to be a reflection of the way in which dictionary knowledge of police can pervade the use of intelligence-led policing tools. This a rgument t hat t he avoidance of p aperwork c an be i nfluenced b y police dictionary kno wledge would a lso pr ovide a n e xplanation a s t o w hy none of t he incidents s tudied t hat occurred du ring t he p eriod w here the linkag e s ystem w as available w ere s ubmitted to ViCLAS f or inc lusion. The V iCLAS f orms a re a ctually small books a nd r equest a n e normous amount of de tail r elating t o many aspects of a case. There is also an abridged version of the ViCLAS form, referred to as the ‘short’ form, which is still several pages long. One detective gave the opinion that this tedious form-filling was one of the reasons they believed ViCLAS was not readily incorporated into NSW Police. There w as, however, l ittle e vidence t o s upport t he i dea t hat t he p ractice of inaction in police work to avoid paperwork was widespread. However there were some instances, s uch as t he J ohnson i nvestigation, w here evidence ha d not b een pr operly recorded and maintained which would appear to be reflective of the practices described, consistent with minimising the paperwork completed. One r eason f or lack of e xamples e vident of t his pr actice of avoiding t hese accountability measures m ay be t he t ype of i nvestigation t hat i s t he focus of t his research. Serial crime cases were not considered routine police work, but part of large investigations. The exception here of t he J ohnson case can b e attributed to t he varied - 287- nature of the offences, whereby on a number of occasions the offender committed break and enter offences without any physical attack on the victims. Therefore, the suggested negligence m ay h ave be en an indoctrinated response t o dealing w ith t hose t ypes o f offences, which can be numerous in some areas. In investigations of s erial c rime, there is f requently hi gh level of m edia coverage. This attention adds to the pre-existing levels of accountability. As previously discussed, S uperintendent S mall ha d c ommented t hat t here w ere pr essures on t he investigation from the me dia, politics, internal pol itics, but stemming f rom the pressures f rom t he c ommunity t o e nsure t hat t he i nvestigation w as carried out t o t he highest s tandard. T his h igher l evel o f a ccountability t o m ore t han j ust t he r eporting requirements may have contributed to the levels of compliance seen. Additionally, the construction of a task force requires that there are individuals responsible f or s pecific duties. In t he t ask force situation, e ach pe rson w ould ne ed t o ensure that their contribution is being made, otherwise the investigative process cannot come together and the outcome of various leads will not be seen in the context of the investigation as a whole. Criminal Profiling One of the investigative techniques that was recognised in the literature review as a tool frequently as sociated with the inve stigation of s erial c rimes w as c riminal pr ofiling. - 288- Criminal pr ofiling, t he deduction of of fender b ehaviour a nd characteristics ba sed on information such as the crime scene, was used in several of the c ases o f interest. The ways in which this tool was perceived by investigators and influenced the actions of the investigation differed. From the interviews and the case study materials, it was evident that the majority of cases of interest used or attempted to use a criminal profile at some stage during the investigation. Criminal pr ofiling was c onducted t o s ome d egree b y t he pol ice a nd by professionals in the field during the Milat investigation. As the closed case file was not available for this case, the original criminal profiles provided to investigators were not available. An account by Bellamy (2004) however, details what is purported to be the criminal pr ofile pr ovided b y an e xperienced c riminal pr ofiler, D r R od M ilton, a forensic psychologist who was bought in after the discovery of Clarke and Walters. Dr Milton was a member of what was referred to as the ‘Think Tank’. Superintendent S mall a lso di scussed hi s p erception of t he c riminal p rofiling pr ocess and the way in which he used the profile during the Milat investigation. His opinion on how a c riminal pr ofile s hould be ut ilised i n the c ontext of a n i nvestigation w as expressed during the interview: One t hing t hat p rofiling d id t hen and d oes now i s it g ives y ou a b it o f a picture of t he offender. Y ou ha ve t o be c areful a nd r emember t hat is never always 100 percent, but it can also give you tips of things to look for if you ever i dentify a suspect. P rofiling, t here m ight be o ne or t wo exceptions - 289- around the world, but profiling doesn’t actually identify the offender. What it will do is give you a bit of a picture of what the offender might look like. In t he case of t he ba ckpacker i nvestigation, the pro filing w as pa rticularly useful in a n umber of w ays. It pr ovided a picture which qu ite f itted I van Milat when he was subsequently identified and indeed the Milat family. But I think in a sense more importantly the profiler assisted to the extent that Rod Milton was v ery conv inced that the offender w ould have k ept i tems belonging to the victims. That the offender would have some of those items, probably c lose by hi m a nd on display i n t he house because t here w ould be this f eeling of pow er w here y ou might g o and visit I van Milat- you m ight visit the offender- and the offender knows that property belonging to people he has murdered is on show. You c an see it. You may e ven be handling i t. And it gives them a sense of satisfaction and feeling of power and the – that extent those sorts of views when they are strongly expressed and reasonably argued they do influence your decisions on: Will we go and search the home? When will we execute a warrant? When will we make an arrest? Under what circumstances would we make it? It is e vident tha t in this c ase the c riminal pr ofile g uided aspects of t he i nvestigation decision m aking. T his k nowledge w as us ed a s a guide t o unde rstanding the of fender. The c riminal pr ofile c an also be s een as a too l to aid the impl ementation of ot her investigative t echniques, and legal a ctions s uch as a s earch warrant. Superintendent Small described the way in which the criminal profile informed the investigation during the Milat investigation: - 290- And sometimes t hey i dentified issues like m otive a nd be haviour. S o t hat I think was very, very important. What you have now also, though, are things like, m ore t han y ou d id then, a re things l ike w here plenty of s tudies ha ve shown that o ffenders commit cr imes i n those a reas w here they a re comfortable w hich means t hey’ve proba bly be en in the ar ea an d they f eel safe and secure. This was again the case with Milat…. It not only came from Rod M ilton bu t i t t ended t o r einforce y our ow n v iews a nd y our ow n assessments which is always comforting to you. So very quickly we had a pr ofile built up where the offender knew the Bowral area and the forest are very well, was well experienced with firearms, had probably lived in the area. Also knew the Liverpool area quite well. Rod Milton a dded t hose dimensions of probably be ing a c ontrol f reak, a v ery controlled person, blue collar worker, a rough age range and things like that. Now a lot of that does not identify individuals, what it does do is put a lot of people into the parameters, people outside it you can’t necessarily rule out…. It allows you to prioritise. The readiness to consider the criminal profile in the investigations could be seen to reflect the t raditional ‘gut i nstinct’ approach t o pol icing. Being able t o i ncorporate experience into the understanding of the crimes, even if not one’s own, may be seen to familiarise detectives with the cases that are rare and therefore unfamiliar. - 291- In a ddition t o t his i mplementation of t he pr ofile, S uperintendent S mall a lso commented on the form of the profile. As discussed in Chapter 3, the criminal profiling process usually results in an actual profile produced, which in some cases may not be amendable. …once a profiler puts it into black and white, you’re going to be locked to it. It focuses your mind and you don’t ask all the other questions. My view was: yes, I wanted Rod to do h is profile; yes I wanted him to put some things in writing; but I didn’t want to be locked to that. The role of the criminal profiler in the Milat investigation emerges as one in support of investigators. Using the Think Tank as a sounding board during the investigation also supports this idea. Investigators drew the conclusion that as the bodies were found in an isolated area, the offender may be local (Bellamy, 2004). Dr Milton agreed with police that the killer was familiar with the area, supporting this supposition. This conclusion may ha ve he lped guide their i nvestigation t owards pos sible s uspects w ho l ived in or were familiar with the area. Small discusses an instance where this relationship between criminal profiler and investigator was at play: For exa mple w e’d all co me t o the v iew t hat he had- the of fender or offenders- had gone to school in somewhere between Belanglo and Liverpool or Belanglo a nd the northern side of Liverpool. That sort of t hing. And we were also of the view that this person would probably be known to teachers, that would have been a problem child, would have come from a large family- - 292- that’s what Rod Milton had put a large family in- and would have had these characteristics. And we a ctually t oyed with the idea of writing to a ll the schools. Although the criminal profile developed by Dr Milton appeared to contribute positively to the de velopment of the inve stigative s trategies e mployed during the M ilat investigation, hi s pr ior i nvolvement i n t he G lover c ase c ould be considered a s l ess successful. The criminal profile developed by Dr Milton was present in the closed case file, and w as c onstructed after t he m urders of t wo of t he w omen. In t his profile, i t i s suggested that the c rimes are consistent with an offender who is in their late teens to early tw enties. He substantiated this with facts from t he case, s uch as t he t ime of t he offences after 3pm, and the proximity of the local high school to the murders that were of consideration at the time this profile was being developed. Also used to substantiate the profile of a young offender was a witness statement from another incident, whereby an elderly man had been attacked late one night. The theory at the time of investigators, as evident in the closed case files, was that the elderly man may have been mistaken for a woman due to the coat he was wearing. If pa rallels are dr awn between the us e of t he cr iminal pr ofile i n the M ilat and Glover i nvestigations, i t can be s een ho w a pr ofile l ater f ound t o be i naccurate c ould have i nfluenced an i nvestigation. S mall was quot ed saying such i nformation from t he criminal pr ofiler a ssured him t hat a l ine of i nvestigation w as w orth p ursuing. T he - 293- criminal pr ofile in the G lover c ase ma y h ave had a s imilar e ffect, bolstering the investigators’ belief t hat t he of fender w as a young m ale, e ven t hough t his w as only supported in the profile by a witness to a crime that may not have been related to the murders. Thus, this can be seen to support the argument that incorrect case theories can lead to misguided efforts in the investigation. Also of int erest in the Glover i nvestigation w as t hat t here ha d be en a ttempts made t o ha ve t he F ederal B ureau of Investigation (FBI) compile the ir own criminal profile. Information h ad be en c ollated a nd s ent t hrough an A ustralian embassy representative; however, from t he cl osed case f ile i t doe s not appe ar t hat t he cr iminal profile was produced by nor received from the FBI. There was not a written record of it in the f ile, nor w as t here an y r eference t o a r eceived profile i n any of t he doc uments relating to the inve stigative s trategies i n place. This f ile, how ever, w as not c omplete therefore the FBI profile may have been received but was not on record. The c riminal pr ofile de veloped b y D r M ilton i n r elation t o t he serial r apes committed by Johnson was markedly different. Here, there were two criminal profiles available. Both had been written after the identification of the offender as a suspect and it i s clear from t he profiles t hat Dr M ilton had been i nformed of t he offender and hi s general background. The initial profile offered the t ypology under which the offender of t hese cr imes w ould be cl assified. The s ubsequent c riminal pr ofile of fered s ome insight into the relationship between Johnson and his wife, drawn from letters obtained - 294- during t he execution of a s earch w arrant on J ohnson’s pr emises. T he i nvestigative impact of such insights are limited. During t he i nitial s tages of t he P resta i nvestigation after f ive cas es ha d be en identified as possibly being serial in nature, Detective Faucett describes her request for a c riminal profile f rom t he Behavioural S ciences U nit o f NSW P olice and t he information received: Detective- …Rang the Behavioural Sciences Unit and asked them to profile it all for us. They came up with …a location map that mapped where the set occurred and where they believed the offender lived. Interviewer- A geographic profile? Detective- Yes. They ha d him l iving so mewhere ov er t his P enrith way, Penrith mountains sort of way. It w as not cl ear f rom t he i nterview w hether the detective h ad expected a ge ographic profile or a criminal profile detailing possible characteristics of the offender. This was also not ava ilable f rom t he cl osed case f ile w hich was i ncomplete w hen accessed. To establish the type of profile they received I asked Detective Faucett: Interviewer- Did t hey a ttempt t o g ive y ou a n a ctual pr ofile- a pe rsonality profile of the person? - 295- Detective- Very, v ery general. B ut i t w asn’t ba sed on a nything ne w. W e knew he w as a s ecurity gua rd. W e k new t hat he ha d pr obably be en i n t he security i ndustry because he had a jacket on that l ooked like a uni form- he had on a black uniform. It worked out he was a l oss pr evention officer. He wasn’t a security guard as in a patrolling security guard. But he was wearing the uniform he ha d a n obsession w ith un iforms a nd police a nd t hat sort of stuff. He also- We new his age range from the boys descriptions, we knew he was not Caucasian descendant because the descriptions the kids had given us pretty much outlined what our profile was going to be so there wasn’t a lot more to add to it but we weren’t given anything else anyway. Although i t a ppears that s ome inf ormation consistent w ith a c riminal pr ofile w as received, it is clear that it did not contribute to the investigation. Following up on t he comment pr eviously m ade a bout t he g eographic pr ofile t o a scertain t he us efulness of such information to the investigation, Detective Faucett was asked: Interviewer- The B ehavioural S ciences U nit, they g ave y ou a g eographic profile that identified the Penrith/mountains area. Where was Presta actually living? Detective- Liverpool. I assumed he would have been from the Liverpool area as t hat’s where t he f irst offence oc curred so I t hought t hat’s where he ’d be familiar. Turns ou t h e w orked i n t he m ountains. H e w orked a s a security guard. He worked at several different locations all over the place. He worked at M acquarie F ields, he w orked a t P icton bu t i t w as on hi s journeys ba ck - 296- from Katoomba to Wattle Bay that he made his pick ups of t hese boys. There’s also an offence that occurred when he was up near Newcastle but it was never reported but it seemed to follow his work location. Finally, dur ing t he Folbigg i nvestigation, t he op portunity w as t aken t o h ave a Scientific C ontent A nalysis, a s pr eviously de scribed, conducted on t he d iaries of t he offender. It w as n ot c onsidered us eful i n t his c ase how ever, a ccording to Detective Ryan, t his pr ocess w as a imed a t g iving t he i nvestigator a pe rspective a bout t he personality o f t he suspect that c ould be i ncorporated i nto t he i nterview t echnique. Although unsuccessful in this case, Detective Ryan was more optimistic about its use in the future. He commented that: … i t w as w orth t rying a nd a t t he e nd of t he da y you t ry a nything a s a n investigative technique to gain all the information that you can possibly get and it enables you to make the right decisions. I’d do it again even though I didn’t get a lot out of it. This sums up the overall perception about the use of criminal profiling, in any form, in an investigation; i t is a me ans to obtain as much i nformation a s pos sible t o a id other aspects of the investigation. - 297- Use of Typologies As i dentified pr eviously, t here ha s be en a s ignificant a mount of r esearch i nto categorising offenders according t o va rying sets o f t ypologies. T hese s erial m urderer and serial rapist typologies were proposed to classify offenders in the anticipation that this m ay aid t he i nvestigation of t hese c rimes. F rom bot h t he i nterviews and t he case studies, an attempt was made to see if the use and understanding of typologies had an important c ontribution t o t he i nvestigation, a s suggested i n t he pr evious r esearch outlined in Chapter 2. As abovementioned, the criminal profile produced in the Johnson case identified the offender according to the category of rapist he best fitted. Johnson was identified as a pow er-reassurance r apist (Groth, B urgess, & H olmstrom, 1977a ), a cat egory discussed i n C hapter 2, and the c riminal pr ofiler w ent on to describe some of th e activities in w hich this t ype of o ffender m ay be i nvolved. T here was no s pecific information pr esented i n t he pr ofile a bout ho w t his know ledge c ould a ssist t he investigation. What was interesting to note was that this profile had been developed after the offender had been identified through fingerprint technology. A search warrant had been executed b y T ask Force A rdmore a t t he o ffender’s pr operty, dur ing w hich i tems belonging to some of h is vi ctims w ere ide ntified and collected. From the c riminal profile provided, it was evident that information about the suspect had been provided to - 298- the pr ofiler. This pr actice is not in line w ith some tr aditional be liefs a bout c riminal profiling, w hereby s uch information m ay be c onsidered t o bi as t he d evelopment of a criminal pr ofile. T he criminal pr ofile is us ually c onsidered as a tool to ide ntify a n offender (Turvey, 2002a ); t herefore t he us efulness of t he t echnique t o t he progress of an investigation in this case was questionable. The ot her case in w hich t he us e of t ypologies was evident w as i n the investigation into Slater conducted by Detective Pryde. Although a criminal profile had not been sought, he did receive advice from the then Crime Faculty that informed him that the crime he was investigating appeared to have been committed by a sexual sadist, a t ypology as de scribed i n the l iterature r eview. A l ist of t ypical be haviours accompanied t his a dvice, although it di d not c ontribute to the a ctual inv estigation. It should be noted that the Crime Faculty had not reviewed the full details of the case and therefore would not have been able to provide a great deal of case-specific information. Modus Operandi & Signature The us e o f t he m odus ope randi i n t he i nvestigation of s erial c rimes w as consistent t hroughout s erial m urder a nd s erial r ape i nvestigations. T his i nformation could be s een t o r einforce de cisions b y investigators a nd t o i ndicate l ines of i nquiry, issues w hich emerged f rom bot h t he i nterviews a nd case s tudies. In di scussing t he investigation into a serial rapist, Detective Faucett describes how they identified related crimes. She states that the “MO was good” (Detective Faucett) when comparing various - 299- reported incidents. Similarly, the MO was used to demonstrate an understanding of the offender’s s uspected criminal behaviour i n previous years as t he d etective s tated that, “That was his MO at the time so that way my assumption” (Detective Faucett). Although t his example illustrates t he us e of t he modus operandi t o clarify and support the investigative direction, this is not always the case. The modus operandi in the Milat case varied and is still subject to speculation. The methods used to commit the murders i ncluded repetitive s tabbing, gun s hots, and in the c ase of H abscheid, decapitation (Bellamy, 2004; W hittaker & K ennedy, 1998) . A ll vi ctims ha d be en murdered w here t heir b ody was f ound with t he e xception of N eugenbauer. E vidence suggests t hat N eugenbauer ha d b een m oved from t he s ite w here he was m urdered (Bellamy, 2004) . S uch v ariance i n m odus ope randi m ay h ave a ttributed t o t he be lief that Milat did not commit these murders alone and may have been part of a team. A ke y feature or signature in this case which was evident from the remains of Milat’s vi ctims, w as that t he pa nts or j eans of vi ctims w ere f ound t o be buttoned but had the zip undone (Bellamy, 2004; Whittaker & Kennedy, 1998). In addition to this, it has been noted by many sources that near each body disposal location, a brick or stone fireplace was present, further linking these crimes. The us e of a s ignature to j ustify t he l inkage of a s eries of cr imes w as al so evident i n ot her c ases s uch a s Glover. In t he G lover c ase, the vi ctim’s s tockings tie d around t heir neck was the r ecognised s ignature. This, however, had the a bility to - 300- mislead investigators, as this feature was not present in all the cases and did not occur in t he a dditional sexual assaults c ommitted by Glover. Although the s everity o f the murders a nd t he vi ctimology w as s ufficient t o l ink t hese c rimes w ithout a c onstant signature, i t c ould be s uggested t hat t his m ay h ave been one of t he r easons wh y the series of as saults and sexual as saults on elderly women were not a ssociated with the murders until towards the end of the investigation. In discussing the use of modus operandi recognition in the case investigated by Detective P ryde, he identifies the i mportance o f officially r ecognising t his f eature for its possible future use: In particular this victim from her statement and other police evidence we had including t he of fence w e c all garrotting. N ow I w anted t o c harge hi m with the garrotting because his MO was that he would choke and strangle and if you can get that in it shows i t m akes a so rt of link between that particular offence- the strangulation component of the sexual offence to other offences he c ommitted no t on ly in this state bu t I w as thi nking a bout the b igger picture an d that w as t his p erson w as i nvolved in ot her sexual assaults t hat were y et unsolved or unf inalised i n ot her states w here h e’d done t he sam e thing. From t his example a nd t he ot hers di scussed, i t is a pparent t hat t he us e of t he m odus operandi was not as necessary to aid in determining whether a number of crimes were considered as s erial i n nature. Rather, t he m odus ope randi contributed m ore t o t he - 301- linking of s uch c rimes. In the case of D aley, J ustice S impson de scribed bot h t he signature and modus operandi: …[the] modus operandi was to lie in wait for women at night when they left work or as they walked home after getting off public transport. Armed with a knife or screwdriver, he would t hen f orce them t o perform or al sex on h im and would ask them questions about their sexual history before robbing them (Williamson, 2002). In t his i nvestigation, t he m odus ope randi w as pu blicised t hrough t he m edia (Cowper, 2000; Toy, 2000). Therefore it can be seen to be used both investigatively, to identify unconnected crimes (Toy, 2000) , a nd a s a c rime pr evention t ool, i dentifying a nd disseminating information relating to the risk faced by the public (Cowper, 2000). Chapter Summary There ha s b een a m ixed reaction to the us e o f i ntelligent-led pol icing pr actices, m ost notable of which was the formal crime linkage system once in place, ViCLAS. Whilst it has be en a cknowledged b y pol ice t hat t here ha ve be en pr oblems i n l inking r elated crimes, s upport f or s ystems s uch a s V iCLAS was not e vident. T he us e of c riminal profiling te chniques in the inve stigation of s erial c rimes w as a lso subject to mixed responses b y pol ice. T here w as i nsufficient e vidence t o s uggest t hat s uch t echniques had a significant impact on the way in which serial crimes are investigated. There was - 302- however, a c lear w illingness a mongst pol ice t o c onsider t his t ool i n f uture investigations. - 303- CHAPTER 9: A NEW TYPOLOGY- VIOLENT SERIAL CRIME INVESTIGATIONS This r esearch set out to de termine w hy s erial c rime inc idents r emain undetected for prolonged periods of time, and what factors impact on the investigation of these crimes. From the research a number of factors were identified as having an impact on t he way in w hich s erial crime i nvestigations ha ve pr ogressed a nd l ed t o a c onviction. T hese factors contribute to our understanding of the processes of investigation that are utilised in the investigation of serial murder and serial rape. In order to enhance our understanding of serial crimes, there have been a range of typologies proposed, as identified in the research literature. Such typologies exist for both serial rape (Groth, Burgess, & Holmstrom, 1977b; Knight, 1999; Knight, Warren, & Reboussin, 1998) and male and female serial murderers (Holmes & DeBurger, 1985, 1998; Kelleher & Kelleher, 1998). One of the main critiques of these typologies is that they do not c ontribute t o our understanding of t he w ays in w hich s erial m urder a nd serial r ape a re i nvestigated. These pr evious t ypologies a re a lmost e xclusively representations of the motivation of the offender. As ha s em erged from t he r esearch, there a re num erous i ssues t hat were identified as impacting on the ability of police to recognise cases of serial murder and serial r ape. Once t his r ecognition was m ade, police w ere abl e t o achieve a pos itive outcome a nd s top t he s eries of of fending, t hus pr eventing f urther vi ctimisation. - 304- Therefore, it is e ssential f or a t ypology to assist the inve stigation through t he recognition of limitations to the investigative process. From t he f indings t hat ha ve e merged f rom t he da ta, i t i s pr oposed t hat a typology of the investigation exists, which describes the investigation as a whole. This typology is developed based on the trends that have emerged from the findings. As with most t ypologies, t hese categories are not m utually e xclusive a nd t here c an be s ome overlap. These c ategories largely r eflect the recognition of the crimes as serial, which can be found to influence the length of the investigation, the span of criminal offending and the time to achieving a positive investigative outcome in the form of the arrest of the offender. There are three categories in this typology and they have been termed as: the S erendipitous Investigation; t he C onspicuous Investigation; a nd t he D elayedRecognition Investigation. The Serendipitous Investigation The f irst c ategory of t his t ypology ha s be en t ermed t he S erendipitous Investigation. These cases are often characterised by: • • • A primary link between cases from sources other than police A linkage that occurs after at least four cases or six months An arrest usually achieved within six months of recognition of crimes as serial by investigators - 305- The underpinning of this category is the way in which these series of crimes are linked. R ather t han be ing l inked t hrough f ormalised l inkage s ystems dur ing a n active investigation, t hese s erial c rimes a re l inked t hrough ot her m ethods or s ources. C ases from t he pr esent r esearch s ample t hat s upport t his i dentification of t he Serendipitous Investigation include: • Bruno Presta o The s exual a ssaults ha d oc curred ov er an e ight m onth pe riod pr ior t o linkage. o Similarities be tween two cases a nd earlier inc idents w ere f irst recognised by a family member of the detective. o Additional c rimes w ere i dentified t hrough the de tective’s ow n knowledge about other cases. o Presta was arrested within weeks of the linkage recognition. • Wayne Anthony Trindall o The sexual assaults occurred over a nine year period. o Recognition of related cases was made by the Diagnostic and Analytical Laboratory (DAL). This was not as a result of an active investigation. o Trindall was identified and arrested three months after the investigation into this series of crimes began. - 306- Although not completely consistent with this category, in the case of Ivan Milat, the me dia h ad drawn links be tween the f irst k nown victims a nd the o ther mis sing persons. H owever, t his w as a t a t ime w hen pol ice w ere r eporting t hat t hey w ere not expecting an y further vi ctims; t herefore t hese cr imes w ere not i nvestigated as a s erial offence at this stage. The Conspicuous Investigation The s econd c ategory, w hich i s r eferred t o a s t he C onspicuous Investigation, r efers t o cases in which the link between crimes is apparent. The characteristics include: • • • A number of victims that emerge over a relatively short period of time. Clear signature or modus operandi evident in each of the crimes. Intensive investigation, usually in the form of a task force, initiated immediately after linkage between crimes made. This category represents those cases that are usually best conform to the ‘ideal’ type of serial offender, who targets vulnerable victims, is considered to have been motivated by a need to fulfil a s exual desire (Goldsworthy, 2002; Mouzos, 2000) or is motivated by the need to achieve power and control over their victim (Egger, 1998). Such crimes are usually t he focus of i ntense m edia attention. T he investigative te am ta sked with working t his c ase is usually w ell r esourced, e ven w hen i nvestigations be come - 307- protracted. Cases from the present research sample that support the identification of the Conspicuous Investigation include: • Paul Charles Denyer o Three w omen w ere m urdered a nd a nother vi olently a ttacked ove r a 20 day period. o There w ere clear s imilarities be tween the cases w ith a s trong modus operandi apparent. o Strike F orce P ulsar i nitiated soon after s econd offence, leading t o Denyer’s arrest 50 days after his first murder. • John Wayne Glover o Five w omen w ere m urdered ov er a s even m onth pe riod, w ith a nother murdered just prior to Glover’s arrest. All women were murdered within the same geographical area. o Recognisable signature apparent in all murders. o Although an int ensive i nvestigation was ini tiated, it w as a lmost s even months before a dedicated task force was formed. This led to the arrest of Glover two months later. - 308- • Ivan Milat o Although murdered over a period of two years, five of the seven victims were di scovered ove r a period of a f ew w eeks a nd l inked t o t he pr ior discovery of the other two victims. All victims were located in the same geographical area of the state forest in which they were found. o Although the modus operandi varied between some of the victims, there was a clear signature linking the crimes. o Task F orce Air w as s et up i mmediately following t he di scovery of t he second set of remains and a link between the earlier discovery of victims was ma de. Milat w as arrested eight mont hs a fter thi s task force w as formed. This latter case demonstrates the importance of linking related crimes, as this was not achieved until after all victims’ remains were discovered. This is unlike the other two cases in this category, where the link between related crimes occurred within months of the m urders t aking pl ace. However, although l inkage o f m issing person cases did not occur, the crimes w ere linked once t hey w ere recognised as vi ctims of cr ime, and therefore are represented by this category. - 309- The Delayed-Recognition Investigation The t hird a nd f inal c ategory pr esented i n t his t ypology i s t ermed t he D elayedRecognition Investigation. T his c ategory refers t o i ncidents of s erial m urder or s erial rape where there was a long period of offending prior to the recognition of the offences as serial in nature. This category is characterised by: • • • Delay in recognition of related crimes exceeding 12 months. Long history of criminal offending in the series. Different j urisdictions i nvolved i n t he i nvestigation o f t he v arious crimes. • Inadequate consolidation of resources. As not ed, t he di fficulties a cknowledged b y t he research l iterature i n r elation t o t he consolidation of r esources a cross different jur isdictions (Egger, 1990a ) have be en identified as a key issue and characterise many of the cases that fall into this category. From the present research sample, the cases of serial murder and serial rape that support the identification of the Delayed-Recognition Investigation include: • Anthony James Daley o A period over 18 months before the crimes were investigated as serial in nature. Not a ll c rimes w ere ini tially linke d to investigated by the Strike Force Sapelo. - 310- the of fender or o Timeframe of known criminal offending was 22 months. o Offences oc curred i n t wo r egions ove r 20 ki lometres a part a nd unde r different local area commands. o Initially t he crimes i n the t wo areas were i nvestigated as un related and investigated by the respective commands. • David Glen Johnson o Formal r ecognition of t he crimes as s erial i n nature di d not occur unt il almost t wo years after t he i nitial of fences. The task f orce w as s et up following problems in the previous investigation by detectives. o The know n t imeframe o f of fending f or t his s eries of c rimes w as t wo years. o Due to the de ficiencies in the or iginal inve stigation, specialised detectives w ere r equired to form the t ask f orce, leading t o a reinvestigation of all evidence and witness testimony. • Snowtown o Initial investigation into the disappearances of three people did not occur until at least six months after the last person was reported missing. o Crimes oc curred over a f ive year p eriod, w ith s ome vi ctims be ing reported missing to police at v arious times. The remains of all but on e victim were discovered up to three years after death. - 311- o Some of t he vi ctims w ere r eported missing an d therefore ha d been investigated b y a de dicated unit, while ot her detectives w ere l ater required when the scope of the crimes emerged. • Kathleen Folbigg o Although each de ath ha d be en r eferred t o pol ice a utomatically, due t o the assignment of each of the deaths as a result of Sudden Infant Death Syndrome, these crimes were not recognised as such for ten years. o The of fences of m urder of t hree c hildren and t he m anslaughter of a fourth occurred over a ten year timeframe. o Folbigg had lived in two residences and attended at least four different hospitals with her children. o Once the inve stigating detective r ecognised the pos sibility tha t the deaths may have been serial in nature, resources from NSW Police were relatively forthcoming, although not dedicated to that case alone. As previously acknowledged, the categories in this typology are not exclusive. Due to the timeframe of offending and other features of the case, the Trindall case can also be seen to fit into this category. • Trindall o Crime l inkage di d not occur t ill up t o ni ne years a fter t he of fences occurred. - 312- o Known criminal offending occurred over a nine year period o Although the ma jority of c rimes oc curred w ithin the s ame a rea o f Sydney, t he c rimes w ere r eported t o a num ber of di fferent pol ice stations. Similarly, a spects of th e M ilat inve stigation can be s een to meet the c riteria f or thi s category as well. • Ivan Milat o The know n t imeframe o f of fending i n t his s eries oc curred ov er a t hree year period, with recognition of the crimes as related coming only after these crimes occurred. o Victims were r eported missing t o di fferent pol ice a gencies i n Australia and overseas. As no t wo i ncidents of s erial r ape or s erial m urder a re i dentical, t here c an be no one category within which these c rimes c an be pl aced. Of t he cases f rom t he r esearch sample, the Milat case is the only one which has features which suggest inclusion in all three categories. This t ypology emphasises the relationship between the ability of police to link related crimes and thus to investigate them as such, and the progression to achieving a good i nvestigative out come. T his s upports s uggestions by E gger (1990a, 1998) that - 313- linkage blindness c an ha ve a de trimental i mpact on t he i nvestigation of s erial m urder and serial rape cases. Also s ignificant t o t his i ssue of l inkage bl indness is the v ictimology of the se crimes a nd t he w ay i n which t his c an be s een t o c ontribute t o t he pr ogression of t he investigation. In t he S erendipitous Investigation category, t he c ases w hich ha ve be en identified all inc lude vi ctims tha t a re s ubject t o strong s tereotyping b y police: A s discussed i n t he r esearch l iterature, pr ostitutes (Bridgett & R obinson, 1999) , male victims of s exual a ssault (McAdam & F itts, 19 99) and m issing pe rsons (Swanton & Wilson, 1989). This supports the research literature which proposed that detectives will use culturally informed judgements based on stereotypes, which are all subject to police cultural understandings about these groups. This dictionary knowledge can be seen to assist pol ice to make ju dgements a bout v ictims, a nd s ubsequently i nform t he w ay i n which they investigate crimes committed against these groups. This us e of know ledge also pr ovides one e xplanation f or w hy t he m ajority of the vi ctims represented i n t he C onspicuous Investigation c ategory were recognised as being part of a series relatively soon. The victims of the Denyer and Glover cases are ‘innocent’ victims- young women and the elderly- and thus achieve a different response due t o t he pol ice c ultural unde rstandings t hat t hese gr oups a re ‘ real’ victims (Egger, 1998; Fox, et al., 2005). Coupled with the fact that these victims were murdered over a comparatively s hort pe riod of t ime a nd di scovered s oon a fter de ath, t hese f actors contributed to the ability to recognise that the crimes were related. - 314- Similarly, the victimology of the crimes could be seen to have some influence in the D elayed-Recognition Investigation c ategory. T his i s pa rticularly evident i n t he Snowtown case where police described the victims as “a group that preyed upon itself” (Boston, 2004), thus demonstrating the knowledge that had informed the way that they investigated the disappearances of these victims. Conversely, the police understanding that accept ed t he explanation pr ovided f or t he de aths of F olbigg’s c hildren, a nd t he knowledge that the murder of children by their mother is rare, contributed to the delay in the investigation of these crimes. As ha s be en d emonstrated i n t his t ypology, t he i ssues of crime l inkage and victimology are not the sole defining issues contributing to the investigative outcome. Other issues such as the allocation of resources, the skills of the investigators and use of investigative tools and techniques all contribute to the ability of investigators to achieve the investigative outcome of an arrest, and are reflected in the timeframe within which this occurs. - 315- RECOMMENDATIONS This current research identifies some areas in the investigation of serial rape and serial murder that need to be addressed by police organisations. There are also further areas of research identified that were beyond the scope of the present research. Recommendations for the Investigation of Serial Crimes 1. Increased crime linkage capabilities required As presented i n t he di scussion, an analysis of t he data has revealed t hat t he ability o f police t o l ink r elated c rimes i mpacts on t he t imeframe o f of fending a nd w hen t he offender i s ar rested. The cr imes t hat w ere r ecognised as s erial i n nature ear lier were shown to increase the chances of an efficient and quickly obtained outcome in the form of arrest. 2. Reintroduction of formalised intelligence-led linkage systems The findings from this research suggest the need for formalised linkage systems that are in line w ith int elligence-led policing pr inciples. W hile it is a cknowledged that the system pr eviously op erational, V iCLAS, w as not i ntegrated i nto A ustralian pol ice organisations l ong t erm and i s no l onger op erational, t his r esearch ha s i dentified t he - 316- need f or m ore e ffective l inking of related c rimes. A s de monstrated b y t he t ypology outlined i n Chapter 9, t he l inkage of r elated c rimes t hrough t raditional i nvestigative practices often result in a delayed recognition, up to two years in this research sample, which c an l ead t o a l onger i nvestigation, gr eater a llocation of r esources, a nd an increased number of victims as the offender continues to offend until arrested. 3. National centralised body with investigative capabilities Due t o the r are na ture of violent s erial cr imes, f ew pol ice a gencies ha ve de tectives experienced in the investigation of these crimes. Very few of the detectives involved in the i nvestigation of t he cas es of f ocus i n this r esearch had any ex perience w ith serial crime inve stigation. Unlike ot her c ountries s uch as t he U nited States and t he F ederal Bureau of Investigation, Australia does not have a national policing organisation with the j urisdiction to investigate crimes s uch as m urder and r ape t hat oc cur acr oss l ocal policing bounda ries or when assistance i s r equested. Such centralised bodies can provide e xperienced i nvestigative t eams w ho ha ve t he s kills a nd knowledge t o investigate serial crimes. 4. Improved education of police officers to recognise groups vulnerable to crime A significant delay in the recognition of related crimes as serial in nature was found to be frequently associated with victims that are considered as vulnerable by the research literature. T his i ncludes pr ostitutes a nd t hose who l ive t ransient l ifestyles. T his - 317- education would assist in informing police cultural knowledge and therefore subsequent police practice. Recommendations for Further Research 1. Closer analysis of the missing persons register and its use investigatively With the i ntroduction of a na tional m issing pe rsons da tabase i n r ecent years i n Australia, it has been hoped that it will be a more efficient and accessible system than its predecessors. While this system theoretically fills some of the gaps left in the wake of ViCLAS, it still faces the s ame challenges of being i ntegrated i nto s tandard pol ice practice. T he n eed for a f ully i ntegrated national m issing pe rsons d atabase was supported b y t he V ictorian O ffice of P olice Integrity (2006), f ollowing t he decommissioning of the ViCLAS system. 2. New initiatives such as the NSW Police Unsolved Homicide Unit need to be reviewed The introduction in recent years of initiatives such as the Unsolved Homicide Unit need to be studied to determine if they have any impact on the investigation of serial crimes. As r ecognised by K ennedy (2007), m odern f orensic t echniques c ould i mprove t he likelihood of achieving closure on outstanding homicide cases. However, how this will impact the inve stigation of s erial crimes is - 318- yet to be de termined, thus f urther consideration is required. How these new initiatives address the problems experienced in linking violent serial crimes is yet to be seen. 3. An expanded study into the investigation of serial crime to validate the proposed typology Further research that analyses the investigation of other incidents of serial murder and serial r ape, both in Australia and internationally, i s ne eded to va lidate f urther t he proposed typology. This was beyond the scope of the present study. - 319- CONCLUSION This study asked how incidents of serial murder and serial rape in Australia remained undetected for up t o ni ne years. T o a nswer t his que stion, t he way i n which pol ice investigated t hese c rimes ha d t o be c onsidered. A n a nalysis of t he r esearch l iterature revealed that there was little prior research into how a r ange of investigative tools and techniques were utilised and how they contributed to the investigative outcome. While forensic techniques have their place in the establishment of links between related crimes, other aspects of the investigation were seen to have a greater impact on the i nvestigative out come. O ne aspect t hat had a s ignificant i nfluence on t he investigative outcome was the allocation of resources, whereby it was found that early identification of the c rimes a s s erial, coupled with the e stablishment of de dicated resources in the form of a task force, achieved an efficient resolution to the case more often. The efficiency of this combination of factors was demonstrated by the proposed typology in which the Serendipitous Investigation referred to cases of serial murder and serial r ape w here t he l ink between related cas es was m ade up to six months after the series b egan, which o ften l ed t o t he a rrest of a s uspect within s ix m onths of t his l ink being m ade. T his w as in c ontrast t o t he D elayed-Recognition Investigation w hich referred to cases where the linkage of related crimes did not occur until 12 m onths or more af ter t he i nitial of fence ha d occurred. However on ce t he cr imes ha d been - 320- recognised as serial in nature and resources allocated in the form of a t ask force, most of these incidents were not resolved quickly with some investigations exceeding twelve months until arrest of the offender was achieved. One factor that contributed to the inability of police to link related crimes was the lack of a formal crime linkage system. This intelligence-led approach, ViCLAS, had been i n pl ace du ring t he t ime pe riod when m ost of t hese o ffences o ccurred; how ever there was no evidence from the research data to suggest that any attempt had been made to use this tool in the investigation of serial murder or serial rape. This was despite the fact t hat this s ystem ha d be en de signed t o t arget s uch of fences a nd contribute t o a reduction in linkage blindness. Another key issue to the investigation of serial rape and serial murder offences was f ound t o b e t he m edia. D uring t hese hi ghly publ icised i nvestigations, t he m edia was f ound t o s erve i mportant f unctions f or pol ice. T hese f unctions w ere a s a di rect investigative tool to gain information from the publ ic a nd for a dditional vi ctims to come forward; to boost public confidence in the police investigation itself; and as a tool for cr isis m anagement to counteract a climate of f ear created by s ensational m edia reporting. The ove rarching f actor that ha s be en de monstrated t o i nfluence all of t hese aspects of t he i nvestigation w as t he w ay i n w hich pol ice cultural knowledge w as demonstrated t o i nform how pol ice i nvestigate s erial m urder a nd s erial rape. P olice - 321- cultural know ledge c an be s een t o i mpact on the pr ogression of t he i nvestigation, especially t hrough t he d ecisions m ade i n r egards t o t he c ultural und erstandings a bout some gr oups of vi ctims. O f pa rticular int erest to this r esearch was the w ay in w hich police c ultural unde rstandings of t he vi ctim i nfluenced ho w t hey i nvestigated vi olent serial crimes. Police cultural knowledge informed the investigation of serial murder and serial rape offences from the research sample. The relatively rare occurrence of serial crime causes p roblems i n t he de velopment of pol ice i nvestigative t echniques. P olice interviews de monstrate t hat t he i nvestigation of newly identified s erial crimes pos ed major issues for investigators. The cases that were characterised by extensive delays in the r ecognition of of fences as pa rt of a s eries of ten involved victims t hat w ere t he subject of strong police cultural understandings. It w as de monstrated t hat t his di ctionary know ledge i nformed de tectives a bout the way in which reported crimes or missing person reports were dealt with when the victims were known prostitutes, considered to be criminals themselves, or were deemed to be likely to disappear of their own accord. This subsequent directory knowledge led to the delay in the recognition that a serial offender was present. Thus, this knowledge directly impacted on and contributed to the delay in the recognition and investigation of serial mur der and serial rape c ases. It i s s uggested t hat i mproved e ducation of pol ice officers to increase their understanding of the victimology of violent serial crimes could assist in the recognition of these offences. - 322- An issue t hat w as evi dent i n the ana lysis o f t he da ta w as t hat of d etective experience. Due to the rarity of violent serial crimes, many detectives do not have the experience i n the i nvestigation of t hese crimes. It is s uggested that a na tional, centralised pol icing body could a ddress t his pr oblem, forming a uni ted s ource of experience and expertise. The f actors i dentified b y this r esearch ha ve i mpacted on t he i nvestigation of serial mur der and serial r ape inc idents in Australia. Whilst the impa ct on the investigation varied between cases, they contributed to the delay in the recognition of these offences as serial in nature and in achieving a positive investigative outcome. As indicated by the proposed recommendations, these factors can be addressed to improve the investigative capabilities of the police investigating these violent crimes. - 323- REFERENCES AAP. (2000, June 30). Serial rapist may be at work in Sydney's west. AAP Newsfeed. AAP. (2006, 12 December). Dupas to go straight to trial for graveyard murder charge. AAP. (2007, 24 December). DNA dispute sparks call for Falconio review. Brisbane Times. ABC. (2002). Publican links missing girl to Ivan Milat at inquest. Retrieved 2-12-2004, from http://www.net.au/news/newsitems/200202/s484675.htm ABC Premium News. (2006, 21 April). 'Sexual predator' jailed for 22 years. ABC Premium News. Abrams, D., Viki, G. T., Masser, B., & Bohner, G. (2003). Perceptions of Stranger and Acquaintance Rape: The Role of Benevolent and Hostile Sexism in Victim Blame and Rape Proclivity. Journal of Personality and Social Psychology, 84(1), 111-125. Adderly, R. W., & Musgrove, P. (2001). Police crime recording and investigation systems: A user's view. Policing: An International Journal of Police Strategies and Management, 24(1), 100-114. American Psychiatric Association. (2000). Diagnostic and Statistical Manual for Mental Disorders (Fourth ed, Text ed.). Washington, DC: American Psychiatric Association. Anderson, W. R. (1999). Can personality disorders be used as predictors of serial killers? Futurics, 23(3/4), 34-43. Anonymous. (1998). A Serial Killer's Perspective. In R. M. Holmes & S. T. Holmes (Eds.), Contemporary Perspectives on Serial Murder. London: Sage Publications Ltd. 324 Australian Bureau of Statistics. (1989). Year Book Australia: Chapter 12- Culture, Recreation, Environment and Tourism. Canberra: Australian Bureau of Statistics. Australian Bureau of Statistics. (2002). Year Book Australia (No. 1301.0). Canberra: Australian Bureau of Statistics. Australian Bureau of Statistics. (2007). Recorded Crime- Victims 2006, 2007 Australian Law Reform Commission. (1992). Multiculturalism and the Law (No. 57). Canberra: AGPS. Bartol, C. R., & Bartol, A. M. (2004). Introduction to Forensic Psychology. Thousand Oaks: Sage Publications. Bazeley, P., & Richards, L. (2000). The NVivo Qualitative Project Book. Thousand Oaks: Sage. BBC. (2003). Green River killer confession. Retrieved 06/11/03, from http://news.bbc.co.uk/1/hi/world/americas/3245301.stm Begg, H., & Norris, G. (2006). The City Stalker- Strike Force Grahame. Australian Police Journal(September), 188-192. Bell, C. (2002). The Murderer Hunters: The Detectives' Inside Stories. London: André Deutsch. Bellamy, P. (2004). Ivan Milat: The Last Ride. Retrieved 2/8/2004, from http://www.crimelibrary.com/serial_killers/predators/milat/discovery_1.html Benns, M. (2003). When the Bough Breaks: The True Story of Child Killer Kathleen Folbigg. Sydney: Bantam. Bernard, H. R. (2000). Social Research Methods: Qualitative and Quantitative Approaches. Thousand Oaks: Sage. Blackburn, R. (2000). The Psychology of Criminal Conduct: Theory, Research and Practice. New York: John Wiley & Sons Inc. 325 Boston, J. (2004). Snowtown: A Bank Vault's Deadly Math. Retrieved 10.01.05, 2005, from http://www.crimelibrary.com/serial_killers/weird/snowtown/index_1.html Bourdieu, P. (1990). The Logic of Practice (R. Nice, Trans.). Cambridge: Polity Press. Boyatzis, R. E. (1998). Transforming Qualitative Information: Thematic Analysis and Code Development. Thousand Oaks: Sage. Bridgett, M., & Robinson, J. (1999). Sex workers and sexual assault: the hidden crime, Restoration for Victims of Crime Conference. Melbourne: Australian Institute of Criminology and the Victims Assistance and Referral Service. Brookman, F. (1999). Accessing and analysing police murder files. In F. Brookman, L. Noaks & E. Wincup (Eds.), Qualitative Research in Criminology (pp. 46-61). Aldershot: Ashgate Publishing Ltd. Brown, M. (2002). The Monster of Belanglo. In Outside the Law: Australian True Crime Stories (pp. 81-112). Victoria: The Five Mile Press. Burns, R. B. (1997). Introduction to Research Methods (3 ed.). Melbourne: Longman. Cameron, B. W. (1999). Profiling: We All Do It. Law and Order, 47(4), 4. Canter, D. (2003). Mapping Murder: The Secrets of Geographical Profiling. London: Virgin Books. Canter, D., Missen, C., & Hodge, S. (1996). Are Serial Killers Special? Policing Today, 2, 1-12. Caputi, J. (1990). The New Founding Fathers: The Lore and Lure of the Serial Killer in Contemporary Culture. The Journal of American Culture, 13(3), 1-12. Caputi, J. (1999). Small Ceremonies: Ritual in Forrest Gump, Natural Born Killers, Seven, and Follow Me Home. In C. Sharrett (Ed.), Mythologies of Violence in Postmodern Media (pp. 147-175). Detroit: Wayne State University Press. Carney, T. (2004). Practical Investigation of Sex Crimes: A Strategic and Operational Approach. USA: CRC Press. 326 Chan, J. (1996). Changing Police Culture. British Journal of Criminology, 36(1), 109134. Chan, J. (1997). Changing Police Culture: Policing in a Multicultural Society. New York: Cambridge University Press. Chan, J. (1999a). Governing police practice: limits of the new accountability. British Journal of Sociology, 50(2), 251-270. Chan, J. (1999b). Police Culture. In D. Dixon (Ed.), A Culture of Corruption: Changing an Australian Police Service (pp. 98-137). Sydney: Hawkins Press. Chan, J. (2003a). Organizational Socialization and Professionalism. In J. Chan (Ed.), Fair Cop: Learning the Art of Policing (pp. 3-40). Toronto: University of Toronto Press. Chan, J. (2003b). Taking On the Culture. In J. Chan (Ed.), Fair Cop: Learning the Art of Policing (pp. 201-247). Toronto: University of Toronto Press. Chan, J., Brereton, D., Legosz, M., & Doran, S. (2001). E-policing: The Impact of Information Technology on Police Practices. Brisbane: Queensland Criminal Justice Commission. Chatterton, M. R. (1995). The Cultural Craft of Policing- Its Past and Future Relevance. Policing and Society, 5, 97-107. Chavez, A. S., & Gibson, D. C. (2004). Polygraph Keys vs. Publicity: Serial Murder Public Relations. Public Relations Quarterly, 49(1), 10-14. Chermak, S. (1998). Predicting Crime Story Salience: The Effects of Crime, Victim, and Defendant Characteristics. Journal of Criminal Justice, 26(1), 61-70. Commissioner for Children Tasmania. (2006). Advice to the Minister for Health and Human Services: A Child Death Review Process for Tasmania. Hobart: Commissioner for Children Tasmania. 327 Cook, P. E., & Hinman, D. L. (1999). Serial Murder. In H. V. Hall (Ed.), Lethal Violence: A Sourcebook on Fatal, Domestic, Aquaintance and Stranger Violence. Boca Raton: CRC Press. Corsianos, M. (2003). Discretion in detectives' decision making and 'high profile' cases. Police Practice and Research, 4(3), 301-314. Cowper, M. (2000, July 5). Women warned as attacks continue. Parramatta Advertiser. Crim Trac. (2008a). Crim Trac: National Criminal Investigation DNA Database. Retrieved 4 June, 2008, from http://www.crimtrac.gov.au/systems__projects_national_criminal_investigation _dna_... Crim Trac. (2008b). Fingerprints. Retrieved 7 July, 2008, from http://www.crimtrac.gov.au/systems_projects/Fingerprints.html Crimes Act 1900 (NSW), 61I (1900). Crome, S. A., & McCabe, M. P. (2001). Adult rape scripting within a victimological perspective. Aggression and Violent Behavior, 6, 395-413. Davies, A. (1992). Rapists' Behaviour: A Three Aspect Model as a Basis for Analysis and the Identification of Serial Crime. Forensic Science International, 55, 173194. Davies, M., & Mouzos, J. (2007). Homicide in Australia: 2005-06 National Homicide Monitoring Program annual report (No. 77). Canberra: Australian Institute of Criminology. Deans, G. (Writer) (2006). The Backpacker Murders. In G. McNeice (Producer). Australia: Crime and Investigation Network. Denzin, N. K., & Lincoln, Y. S. (2005). Introduction: The Discipline and Practice of Qualitative Research. In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage Handbook of Qualitative Research (3rd ed., pp. 1-32). Thousand Oaks Sage. 328 Department of Immigration and Citizenship. (2007). Fact Sheet 86. Overstayers and People in Breach of Visa Conditions. Canberra: Department of Immigration and Citizenship. Deu, N. E., R.J. (1997). The role of criminal fantasy in predatory and opportunistic sex offending. Journal of Interpersonal Violence, 12(1), 19-29. Devery, C. (2003). Learning at the Academy. In J. Chan (Ed.), Fair Cop: Learning the Art of Policing (pp. 80-144). Toronto: University of Toronto Press. Dixon, D. (1999a). Issues in the Legal Regulation of Policing. In D. Dixon (Ed.), A Culture of Corruption: Changing an Australian Police Service (pp. 36-68). Sydney: Hawkins Press. Dixon, D. (1999b). Reform, Regression and the Royal Commission into the NSW Police Service. In D. Dixon (Ed.), A Culture of Corruption: Changing an Australian Police Service (pp. 138-179). Sydney: Hawkins Press. Dixon, D. (Ed.). (1999c). A Culture of Corruption: Changing an Australian Police Service. Sydney: Hawkins Press. Douglas, J. E., & Olshaker, M. (1997). Journey Into Darkness. New York: Scribner. Easteal, P. W. (1992). Rape (No. 1). Canberra: Australian Institute of Criminology. Egger, S. A. (1984). A Working Definition of Serial Murder and the Reduction of Linkage Blindness. Journal of Police Science and Administration, 12(3), 348357. Egger, S. A. (1990a). Linkage Blindness: A Systemic Myopia. In S. A. Egger (Ed.), Serial Murder: An Elusive Phenomenon (pp. 163-176). London: Praeger. Egger, S. A. (1990b). Serial Murder: A Synthesis of Literature and Research. In S. A. Egger (Ed.), Serial Murder: An Elusive Phenomenon (pp. 3-34). London: Praeger. Egger, S. A. (1998). The Killers Amoung Us: An Examination of Serial Murder and Its Investigation. New Jersey: Prentice-Hall. 329 Engel, R. S., Calnon, J. M., & Bernard, T. J. (2002). Theory and Racial Profiling: Shortcomings and Future Directions in Research. Justice Quarterly, 19(2), 249273. Erainçlioglu, Z. (2004). The Illustrated Guide to Forensics: True Crime Scene Investigations. Dubai: Carlton Books Limited. Ericson, R. V. (1993). Making Crime: A Study of Detective Work. Toronto: University of Toronto Press. Evans, C. (1996). The Casebook of Forensic Detection. New York: John Wiley & Sons Inc. Fattah, E. A. (1997). Criminology: Past, Present and Future. London: Macmillan Press Ltd. Fineman, M., & McCluskey, M. (1997). Feminism, Media and the Law. New York: Oxford University Press. Finnane, M. (1999). From Police Force to Police Service? Aspects of the Recent History of the New South Wales Police. In D. Dixon (Ed.), A Culture of Corruption: Changing an Australian Police Service (pp. 6-35). Sydney: Hawkins Press. Fisher, J. C. (1997). Killer Among Us: Public Reactions to Serial Murder. Westport: Praeger Publishers. Fleming, L. (1994). Serial Murder. Australian Police Journal, 48(2), 59-72. Fontana, A., & Frey, J. H. (2003). The Interview: From Structured Questions to Negotiated Text. In N. K. Denzin & Y. S. Lincoln (Eds.), Collecting and Interpreting Qualitative Materials (2 ed.). Thousand Oaks: Sage Publications. Fox, J. A., & Levin, J. (1998). Multiple Homicide: Patterns of Serial and Mass Murder. Crime and Justice, 23, 407-455. 330 Fox, J. A., & Levin, J. (1999). Serial Murder: Popular Myths and Empirical Realities. In M. D. Z. Smith, M.A. (Ed.), Homicide: A Sourcebook of Social Research (pp. 165-175). London: Sage Publications. Fox, J. A., Levin, J., & Quinet, K. (2005). The Will to Kill: Making Sense of Senseless Murder (2nd ed.). Boston: Peason Educatioon Group Inc. Genge, N. E. (2004). The Forensic Case Book: The Science of Crime Scene Investigation. London: Ebury Press. Giannangelo, S. J. (1996). The Psychopathology of Serial Murder: A Theory of Violence. London: Praeger Publishers. Gibbons, D. C. (1988). Some Critical Observations on Criminal Types and Criminal Careers. Criminal Justice and Behavior, 15(1), 8-23. Gibson, D. C., & Chavez, A. S. (2004). Serial Murder Public Relations Tactics. Public Relations Quarterly, 49(4), 11-18. Godwin, G. M. (2000). Hunting Serial Predators: A Multivariate Classification Approach to Profiling Violent Behavior. London: CRC Press. Golafshani, N. (2003). Understanding Reliability and Validity in Qualitative Research. The Qualitative Report, 8(4), 597-607. Goldsworthy, T. J. (2002). The Hunters Amoung Us: Serial Killers. Australian Police Journal, 56(4), 274-284. Gray, D. E. (2004). Doing Research in the Real World: Qualitative and Quantitative Approaches. Thousand Oaks: Sage. Greer, C. (2003). Sex crime and the media. Collumpton: Willan Publishing. Groth, A. N., Burgess, W., & Holmstrom, L. L. (1977). Rape: power, anger and sexuality. The American Journal of Psychiatry, 134(11), 1239-1243. 331 Hagan, M. (1992). The Police Perspective: Special Issues in Serial Murder. Paper presented at the Homicide : patterns, prevention and control : proceedings of a conference, Canberra. Hale, R., & Bolin, A. (1998). The Female Serial Killer. In R. M. Holmes & S. T. Holmes (Eds.), Contemporary Perspectives on Serial Murder (pp. 33-58). Thousand Oaks: Sage Publications. Hazelwood, R., & Burgess, W. (2001). Practical Aspects of Rape Investigation: A Multidisciplinary Approach (3 ed.). Boca Raton: CRC Press. Henderson, M., & Henderson, P. (1998). Missing People: Issues for the Australian Community. Canberra: National Missing Person Unit: Australian Bureau of Criminal Intelligence. Henderson, M., Henderson, P., & Kiernan, C. (2000). Missing Persons: Incidence, Issues and Impacts. Trends & Issues in Crime and Criminal Justice, 144. Herald Sun. (1992, September 22). Tourists 'butchered' Police scour bush as body is identified. Herald Sun. Hickey, E. W. (1997). Serial Murderers and Their Victims (2nd ed.). Belmont: Wadsworth Publishing Company. Hickey, E. W. (2002). Serial Murderers and Their Victims (3rd ed.). Belmont: Wadsworth. Hodder, I. (2003). The Interpretation of Documents and Material Culture. In N. K. Denzin & Y. S. Lincoln (Eds.), Collecting and Interpreting Qualitative Materials (2 ed., pp. 155-175). Thousand Oaks: Sage Publications. Holmes, R. M., & DeBurger, J. E. (1985). Profiles in Terror: The Serial Murderer. Federal Probation, 49(3), 29-34. Holmes, R. M., & DeBurger, J. E. (1998). Profiles in Terror: The Serial Murderer. In R. M. Holmes & S. T. Holmes (Eds.), Contemporary Perspectives on Serial Murder (pp. 5-16). Thousand Oaks: Sage Publicarions. 332 Holmes, R. M., Hickey, E. W., & Holmes, S. T. (1998). Female Serial Murderesses: The Unnoticed Terror. In R. M. Holmes & S. T. Holmes (Eds.), Contemporary Perspectives on Serial Murder. Thousand Oaks: Sage Publications. Holmes, R. M., & Holmes, S. T. (1998). Serial Murder (2 ed.). Thousand Oaks: Sage Publications. Hoysted, P., & Kidd, P. (2002). Shallow Graves: The Concealments of Killers. Victoria: The Five Mile Press. Innes, M. (1999). The media as an investigative resource in murder enquiries. The British Journal of Criminology, 39(2), 269-287. Innes, M. (2002). The 'Process Structures' of Police Homicide Investigations. British Journal of Criminology, 42, 669-688. Innes, M. (2003). Investigating Murder: Detective Work and the Police Response to Criminal Homicide. Oxford: Oxford University Press. James, M., Anderson, J., & Putt, J. (2008). Missing Persons in Australia (No. 86). Canberra: Australian Institute of Criminology. Jenkins, P. (1994). Using Murder: The Social Construction of Serial Homicide. New York: Aldine de Gruyter. Kelleher, M. D., & Kelleher, C. L. (1998). Murder Most Rare: The Female Serial Killer. London: Praeger. Kennedy, L. (2007, 4 June). Police cool on cold cases. Sydney Morning Herald. Keogh, K. (1997, 28 February). Hunt for rapist: Police special task force formed after 28 attacks. The Daily Telegraph, p. 9. Keppel, R. D., & Birnes, W. J. (2003). The Psychology of Serial Killer Investigations: The Grisly Business Unit. Sydney: Academic Press. 333 Keppel, R. D., & Walter, R. (1999). Profiling Killers: A Revised Classification Model for Understanding Sexual Murder. International Journal of Offender Therapy and Comparative Criminology, 43(4), 417-437. Keppens, J., & Schafer, B. (2004). "Murdered by persons unknown"- Speculative Reasoning in Law and Logic. In T. Gordon (Ed.), Legal Knowledge and Information Systems. Jurix 2004: The Seventeenth Annual Conference (pp. 109118). Amsterdam: IOS Press. Kidd, P. B. (1999). Never to be Released: Volume 2. Sydney: Harper Collins Publishers. Kidd, P. B. (2000). Australia's Serial Killers: The Definitive History of Serial Multicide in Australia. Sydney: Pan Macmillan Australia Pty Ltd. Kidd, P. B. (2005). The Australian Crime File: A 'best of' collection of notorious true crime stories. Rowville: The Five Mile Press. Kidd, P. B. (2008). John Wayne Glover: The Granny Killer, from http://www.trutv.com/library/crime/serial_killers/predators/glover/killer_1.html Kiger, K. (1990). The Darker Figure of Crime: The Serial Murder Enigma. In S. A. Egger (Ed.), Serial Murder: An Elusive Phenomenon (pp. 35-52). London: Praeger. Killmier, B. (1997). Tracking Violent Crime. Paper presented at the Second national outlook symposium : violent crime, property crime and public policy, Canberra. Knight, R. A. (1999). Validation of a Typology for Rapists. Journal of Interpersonal Violence, 14(3), 303-330. Knight, R. A., Warren, J. I., & Reboussin, B. J. (1998). Predicting Rapist Type From Crime-Scene Variables. Criminal Justice and Behavior, 25(1), 46-80. Kocsis, R. N. (1999). Psychological Profiling Techniques for Australian Serial Murder Investigations. Victoria Police Association Journal, 65(10), 22-23,25. 334 Kocsis, R. N. (2002). Profiling Australian Serial Rapists. New South Wales Police News, 82(2), 19-21. Kocsis, R. N., Cooksey, R. W., & Irwin, H. J. (2002). Psychological Profiling of Offender Characteristics From Crime Behaviors in Serial Rape Offences. International Journal of Offender Therapy and Comparative Criminology, 46(2), 144-169. Krueger, R. B., & Kaplan, M. S. (2000). The Nonviolent Serial Offender: Exhibitionism, Frotteurism, and Telephone Scatologia. In L. B. Schlesinger (Ed.), Serial Offenders: Current Thoughts and Recent Findings (pp. 103-120). London: CRC Press. Lane, B. (2004). The Encyclopedia of Forensic Science: A Guide to the Science Behind Crime Investigation. London: Magpie Books. Lee, J. A. (1990). Report of the Royal Commission Inquiry into the Arrest, Charging and Withdrawal of Charges Against Harold James Blackburn and Matters Associated Therewith. Sydney: NSW Government. Legislative Council. (1996). Sexual Violence: Addressing the Crime (No. 9): Parliament of NSW Standing Committee on Social Issues. Lerner, M., & Miller, D. (1978). Just world research and the attribution process: looking back and ahead. Psychological Bulletin, 85, 1030-1051. Levin, J., & Fox, J. A. (1999, 11/05/99). A Culture That's Ripe to Sustain Evil. The Boston Herald, p. 25. Leyton, E. (1986). Hunting Humans: The Rise of the Modern Multiple Murderer. London: Penguin Books. Lievore, D. (2002). Non-reporting and hidden recordings of sexual assault in Australia. Paper presented at the Third Australasian Women and Policing Conference: Women and Policing Globally, National Convention Centre, Canberra. Lloyd-Goldstein, R. (2000). Serial Stalkers: Recent Clinical Findings. In L. B. Schlesinger (Ed.), Serial Offenders: Current Thoughts and Recent Findings (pp. 167-186). London: CRC Press. 335 Manning, J. (1997). Serial rape: offender profiling, Second national outlook symposium: violent crime, property crime and public policy. Canberra: Australian Institute of Criminology. Marché, G. E. (1994). The Production of Homicide Solutions: An Empirical Analysis. The American Journal of Economics and Sociology, 53(4), 385-401. Marciniak, L. M. (1998). Adolescent Attitudes Toward Victim Precipitation of Rape. Violence and Victims, 13(3), 287-300. Marks, K. (2003, 22/05/2003). Children killed for 'ruining social life'. Irish Independent. Mavromatis, M. (2000). Serial Arson: Repetitive Firesetting and Pyromania. In L. B. Schlesinger (Ed.), Serial Offenders: Current Thoughts and Recent Findings (pp. 67-102). London: CRC Press. McAdam, H., & Fitts, A. (1999). Why wreck a man's life? The paradox of male sexual assault, Restoration for Victims of Crime Conference. Melbourne: Australian Institute of Criminology and Victims Referral and Assistance Service. McGarry, A. (2005). The Snowtown Murders. Sydney: Australian Broadcasting Corporation. McLaughlin, E., & Muncie, J. (2001). The Sage Dictionary of Criminology. London: Sage Publications. Miller, J., & Schwartz, M. D. (1995). Rape myths and violence against street prostitutes. Deviant Behavior: An Interdisciplinary Journal, 16, 1-23. Mobbs, J. D. (2001). Crimtrac- Technology and Detection. Paper presented at the 4th National Outlook Symposium on Crime in Australia, New Crimes or New Responses, Canberra. Money, J. (1990). Forensic Sexology: Paraphilic Serial Rape (Biastophilia) and Lust Murder (Erotophonophilia). American Journal of Psychotherapy, XLIV(1), 2636. 336 Mouzos, J. (1999). Mental Disorder & Homicide in Australia. Trends & Issues in Crime and Criminal Justice, 133. Mouzos, J. (2000). Homicidal Encounters: A Study of Homicide in Australia 19891999. Canberra: Australian Institute of Criminology. Mouzos, J. (2001). Investigating Homicide: New Responses For An Old Crime. Paper presented at the 4th National Outlook Synopsium on Crime in Australia- New Crimes or New Responses, Canberra. Mouzos, J., & Houliaras, T. (2006). Homicide in Australia: 2004-05 National Homicide Monitoring Program (NHMP) annual report. Canberra: Australian Institute of Criminology. Mouzos, J., & Muller, D. A. (2001). Solvability Factors of Homicide in Australia: An Exploratory Analysis. Trends and Issues in Crime and Criminal Justice, 216. Neuman, W. L. (2003). Social Research Methods: Qualitative and Quantitative Approaches (5th ed.). Boston: Allyn and Bacon. New South Wales Police. (2005). Forensics. Retrieved 4/10/2005, from http://www.police.nsw.gov.au/about_us/structure/support_command/forensic_s ervices/... Newiss, G. (1999). Missing presumed...? The police response to missing persons (No. Paper 114). London: Policing and Reducing Crime Unit. NMPU. (2005). Missing People: Issues for the Australian Community. Retrieved 15/10/2005, from http://www.missingpersons.gov.au/STUDY.pdf Norris, G. (2006). Personal Communication: Conversation re: Wayne Anthony Trindall. Parramatta. Norris, J. (1997). Serial Killers: The Growing Menace. London: Arrow Books Ltd. NSW Child Death Review Team. (2007). Annual Report 2006. Sydney: NSW Commission for Children and Young People. 337 NSW Commission for Children and Young People. (2008). Child Death Review Team. Retrieved 7th June, 2008, from http://www.kids.nsw.gov.au/kids/about/who/childdeath.cfm O'Leary, J. (1978). Community Perceptions of the Problem. Paper presented at the Police Practices and the Preservation of Civil Rights, Washington D.C. O'Loughlin, N., & Billing, P. (2000). Positive Police Culture: Correcting Perceptions. In T. Coady, S. James, S. Miller & M. O'Keefe (Eds.), Violence and Police Culture (pp. 63-108). Carlton South: Melbourne University Press. Office of Police Integrity. (2006). Report on Victoria Police Missing Persons Investigations. Victoria: Office of Police Integrity. Ogan, J., & Alison, L. (2005). Jack the Ripper and the Whitechapel murders: a very Victorian critical incident. In L. Alison (Ed.), The Forensic Psychologist's Casebook: Psychological profiling and criminal investigation (pp. 23-46). Cullompton: Willan Publishing. Page, A. D. (2007). Behind the Blue Line: Investigating Police Officers' Attitudes Toward Rape. Journal of Police and Criminal Psychology, 22, 22-32. Palermo, G. B. (1997). The berserk syndrome: A review of mass murder. Aggression and Violent Behavior, 2(1), 1-8. Paoline III, E. A. (2003). Taking stock: Toward a richer understanding of police culture. Journal of Criminal Justice, 31, 199-214. Parramatta Advertiser. (2000, July 12). Staff fear rapist attacks: Screams for help scare rapist away. Parramatta Advertiser. Petherick, W. (2005). Criminal Profile: Into the Mind of the Killer. Pranham: Hardie Grant Books. Petraitis, V. (1995). The Frankston Murders: The True Story of Serial Killer Paul Denyer. Carrum Downs: Nivar Press. 338 Pinto, S., & Wilson, P. (1990). Serial Murder. Canberra: Australian Institute of Criminology. PMSEIC. (2000). Science, Crime Prevention and Law Enforcement, Prime Minister's Science, Engineering and Innovation Council: Fifth Meeting. Pogrebin, M. (2002). Qualitative Approaches to Criminal Justice. Thousand Oaks: Sage Publications. Polk, K. (1997). A reexamination of the concept of victim-precipitated homicide. Homicide Studies, 1(2), 141-168. Pudney, J. (2005). Snowtown: The Bodies in Barrels Murders. Sydney: Harper Collins. Putt, J., & Higgins, K. (1997). Violence Against Women in Australia: Key Research and Data Issues (No. 6). Canberra: Australian Institute of Criminology. R v Daley, 279 (NSWSC 2002). R v Presta, 40 (NSWCCA 2000). Ratcliffe, J. H. (2003). Intelligence-led Policing. Trends & Issues in Crime and Criminal Justice, 248. Ravens, T. (2006, 21 April). Serial predator jailed for 22 years. AAP General News Wire, p. 1. Rayner, G. (2008, 19 January). Ipswich murders: DNA link to Steve Wright. Telegraph.co.uk. Reiner, R., & Newburn, T. (2008). Police research. In R. D. King & E. Wincup (Eds.), Doing Research on Crime and Justice. Oxford: Oxford University Press. Reppetto, T. A. (1978). The Detective Task: State of the Art, Science or Craft? Policing: An International Journal of Police Strategies and Management, 1(3), 5-10. 339 Ressler, R. K., Burgess, A. W., & Douglas, J. E. (1988). Sexual Homicide: Patterns and Motives. New York: The Free Press. Ressler, R. K., & Schactman, T. (1992). Whoever Fights Monsters. New York: St. Martin's Press. Reuss-Ianni, E., & Ianni, F. A. J. (1983). Street Cops and Management Cops: The Two Cultures of Policing. In M. Punch (Ed.), Control in the Police Organisation (pp. 251-274). Cambridge: MIT Press. Riley, M. (1997, March 7). How police sat on their hands after rapist warning. Sydney Morning Herald, p. 1. Rossmo, D. K. (1995). Place, Space, and Police Investigations: Hunting Serial Violent Criminals. In J. E. W. Eck, D. (Ed.), Crime and Place (Vol. 4). New York: Willow Tree Press, Inc. Rowlands, L. (2000, 2 August). Serial rape charges. The Daily Telegraph, p. 21. Sackmann, S. A. (1992). Culture and Subcultures: An Analysis of Organizational Knowledge. Administrative Science Quarterly, 37(1), 140-161. Sanders, A., & Young, R. (2002). From Suspect to Trial. In M. Maguire, R. Morgan & R. Reiner (Eds.), The Oxford Handbook of Criminology (3rd ed., pp. 10341075). Oxford: Oxford University Press. Sarantakos, S. (2005). Social Research. New York: Macmillan Education Australia. Schein, E. H. (1985). Organizational Culture and Leadership. San Francisco: JosseyBass. Schein, E. H. (1992). Organizational Culture and Leadership (2nd ed.). San Francisco: Jossey-Bass. Schlesinger, L. B. (2000a). Serial Burglary: A Spectrum of Behaviors, Motives, and Dynamics. In L. B. Schlesinger (Ed.), Serial Offenders: Current Thoughts and Recent Findings (pp. 187-206). London: CRC Press. 340 Schlesinger, L. B. (2000b). Serial Homicide: Sadism, Fantasy and the Compulsion to Kill. In L. B. Schlesinger (Ed.), Serial Offenders: Current Thoughts and Recent Findings (pp. 3-22). London: CRC Press. Scott, B. D. (1996). Evaluating Information on the Internet. Computers in Libraries, 16(5), 44-46. Shears, R. (1996). Highway to Nowhere. Sydney: Harper Collins Publishers. Sheather, M. (1994, 13/6/1994). Facing His Accusers: Ivan Milat was charged last week with the murders of seven backpackers. Who. Sheptycki, J. (2004). Organizational Pathologies in Police Intelligence Systems: Some Contributions to the Lexicon of Intelligence-led Policing. European Journal of Criminology, 1(3), 307-332. Simpson, P. L. (2000). Psycho Paths: Tracking the Serial Killer Through Contemporary American Film and Fiction. Carbondale: Southern Illinois University Press. Small, C. (1998). Task Force Air: Case Review: Unpublished document. Smith, N. (2001). Reading between the lines: an evaluation of the Scientific Content Analysis technique (SCAN) (No. 135). London: Home Office. Snook, B., Cullen, R., Mokros, A., & Harbort, S. (2005). Serial murderers' spatial decisions: factors that influence crime location choice. Journal of Investigative Psychology and Criminal Profiling, 2(3), 147-164. Snyder, D. (2005). DNA Links Ga. Man to at Least 9 Md. Rapes. Washingtonpost.com. Stevens, D. J. (1994). Predatory Rapists and Victim Selection Techniques. The Social Science Journal, 31(4), 421-433. Stewart, J. (1999). Police responses to sexual assault in New South Wales. The Journal for Women and Policing(3), 37-41. 341 Stewart, J. (2001). Milat linked to 1970s missing persons case. Retrieved 3-11-2003, from http://www.abc.net.au/pm/s311748.htm Stone, M. H. (2001). Serial Sexual Homicide: Biological, Psychological, and Sociological Aspects. Journal of Personality Disorders, 15(1), 1-18. Sunday Mail. (1989a, December 3). Granny killer 'sure to strike again'. Sunday Mail. Sunday Mail. (1989b, 26 November). Police step up 'granny killer' hunt. Sunday Mail. Swanborough, G. (Writer) (2004). Paul Denyer. In F. Baker & G. Swanborough (Producer), Forensic Investigators. Australia: Southern Star Entertainment. Swanton, B., & Wilson, P. (1989). Research Brief: Missing Persons. Canberra: Australian Institute of Criminology. Sydney Morning Herald. (2006, 24 October). I'm like Lindy, says 'black widow'. Sydney Morning Herald. Thornton, B., Robbins, M. A., & Johnson, J. A. (1981). Social perception of the rape victim's culpability: the influence of respondent's personal-environmental causal attribution tendencies. Human Relations, 34, 225-237. Tilley, N. (2003). Community policing, problem-oriented policing and intelligence-led policing. In T. Newburn (Ed.), Handbook of Policing (pp. 311-339). Cullompton: Willian Publishing. Tong, S., & Bowling, B. (2006). Art, Craft and Science of Detective Work. The Police Journal, 79, 323-329. Toy, N. (2000, July 1). Serial rapist's picture spearheads police hunt. The Daily Telegraph, p. 16. Turvey, B. (1998). Deductive Criminal Profiling: Comparing Applied Methodologies Between Inductive and Deductive Profiling Techniques. Retrieved 15th June, 2004, from http://www.corpus-delicti.com/Profiling_law.html 342 Turvey, B. (2002a). Criminal Profiling: An Introduction to Behavioral Evidence Analysis. San Diego: Academic Press. Turvey, B. (2002b). Serial Homicide. In B. Turvey (Ed.), Criminal Profiling: An Introduction to Behavioral Evidence Analysis (pp. 513-528). San Diego: Academic Press. Turvey, B. (2002c). Serial Rape. In B. Turvey (Ed.), Criminal Profiling: An Introduction to Behavioral Evidence Analysis (pp. 529-545). San Diego: Academic Press. Turvey, B. (2002d). Victimology. In B. Turvey (Ed.), Criminal Profiling: An Introduction to Behavioral Evidence Analysis. San Diego: Academic Press. Van Maanen, J. (1978a). Kinsmen in repose: occupational perspectives of patrolmen. In J. Manning & J. Van Maanen (Eds.), Policing: A View From the Street. Santa Monica: Goodyear Publishing Company Inc. Van Maanen, J. (1978b). Observations on the Making of Policemen. In P. K. Manning & J. Van Maanen (Eds.), Policing: A View From the Street (pp. 292-308). Santa Monica: Goodyear Publishing Company Inc. Walsh, B. (1993, 23 October). A town haunted by forest of death. Herald Sun. Warneminde, M. (1997, March 9). Rapist may be policeman. Sun Herald, p. 6. Warren, I., & James, S. (2000). The Police Use of Force: Contexts and Constraints. In T. Coady, S. James, S. Miller & M. O'Keefe (Eds.), Violence and Police Culture (pp. 63-84). Carlton South: Melbourne University Press. Warren, J. I., Reboussin, B. J., Hazelwood, R., Cummings, A., Gibbs, N. A., & Trumbetta, S. L. (1998). Crime Scene and Distance Correlates of Serial Rape. Journal of Quantitative Criminology, 14(1), 35-59. Whittaker, M., & Kennedy, L. (1998). Sins of the Brother: The definitive story of Ivan Milat and the backpacker murders. Sydney: Pan Macmillan Australia Pty Ltd. 343 Williams, K. S. (1997). Textbook on Criminology (3rd ed.). London: Blackstone Press Ltd. Williamson, B. (2002, April 27). Serial rapist gets 22 years for attacks. The Daily Telegraph. Wood, J. (1997). Royal Commission into the New South Wales Police Service: Final Report. Sydney: Government of the State of New South Wales. Writer, L. (1992). Garden of Evil: The Granny Killer's Reign of Terror. Randwick: Ironbark Press. www.austlii.edu.au. The Australasian Legal Information Institute www.crimelibrary.com. truTV Crime Library Zedner, L. (2002). Victims. In M. Maguire, R. Morgan & R. Reiner (Eds.), The Oxford Handbook of Criminology (3rd ed., pp. 419-456). Oxford: Oxford University Press. 344 APPENDIX A 345 UWS HUMAN RESEARCH ETHICS COMMITTEE PROTOCOL APPLICATION FORM To Obtain an Ethics Clearance Research projects involving human participants • • Staff, PhD, Masters Hons candidates see Instruction Sheet 1. 4th Yr Undergrad Hons candidates see Instruction Sheet 2. You are reminded that your project should not commence without prior written approval 1. Title of Project Serial Crime in Australia: Investigative Issues and Practice 2. Indicate whether Staff or Student Application Staff and complete the appropriate section of Question 2 Student 2. a Chief Investigator/s (Staff) Name UWS School Contact details (Telephone, Email) Qualifications UWS School Contact Details (Telephone, Email) Qualifications UWS School Contact details (Telephone, Email) Research Degree being undertaken 2. b Associate Investigator/s (Staff) Name 2. c Chief Investigator (Student) Name 346 Natalie Scerra School of Applied Social & Human Sciences 4. 5. Doctor of Philosophy Associate Professor Meg Smith Name of Student Supervisor 3. n.scerra@uws.edu.au 977924214 0401734779 Address for Correspondence *Students must provide an external mail address 67/169 Horsley Road Panania NSW 2213 Anticipated Duration of the Project Proposed commencement date: June 2005 Proposed completion date: December 2006 Funding Is the research project the subject of an application to an internal or external grants body, or will be granted other funding support? Yes If yes, Name of the funding agency What was the outcome? No _______________________________________________ Approved Pending Refused Are there any constraints placed on the release of research data by the funding body Yes No If yes, provide details 6. Lay Summary of Project Write in words a layperson would understand a brief description of the project, ie use ‘Plain English’ First and foremost, this research aims to compile and systematically analyse a previously disparate body of procedures, practices and experiencebased knowledge of investigators on serial crime. Thus, this research will bridge the gap in knowledge of the methods employed to investigate serial crimes in Australia and the issues and problems that arise in the practices and techniques employed. The aim of this research is to determine what factors impact upon, and contribute to the investigation of serial crime in Australia. These include the investigative practices and policies such as: the allocation of resources; the identification criteria for serial crimes; and the methods utilised to link these crimes. The investigative aids that will be considered include: forensic techniques; and criminal profiling. These practices and techniques will 347 be considered to determine what, if any, issues or problems arise in the utilisation of these techniques, in particular how it relates to organisational practices. Serial crimes such as murder and rape, which are the focus of this research, have become increasingly evident in Australia. A serial crime can be defined where two or more crimes occur over a period of time, which can be linked through the similarities between each case, and the belief that these offences were committed by the one offender. Serial crimes provide a number of issues for investigators. Due to the random nature of these crimes, investigators often face many challenges when investigating an incidence of serial crime. The contribution that the various police practices and techniques have to the investigation of serial crime as a whole has been neglected by past research. This research aims at expanding upon previous research. Previous academic research in the area of serial crime has focused on the offenders of these crimes, in particular the motivations for committing the crimes and the background of the offenders. Other researchers have focused their attention on the accuracy of investigative aids such as criminal profiling, but not in the context of other investigative practices or issues. A study conducted by the Australian Institute of Criminology found that investigators believed that there were problems with the time delay in receiving the forensic report, and that it was often phrased in a way which they couldn’t understand. In this same study investigators identified the allocation of resources as a factor which investigators believed affected the outcome of the investigation. This study however did not consider serial crime in any capacity as it related only to homicide cases in general. 7. Scientific, Educational or Cultural Aims of Project Refer to Guidelines Section 7. for information on details required There are a number of aims and objectives of this research: • To gain an understanding of what police practices and policies affect the investigative process and to what extent these practices and policies are seen to affect the outcome • To gain an understanding of how and to what extent investigative aids such as forensic techniques and analysis and criminal profiling are utilized in a serial crime investigation in NSW, what are the issues or problems with utilizing these aids and how do they contribute to the investigative outcome • To compile and systematically analyse a previously disparate body of procedures, practices and experience-based knowledge of investigators on serial crime 348 • 8. To fill the gap in criminological knowledge of the issues that arise in the investigation of serial crime in Australia and how these issues affect the investigative outcome and how these issues relate to organisational practices. Value and Benefits of Project Refer to Guidelines Section 8 for information on details required (use approx. the space provided and type in 12 point font) This study aims to bridge the gap in criminological research of serial crime investigation in Australia through the compilation and systematic analysis of procedures, practices and experience-based knowledge of investigators of these crimes. It considers the problems police face in the investigation of serial crime and how these problems can be countered. How future investigations of serial crime can be improved will be addressed through identifying gaps in knowledge, resources, or training of NSW Police from their previous experience. 9. Details of Research Methodology/Procedures and Recruitment Procedures Refer to Guidelines Section 9 for information on details required. There are two methods that will be employed in this qualitative study: case studies of serial crime incidents and in-depth semi-structured interviews of members of the NSW Police who have worked on serial crimes in an investigative or forensic capacity. This use of two methods in conjunction with the extensive literature review, triangulation ensures a complete set of data is obtained. Case studies will provide information in relation to the issues surrounding investigative techniques and policies. This will provide information about the pressures on police to solve the crime through the utilisation of secondary data such as biographical and media accounts of serial crimes which detail factors relating to the investigation. Thematic analysis will be employed to analyse case study data, the results of which will be compared with the interview data. Such information will also be used in the development of interview schedule. A preliminary thematic analysis of secondary case study data in conjunction with a thorough literature review has been utilised to develop the draft interview schedule (attached). The second research method is in-depth semi-structured interviews, which will provide primary data to compare with the results obtained from the first stage of research. Interviews will be voice-recorded and transcribed at a later date. Interviews will then be subject to a thematic analysis the findings of which will be compared to the themes and data gathered from the case studies. Many issues may not be apparent in these accounts and thus the interviews will 349 enable a richer, more accurate picture of the issues. Data from interviews may also identify lacuna arising from the case studies. By conducting such a comparison between the investigator’s opinion of such practices and the recording of such practices by a third person, the accuracy and applicability of such opinions can be determined. A computer program specifically designed to analyse qualitative data such as QSR NVivo will be used where appropriate. Participants will be recruited through the NSW Police. Contact has been made with the research coordinator of the NSW Police, Mr. Chris Devery, who has agreed to recommend suitable participants or identify the squads that need to be contacted to access suitable participants. Recruitment will utilise a ‘snowball sample’ to access a larger number of suitable participants. Participants will include NSW Police officers and forensic technicians who have worked on a serial crime investigation. Research approval will need to be gained from NSW Police. However, I have been advised that approval can only be sought after receiving approval from the UWS Ethics Committee. Prior to the interviews participants will receive a consent form and Information Statement detailing the aims of the research. Participants may withdraw at any time. Participants will be given ample time to read the information statement, ask the researcher any questions they may have and consider the consent form. The consent form and the Information Statement are attached. Interviews will be conducted at a time and place convenient and suitable for the purpose. The anonymity of participants will be assured. Information collected from the interviews will be coded and de-identified prior to thematic analysis. Confidentiality will be maintained at all times. 10. Participant Sample Description and Sample Size Refer to Guidelines Section 10 for further information on details required. You must clearly define the ‘type’ of participants, and in particular if the participants are children or young people, persons with an intellectual or mental impairment, highly dependent on medical care, persons in dependent or unequal r elationships, p ersons from a distinct collective, or Aboriginal and Torres Strait Islander peoples. The participants will be adults, over the age of 18, of no specific race, gender, or age. Persons with an intellectual or mental impairment, highly dependent on medical care, persons in dependent or unequal relationships, or Aboriginal and Torres Strait Islander peoples will not be targeted. 350 Participants will be drawn from the research population employed by NSW Police currently or previously as identified by NSW Police who have contributed to the investigation of serial crime. Forensic technicians employed by the NSW Police will also be recruited as participants in this research. The number of participants will be dependant upon the availability of suitable participants. However it is anticipated that interviews be obtained from as many suitable participants as possible (for example chief investigator, senior forensic technician). As this research is aimed at collating data that will generate a saturation of qualitative data rather than attempting to create any sort of statistical or quantitative outcomes, it is not the intention to interview large numbers of individuals. The participants will be asked to take part in semi-structured interviews. All information gathered will be treated in strictest confidence and stored securely. The respondents will be made aware that should they decide, at any stage of the research, that they now longer wish to participate then they can withdraw. Respondents will be made aware that no data will be reported that compromises their anonymity. Values and Ethics – Guidelines for Ethical Conduct in Aboriginal and Torres Strait Islander Health Research Researchers, both staff and students who are conducting research with Aboriginal and Torres Strait Islander participants are required to review the Values and Ethics – Guidelines for Ethical Conduct in Aboriginal and Torres Strait Islander Health Research. (see Guidelines for more detail) Criminal Checks In s ome ci rcumstances, i nvolving c hildren, i t may b e n ecessary for t he r esearchers t o h ave undertaken a cr iminal record check. Y ou should indicate if this is required and provide a copy of the documentation. (use approx. the space provided and type in 12 point font) The researcher/s will be required to undertake a criminal check Yes No 11. Evaluation of Potential Harm or Risk of Harm from Research Procedures Refer to Guidelines Section 11 for information on details required Does the project involve any of the following research procedures? a) the possibility of physical stress/distress, discomfort to the participants Yes If yes, how will the potential harm or risk of harm be addressed? 351 No b) the possibility of psychological/mental stress/distress, discomfort Yes No to the participants. If yes, how will the potential harm or risk of harm be addressed?: It is not anticipated that interviews will trigger any form of discomfort as interviewees are experienced criminal investigators who deal with these issues on a day-to-day basis. However, as interviewees will be required to discuss criminal cases that they have investigated in case it may trigger their emotional responses to the crimes, the New South Wales Police Service have their own internal counselling services that members have access to. However if interviewees do not wish to utilise the services provided by the NSW Police then participants can contact Dr. Fiona Rummery Senior Lecturer at the University of Western Sydney by phone on (02) 9685 9120 or by email f.rummery@uws.edu.au . Dr. Rummery has extensive counselling experience. This will also be addressed by including the contact number of my supervisor who will put them in contact with a qualified counselor, on the Information Statement attached to the survey. Participants will also be informed that they can withdraw at any time without penalty or prejudice. c) deception of participants at any stage of the project Yes If yes, give details, will the participants be debriefed after the procedure? No No d) the use of drugs or invasive procedures Yes For example, chemical compounds, drugs, ionising or non-ionising radiation, other biological agents, human tissue, special diets or modified foods, sensory evaluation of genetically modified food. Evidence of clearance from the University Biosafety Committee should be submitted with your application. If yes, give details as to how any potential harm or risk of harm will be minimised: 12. Potential Ethical Issues 352 You should address each potential ethical issue as it relates to your project (use approx. the space provided and type in 12 point font) 12 a. Integrity, Respect for persons, Beneficence and Justice Refer to Guidelines Section 12 a for further information on details required. You must provide a statement that demonstrates that you have considered the ethical principles of Integrity, Respect for Persons, Beneficence and Justice as it relates to your research project. The researcher is committed to the principles that guide reputable research, including the principle of beneficence. The researchers supervisors and professional advisor (Mr. Devery) have many years of experience and commitment to research and understand the importance of ethical conduct toward research participants. The researcher will therefore be operating within a framework that reflects the respect for persons and justice that the research requires. All research will be undertaken with care and respect for the respondents’ professional and personal circumstances, with attention paid to the nature of the profession of police work. This research will be conducted with the greatest respect to persons involved and with high integrity. There are no benefits to be gained by the researcher from being dishonest with participants. The researcher is aware of providing for the welfare and rights of all participants. All participants will have the opportunity to ask the researcher any questions they may have and the researcher will respect and consider any such questions seriously. None of the research carried out in relation to this project will disrespect any of the participants. 12 b. Consent You should refer to Guidelines Section 12b for information on detail required. Provide d etails a bout how you will o btain t he agreement o f your p articipants to ta ke p art in th e research. Provide as attachments copies of Information Letters, Consent forms, etc. as set down in this section. All participants will receive a consent form and Information Statement prior to the commencement of interviews. The Information Statement will outline in plain English the aims of this project and the role of participants in this project. Two copies of this will be available, so that participants can retain one copy for themselves and the researcher will retain the other copy. Participants will be given ample time to read the information statement, ask the researcher any questions they may have and consider the consent form. Participants will have the option to remain anonymous if they should wish. This will be outlined on the Information Statement and consent form. If a participant requests to be anonymous, their anonymity and confidentiality will be maintained at all times. There is no benefit to be gained by this study by the 353 identification of participants or by breaching their right to anonymity and confidentiality. 12 c. Research Merit and Safety You should refer to Guidelines Section 12c for information on details required. Your r esponse s hould d emonstrate t he r esearch i s j ustifiable a nd i s d esigned as t o en sure t hat any risks to participants are balanced by the likely benefit/s to be gained. T he research should be co nducted o r s upervised only b y p ersons with e xperience, q ualifications a nd co mpetence appropriate to the research. The risks involved in this research are extremely minimal, as the ethical approach in regards to participant consent and the access to counseling services, reduces the likelihood of this occurring. In 2002 I have conducted and completed research as part of my Bachelor of Social Sciences (Honours) Criminology thesis. The principle supervisor, Associate Professor Meg Smith has extensive experience in conducting research. Associate Professor Meg Smith has also, amongst other things, conducted her own research for her Ph.D. d. Ethical Review and Conduct of Research Refer to Guidelines Section 12 d for information on details required. Provide details about h ow the results of the research will be di sseminated a nd to whom. Y ou must provide details of how the data collected will be stored securely, who will have access and the disposal time line. The contents of the interviews will be voice-recorded and transcribed to enable analysis by the Chief Investigator. All identifying material will be locked in a filing cabinet in my office for the duration of my PhD research. After completion it will be stored in the same manner by my supervisor. Participants will not be identified unless they wish to be and give written permission to be identified. All recordings and related transcripts and related materials will be stored in a locked filing cabinet for a five year period. All material will be accessible by the researcher. A copy of this thesis will be made available to participants at the University of Western Sydney library or by contacting the School of Applied Social Sciences, from early 2007. Participants will be informed of this access to the thesis on the Information Statement. 13. Privacy requirements Refer to Guidelines Section 13 for information on details required for Questions a & b a. Will you be accessing data held by a Commonwealth Department or Agency? Yes 354 No b. Will you be accessing data held by a State Department or Agency? Yes No If the researcher/s answer yes to question a or b they should be aware of and agree to abide by the Privacy Principles in dealing with personal Information. c. Does your research involve the use of existing records, which identify individuals but which are not in the public domain (eg medical or other personal records) that are held by an organisation. Yes Provide details if you answer yes to a, b or c No B. It is my intention upon approval from NSW Police to access the NSW Police Library in Goulburn. Access to this library will provide me with further information regarding police policies and procedures relating to investigative methods. It is not envisaged that I will be accessing any personal data on individuals from the NSW Police Library. Information obtained from the library, although not readily available to the general public, should not consist of restricted or confidential material. I am, however, aware of the Privacy Principles in regards to personal information, and I agree to abide by those principles when dealing with any personal information I may obtain. It is also my intention to discover from NSW Police if access to closed/completed police case files will be available to contribute to the data gathered for the case study component of this study. The information that will be sought from these case files would be specific to investigative practices only and confidential information or information of current operational interest will be treated with the greatest respect and confidentiality. Access to this information will be conditional to permission being granted by NSW Police and conformity to any guidelines or protocol that they establish to ensure the sensitivity and privacy of the data is respected. Again, I am, aware of the Privacy Principles in regards to personal information, and I agree to abide by those principles when dealing with any personal information I may obtain. 14. Ethics Approval from Another Institution Does this research require the ethics approval of another institution? If yes, give details including whether or not you have submitted an application to that institution for ethics approval. (please )Yes No Name of Institution/Human Ethics Committee Application Approved/disapproved Application Pending New South Wales Police Ethics approval from NSW Police is pending ethics approval from the UWS Ethics Committee. Ethics approval from NSW Police is pending ethics approval from the UWS Ethics Committee. 355 15. Chief Investigator/s Declaration (Staff or Student) I/we certify that the information given in this application is correct to the best of my/our knowledge. I/we acknowledge that I/we must notify the Committee if there is any ethically relevant variation or if the project is discontinued prematurely. I/we have read, and agree to abide by the relevant code of practice for research involving humans. Chief Investigator Signature/s: Signed................................................ Signed................................................ Signed................................................ Date……………………………………….. 16. Student Supervisor Declaration I certify that the information given in this application is correct to the best of my knowledge. Supervisor’s Signature: Signed.........................................……………………...... Date………………………………. Before lodging the application please complete this checklist All questions have been answered Chief investigator/s and if applicable Student Supervisor have signed the declaration Supporting documentation is attached Required number of copies plus the original are attached The Human Ethics Officer has reviewed the application Yes No Staff, P hD o r M asters ( Hons) can didate - Application h as b een s ubmitted to th e UWS H uman Ethics Research Committee (See Instruction Sheet 1.) Students – Bachelors (Hons) candidates, Masters (Course Work) - Application has been submitted to the UWS Human Research Ethics Student Panel (See Instruction Sheet 2.) Application submitted.....................................(Date) End of Application Research Services Web Page: http://www.uws.edu.au/about/adminorg/devint/ors/ethics/humanethics 356 Interview Schedule- Police Semi structured interview schedule. Questions in capital letters are the questions that will be asked. The questions in point form below each main question are prompts to aid the interviewer. Name Age Rank/Position Number of years with NSW Police HOW MANY SERIAL CRIME INVESTIGATIONS HAVE YOU BEEN PART OF? • What type of crimes were involved- murder, rape, arson HAVE THESE CASES BEEN SOLVED, LEADING TO THE PROSECUTION OF AN ALLEGED OFFENDER? - Yes • Was a conviction obtained • If a conviction was obtained who was the offender • If a conviction was not obtained, what was the reason for this - No • What were some of the reasons for this • Was there a lack of evidence, no viable suspect AT WHAT STAGE OF THE INVESTIGATION WERE DID YOU BECOME INVOLVED ● After how many reported victims • In what capacity/ what was your role • Had you any previous experience in serial crime investigations • Were you reassigned to another case at any time during the investigation • What was the reason for this- progress of investigation at the time of reassignment AT WHAT STAGE WERE THE CRIMES CONSIDERED TO BE RELATED OR SERIAL • How long after the second reported victim • Was there any problems in linking these crimes • Was the violent crime linkage analysis system used to help link these crimes • Did this effect/impact the investigation in any way • Were more manpower allocated/resources • Was a taskforce set up SERIAL MURDER- WERE ANY OF THE VICTIMS REPORTED MISSING FOR A PERIOD OF TIME BEFORE THE DISCOVERY OF THEIR REMAINS 357 • • • • Was their disappearance investigated in relation to a serial crime before the remains were found If so, what suggested that this victim was linked to other cases If not, what is your opinion as to why this was If more than one victim was reported missing prior, were these disappearances linked WHAT FORENSIC TECHNIQUES WERE EMPLOYED IN THIS INVESTIGATION • Was DNA profiling used to link related crimes • Was DNA profiling used to identify a suspect • To what extent was fingerprint analysis used and how beneficial was it to the investigation • Was other techniques used such as blood pattern/stain analysis, ballistics. • Were there any forensic techniques not used but you feel would have been beneficial to the investigation FOR THE FORENSIC TECHNIQUES USED IN THE INVESTIGATION DID YOU RECEIVE A REPORT • Did you understand the content of the report and how it applied to your investigation • Did you understand the terminology • Was the terminology explained to you at any time • Did a lack of understanding of the terminology effect the investigation in any way • Was there a delay in receiving such a report WAS CRIMINAL PROFILING USED AS AN AIDE IN THE INVESTIGATION • How did the profile contribute to the investigation • How much weight was given to the profile • Do you consider the profile to be accurate • If the profile was inaccurate, did this impact the investigation negatively LOOKING BACK ON THE INVESTIGATION IN ITS ENTIRETY, WHAT BENEFITED THE INVESTIGATION MOST • What techniques or practices contributed positively • What policies allowed greater contributions- taskforce/manpower WHAT DO YOU BELIEVE IMPACTED THE INVESTIGATION NEGATIVELY • Did policy effect the investigation- resource allocation, manpower reassignment • Were there any techniques not used that you feel would have been beneficial 358 Interview Schedule- Forensic Technicians/Consultants Semi structured interview schedule. Questions in capital letters are the questions that will be asked. The questions in point form below each main question are prompts to aid the interviewer. Name Age Rank/Position Number of years employed by NSW police in forensics • As a civilian • As a NSW Police member HOW MANY SERIAL CRIME INVESTIGATIONS HAVE YOU WORKED ON? • What was your role • What type of crimes were involved- murder, rape, arson HAVE THESE CASES BEEN SOLVED, LEADING TO THE PROSECUTION OF AN ALLEGED OFFENDER? - Yes • Was a conviction obtained • If a conviction was obtained who was the offender • If a conviction was not obtained, what was the reason for this - No • What were some of the reasons for this • Was there a lack of evidence, no viable suspect DID ANY OF THE FORENSIC EVIDENCE CONTRIBUTE TO AN ARREST OR CONVICTION • What type of forensic evidence contributed to this • To what extent • Was it used as evidence in court AT WHAT STAGE OF THE INVESTIGATION DID YOU BECOME INVOLVED? • From first reported victim • Part of a team that was involved from the beginning • What form of forensic analysis did you conduct • What other forms of forensic analysis were conducted by others according to your knowledge WERE YOU PHYSICALLY INVOLVED IN COLLECTING EVIDENCE FROM ALL RELATED CRIMES? 359 • • • If yes, do you believe that this is an important measure in collecting evidence from related cases If no, was the other technician/s from your office/team Do you believe that this could impact an investigation- different perspectives, not aware of similar evidence WERE THERE ANY ISSUES/PROBLEMS IN THE COLLECTING OF EVIDENCE • Were crime scenes secured adequately to protect the evidence DID THE FACT THAT THESE CASES WERE CONSIDERED AS SERIAL IMPACT THE WAY IN WHICH YOU TREATED/VIEWED THE EVIDENCE • How, to what extent WHAT WAS THE AVERAGE TIMEFRAME BETWEEN THE COLLECTION OF EVIDENCE AND THE AVAILABILITY OF A REPORT TO POLICE • Is this a common/standard timeframe • Were any problems associated with processing the evidence • In your experience did the police have any issues in the timeframe TO YOUR KNOWLEDGE DID THE POLICE HAVE ANY ISSUES IN UNDERSTANDING THE FORENSIC REPORTS • If so, what type of problems arose • Were issues explained to them if they requested it CONSIDERING YOUR ROLE AS A WHOLE IN THE INVESTIGATION, IS THERE ANYTHING THAT YOU CONSIDER MAY HAVE IMPROVED THE FORENSIC ASPECT • Improvement in services, access to evidence 360 Information Sheet for Participants Study of the Investigation of Serial Crime in Australia Project Description This s tudy i s part of a Doctor of P hilosophy p roject at t he University o f Western Sydney. The aim of this research is to determine what factors affect the investigation of s erial crime in Australia: in particular, the impa ct of forensic techniques and criminal profiling on the outcome of an investigation. The impact of pol icies, pr ocedures a nd pr actices of t he N SW Police S ervice, a nd how t hey contribute to an investigation of serial crime will also be considered. In order to participate in this study, you must have contributed to the investigation of serial crime in NSW, in any capacity. Your Role & How It May Affect You As a pa rticipant, you will be a sked t o answer several qu estions w hich w ill identify what you, as a member of, or employee of, the NSW Police, perceive as issues s urrounding t he i nvestigation of s erial crime. For e ach question, you will also ha ve t he opportunity t o add a ny c omments y ou may f eel necessary. T his interview will take approximately 45-60 minutes to complete. If you do not w ish t o pa rticipate, w e r espect y our decision. T here i s no p enalty or disadvantage to you, for not participating. You may also withdraw from participation at any t ime, w ithout a ny pe nalty or di sadvantage t o y ou. A s a pa rticipant, y ou may be completely anonymous if you so wish. Any information provided by you will be kept in a secure place at the University of Western Sydney. The c ompleted thesis w ill be av ailable i n early 200 7 from t he U niversity of Wes tern Sydney library or by c ontacting t he S chool of A pplied S ocial a nd H uman Sciences by mail, to School of Applied Social and Human Sciences, University of Western Sydney, Locked Bag 1797, Penrith DC, NSW 1797 361 Further Information Any questions concerning the project can be directed to the Chief Investigator Natalie Scerra by email n.scerra@uws.edu.au or the Project Supervisor Associate Professor M eg S mith b y phone on 0297726299 or m obile 0411 030 256 . Or b y mail, to the School of Applied Social and Human Sciences, University of Western Sydney, Bankstown c ampus, Locked B ag 1797, Penrith DC, NSW, 1797, or b y email m.smith@uws.edu.au. If participation in this study has caused you distress and you wish to seek counseling you can either contact Associate Professor Meg Smith who will provide suitable contacts for counseling services. Alternatively, you can contact Dr. Fiona Rummery at the University of W estern S ydney by ph one on (02) 9685 9120 or by e mail f .rummery@uws.edu.au . Note- This s tudy ha s be en a pproved by t he U niversity o f Western S ydney H uman R esearch E thics Committee. I f you have any co mplaints or reservations about the ethical conduct of this research, you may contact the Ethics Committee through the Research Ethics Officers (tel: 02 4736 0883). Any issues you raise will be treated in confidence and investigated fully, and you will be informed of the outcome. 362 SIGNED AGREEMENT/ CONSENT I (the participant) have read the information supplied on the Information Sheet and any questions I have asked have been answered to my satisfaction. I agree that I am participating in a PhD study and that any information I provide to the study may be published in the PhD thesis or any related publications. I understand that a copy of the thesis will be available at the University of Western Sydney library, from late 2006. I agree to participate in this study, realising that I may withdraw at any time without penalty or prejudice. Please read the statements below and select your preference. Your confidentiality will be respected at all times. I wish to remain anonymous and not be identified in the resulting publication/s. I wish to be identified with my statements and give permission for the use of my name in the resulting publication/s. Signature ______________________________ Date ______________ For use by the researcher only: Name____________________________________ Contact Phone Number_____________________________ Please sign this statement and return to researcher. 363 APPENDIX B 365 Example of Preliminary Thematic Cluster Victimology Victim "groups" "stereotypes" ↙ ↓ ↘ Missing persons ↖ ↑ ↗ Police stereotypes associated with victim Innocent ↕ positive Formal days ↙ ↖ ← Other linkage ↕ Impact ↕ Crime Linkage weeks ↖ ↑ ↔ months ↗ ↗ Worthless/Deserving ↕ ↘ negative → Informal ↗ years Formalised linkage system Diagram 1: Preliminary Thematic Cluster- Victimology This is a n example of the preliminary th ematic c lusters us ed as a guideline to analyse the da ta. This di agram ide ntifies the is sues r elating to the vi ctimology of the cases of serial crime. As these factors did not occur in isolation from the other factors which impact on the investigation of serial crimes. Therefore there would be an overlap and repetition of some factors. 366 APPENDIX C 367 Letter to NSW Police Requesting Access to Closed Case Files Natalie Scerra University of Western Sydney Locked Bag 1797 Penrith South DC 1797 To Whom It May Concern: I am a PhD candidate at the University of Western Sydney and would like to conduct interviews of NSW Police employees who have contributed to serial crime investigations. My research aims at addressing the impact of policies, procedures and practice on the investigation of serial crime in Australia. In particular, how forensic techniques and criminal profiling contribute to an investigation of serial crime will also be considered. In order to conduct this research I am requesting permission to access closed/completed cases files from NSW Police. These case files would need to pertain to the serious serial crimes of murder and rape. These case files will only be utilized for the purpose of gaining information about the investigative practices and techniques used in the individual investigations. In accordance with the University of Western Sydney Ethics Approval, I would like to emphasise that the information that will be sought from these closed case files would be specific to investigative practices only and confidential information or information of current operational interest will be treated with the greatest respect and confidentiality and in adherence to State Privacy Laws. The inclusion of confidential information or information of current operational interest would not benefit or contribute to the research in any way. It is understood that access to this information will be conditional to permission being granted by NSW Police and conformity to any guidelines or protocol that they establish to ensure the sensitivity and privacy of the data is respected. If you have any queries you can contact me, the Chief Investigator, by phone on (02)9772 6411 or email n.scerra@uws.edu.au or the Project Supervisor Associate Professor Meg Smith on (02)9772 6299 or m.smith@uws.edu.au. Both the Chief Investigator and the Project Supervisor are contactable by mail to, School of Applied Social and Human Sciences, University of Western Sydney, Bankstown campus, Locked Bag 1797, Penrith DC, NSW, 1797. Yours Sincerely Natalie Scerra Note- This study has been approved by the University of Western Sydney Human Research Ethics Committee. If you have any complaints or reservations about the ethical conduct of this research, you may contact the Ethics Committee through the Research Ethics Officers (tel: 024736 0883). Any issues you raise will be treated in confidence and investigated fully, and you will be informed of the outcome. SERIAL CRIME IN AUSTRALIA Investigative Issues & Practice Research Proposal to NSW Police Submitted by Natalie Scerra University of Western Sydney Phone: (02) 9772 6411, or 0401 734 779 Email: n.scerra@uws.edu.au 388 Serial Crime in Australia: Investigative Issues & Practice AIMS 349 PROPOSED OUTCOMES & BENEFITS BACKGROUND 349 350 Australian Police Context 351 Defining & Linking Related Crimes 351 Police Policy and the Allocation of Resources 352 The Contribution of Forensic Techniques 353 Criminal Profiling as an Investigative Aide 354 The Victim 354 NEED FOR RESEARCH METHODOLOGY 355 356 Time Frame 356 1. Case Studies 356 Sources of Information 357 Sample 357 2. Interviews 358 Consent 358 Sample 359 Interview Method 359 Analysis 360 Confidentiality and Privacy REFERENCES 360 ERROR! BOOKMARK NOT DEFINED. i AIMS First a nd foremost, this research aims to compile a nd systematically analyse a previously disparate body of procedures, practices and experience-based knowledge of investigators on serial crime. Thus, this r esearch will br idge t he gap i n academic knowledge of t he m ethods e mployed t o i nvestigate s erial c rimes i n A ustralia a nd t he issues and problems that arise in the practices and techniques employed. The aim of this research is to determine what factors impact upon, and contribute to the investigation of serial crime in Australia. These include the investigative practices and policies such as: the allocation of resources; the identification criteria for serial crimes; and the methods utilised to link these c rimes. The inve stigative aids tha t w ill be c onsidered include: forensic t echniques; a nd c riminal pr ofiling. T hese pr actices a nd t echniques w ill be considered from an academic viewpoint to determine what, if any, issues or problems arise in the utilisation of these techniques, in particular how it relates to organisational practices. PROPOSED OUTCOMES & BENEFITS This r esearch aims t o i dentify how t he va rious a spects of t he i nvestigative procedures and processes interact and contribute to the outcome of an investigation of serial c rime in Australia. As s erial c rime oc curs inf requently in comparison to other offences, t he a pplication of m odels s uch as intelligence-led pol icing a nd crime management have not been adequately considered by previous academic research in the sphere of serial crimes. It is hoped that this research can lead to more efficient and applicable investigative policy for serial crimes. Not only may this lead to the identification and arrest of serial offenders, but the sooner such an arrest occurs in a serial offender’s career, the greater impact such investigations have as crime prevention measures by reducing the number 349 of victims targeted by such offenders. As serial crime creates uncertainty and fear in a community exposed to these cr imes, efficient detection will eas e t he i ssues of t he community. BACKGROUND Much academic research in this f ield has f ocused on the o ffenders of t hese crimes, in particular the motivations for committing the crimes and the background of the of fenders (see F ox & Levin, 1998; H ickey, 1997; H olmes & H olmes, 1998) . Another f ocus of researchers ha ve be en on t he accuracy of i nvestigative aids s uch a s criminal profiling (Kocsis, 1999; Kocsis, Cooksey, & Irwin, 2002). However s erial cr imes, in particular s erial m urder and s erial r ape, provide a number of i ssues f or i nvestigators. T hese problems a rise f rom t he i nherent characteristics shared by many of these crimes. Characteristics such as the lack of prior relationship between the victims and offender, the lack of apparent motivational drives, and t he of fender’s m obility, of ten hi nder t raditional i nvestigative pr ocedures (Egger, 1998). For example, interviewing family and friends for an indication of motivation is unlikely to be useful. According to Mouzos (2000) Australian police agencies accept the Federal Bureau of Investigation’s (FBI) definition as outlined by Cook and Hinman (1999, quoted in Mouzos, 2000), which states that: there are 2 or more separate homicide events which occur over a period of time (hours, days, weeks, or ev en years). There i s a coo ling-off pe riod be tween e pisodes. These crimes are predatory. The offender frequently stalks his victims. The motive is clearly psychological. The o ffender’s be haviour a nd the crime scen e ev idence t ypically indicate sexual and sadistic features, and may i nvolve t orture and mutilation of t he victim (p.364). 350 In New South Wales, the legal definition of rape, referred to as sexual assault, I “any person who has sexual intercourse with another person without the consent of the other pe rson and w ho knows t hat t he ot her pe rson doe s not c onsent t o t he s exual intercourse…” ("Crimes Act 1900 ( NSW)," Section 61I). The definition for serial rape is however, more difficult to operationalise. Unlike serial murder, definitions of serial rape are not f reely available i n the publ ications of ac ademics. Instead, t he t erm s erial rape is often presumed by many to be self-explanatory. Therefore i n simple t erms, serial r ape can be cons idered as r epeated r ape offending. H owever, w hen w e r efer t o s erial r apists, w e us ually r efer t o predatory rapists w ho vi ctimise s trangers. R apists w ho repetitively offend a gainst s omeone known t o t hem, e xample g irlfriend, or t hose m ake a h abit of r aping many s exual partners over a period of time, a rarely considered as serial rapists. Australian Police Context It is also the researcher’s intention that the findings from this study be available to NSW Police. All of the factors that will be discussed are considered as components of the int egrated crime management mode l. The int egrated crime ma nagement m odel considers t hat f unctions s uch a s t he i nvestigative, i ntelligence, f orensic a ll w ork together t o contribute t o a m ore effective way o f m anaging a nd responding t o crime. This research ma y be ut ilised by Australian Police forces to consider ho w t his model performs when confronted by serious serial crime. Defining & Linking Related Crimes The mobility of serial offender’s may also hinder investigations as investigators of one case may not be aware of the occurrence and/ or evi dence f rom an other related case. One i ssue w hich will be addr essed is t he m ethods or s ystems employed by investigators to determine that cases of serial murder or serial rape are occurring. This 351 relates both to the technologies employed such as the Violent Crime Linkage Analysis System ( ViCLAS) (Killmier, 1997; M ouzos, 2001) and the pol icies r elating to the procedure to make such a determination. By determining that cases are linked and thus may be serial in nature, it is also aimed that it can be ascertained as to what impact this has on t he i nvestigative pr ocesses e mployed and t he pr ocedures t hat ut ilised i n t he investigation of these crimes. Particularly if these differ from conventional investigative processes or procedures employed in the investigation of non-serial crimes. Previously, l ittle r esearch ha s b een conducted i nto how de fining a nd l inking serial crimes can contribute to an investigation of serial crime in Australia. One study conducted b y A dderley and M usgrove (2001) considered several com puter s oftware systems implemented in police forces internationally. One o f t he m ain pr oblems with such linkage systems that were identified in this study was the ability of the investigator to ask the right question of the system. As such systems require that specific questions must be a sked b y t he u sers, i t w as f ound t hat often t he us er would no t know w hat question was relevant thus affecting the ability of the system to be used to link related crimes. Adderly and M usgrove however, di d not consider t he ViCLAS s ystem t hat i s currently utilised in Australia. Thus one aim of this research is that an understanding of the issues Australian police have with a similar system can be gained. The issues of defining and linking serial crime will be considered in the context of pr evious r esearch in the a rea of i ntelligence-led pol icing, as m any o f the conc erns identified above arose from this framework of policing. Police Policy and the Allocation of Resources Another i ssue, w hich m ay i mpact upon a n i nvestigation of s erial c rime, is t he length of time between related cases, which may be up to several months or years. This period of inactivity of the offender may lead to a reduction in police resources. Police policy m ay b e one t hat be lieves t hat pol ice r esources m ay b e m ore f ruitful on m ore recent and possibly solvable cases. A study conducted by Mouzos and Muller (2001) on 352 factors related to the solvability of murder investigations in Australia, found that one of the m ain issues r aised by i nvestigators w as t he l ack of r esources av ailable. This includes a lack of appropriately experienced investigators, support staff, and analysts. Another i mportant resource w hich was i dentified b y M ouzos a nd M uller a s hampering investigations was the lack of time devoted to the process due to the policy that a n i nvestigator c an work on onl y on e m urder a t a t ime. A s t his i ssue doe s not appear to be particularly apparent in serial crime investigations in which task forces are usually formed, this research will aim to determine how this differs. However it is also recognized that there ma y be similar issues relating to support staff, particularly if an investigation loses momentum over a period of time. The Contribution of Forensic Techniques The use of investigative technologies can aid serial cases by providing evidence when t here a re no ot her a venues t o e xplore. T hese t echnologies i nclude t race/fibre evidence, D NA p rofiling, f ingerprinting, a nd bl oodstain a nalysis. A f actor c onsidered important in the inve stigation of s erious c rimes s uch as m urder i s t he collection of evidence from the crime scene (Mouzos & Muller, 2001). However, the contribution of forensic t echniques t o serial cr ime i nvestigations i s cons idered to be a g iven fact and reasearch into this area is in its infancy. According to Mouzos and Muller, investigators considered the securing of a crime scene or scenes as necessary to solve a crime such as murder. The need for forensic specialists at the crimes scene/s was also considered as an i mportant f actor. T his e nsured t hat e vidence w as pr eserved, c ollected a nd uncontaminated. C onversely one o f t he m ajor i mpediments t o a m urder i nvestigation was found to be the length of time taken for the forensic analysis to be completed. However, the study conducted by Mouzos and Muller (2001) was not specific to serial crimes and thus one aim is to determine if there findings regarding resources and the forensic process have been considered an issue by investigators of serial crime. The impact of this information on the investigation will also be considered to determine the 353 extent of which forensics contribute to the outcome of a serial crime investigation. For example was forensic evidence the key factor linking related crimes? Another as pect of t he us e of f orensic t echniques w ill a lso be c onsidered. A timeline of the a vailability o f va rious f orensic te chniques w ill be de termined and considered in relation to how this development affected investigative capabilities over time. Criminal Profiling as an Investigative Aide The us e of m any o f t hese t echnologies are s tandard i n t he i nvestigation of serious crimes, however the use of criminal profiling is a tool which is often utilised in serial crimes in which there are very few leads. The aim of criminal profiling is to limit the focus of the investigation, to suspects which conform to the profile, in an attempt to limit t he s cope of a n i nvestigation pa rticularly w hen t here a re num erous pos sible suspects (Turvey, 2002a ). H owever, c riminal pr ofiling i s not a n e xact s cience, a nd sectors s uch as t he m edia of ten l ay m uch w eight on this t echnique. T here ha ve be en incidents where the criminal profile has been criticized by the media and other outlets for mis leading an investigation. An example of thi s w as the or iginal p rofile of the ‘Granny Killer’ stated that the offender would be young when the offender was middleaged. T hus t he c ontribution t hat this t echnique makes t o a n i nvestigation, how it is utilized and to what extent, is one of the aims of this research. There ha ve be en a num ber of s tudies i nto t he a ccuracy of t he c riminal profile and the l evel o f ‘ expertise’ r equired to form on e, however t his r esearch ai ms t o go beyond this previous form of analysis to consider how useful this tool can be during the actual investigative process. The Victim In addition, the vi ctimology of the se c rimes ma y ma ke it di fficult f or investigators to recognise or establish cases of serial crime. For example some people 354 are more likely than others to live transiently, thus their sudden absence will not cause alarm (Fox & Levin, 1 998). A r ecent ex ample of t his i ssue w as ap parent i n an American case of serial murder, where the offender stated that: I also picked prostitutes as victims because they were easy to pick up without being noticed. I knew they would not be reported missing right away and might never be reported missing. I picked prostitutes because I thought I could kill as many of them as I wanted to without getting caught. (BBC, 2003) This research will aim to determine if the victimology of these crimes impacted the investigation. In particular if victims were r eported missing a fter a p eriod of time had e lapsed t hus i mpacting t he i nvestigative pr ocess. Additionally t he i ssue o f determining a possible link between several reported cases of missing persons through the utilisation of resources such as ViCLAS will be considered. Specifically, if at any stage dur ing a s erial crime i nvestigation, a r eview of m issing p erson r ecords i s conducted t o de termine i f t here a re a ny s imilar c ases. T his w ould a lso apply t o da ta relating to serious sexual assaults or murders. NEED FOR RESEARCH As demonstrated, there are several gaps in the current academic knowledge of the investigation of serial crime. Previous research into the different aspects which can contribute to the investigative process have been limited and few have considered serial crime as a separate investigative issue. As previously stated, serial crime poses several issues for investigators which may not be apparent in other investigations and thus this research aims to identify how serial crime investigation can be seen to differ. 355 METHODOLOGY • • Case Studies Interviews There are tw o methods tha t w ill be e mployed in this qua litative s tudy: c ase studies of serial crime incidents and in-depth semi-structured interviews of members of the various A ustralian Police agencies who have w orked on serial cr imes i n a n investigative or f orensic c apacity. T his us e of t wo m ethods i n c onjunction w ith t he extensive literature review, triangulation ensures a complete set of data is obtained. It has been determined that interviews would be an accurate method to determine police practices and t he us e o f i nvestigative t ools as s tated by t he i nvestigators t hemselves. The opi nions of t hese i nvestigators a re e ssential i n e stablishing how be neficial s uch tools were, if police policy as previously outlined, impacted upon the investigation and to what extent. Time Frame A time frame from which both methodologies will be structured is 15 years. This is a le ngthy p eriod of time , however the r esearcher f eels jus tified in setting thi s tim e frame as the incidence of serial crime is so rare. This also allows for the researcher to gain a sufficient sample for both the interviews and case studies. 1. Case Studies Case studies of serial crimes in Australia will be conducted. These case studies may provide further evidence i n relation t o the i ssues s urrounding i nvestigative t echniques and policies, as m any bi ographical a ccounts of s erial cr imes, detail f actors r elating t o the investigation. By conducting such a comparison between the investigators opinion of such practices and the recording of such practices by a third person, the accuracy and applicability of such opinions can be determined. Such information can also aid in the development of i nterview que stions a s w ell a s f urthering t he f indings f rom s uch 356 interviews. Many issues may not be apparent in these accounts and thus the interviews will enable a richer, more accurate picture of the issues. Similar s tudies in A merica w hich c onsidered t he ps ychological a nd of fender characteristics of s erial murder ha ve al so utilised case s tudy research. Giannangelo (1996) utilised case studies of American serial killers to determine what psychological characteristics w ere apparent. Whilst E gger’s (1998) use of c ase s tudies e nabled hi s development of unde rstanding a range of i ssues r elated t o s erial m urder i ncluding offender characteristics. Sources of Information In order to conduct case study research, a variety of materials will be drawn on t o provide de tails o f the se crimes. W hen conducting a do cument a nalysis, t here a re t wo forms of evidence, being documents and records (Hodder, 2003). Records ar e of ficial texts, whereas documents are texts which are formed for personal use. Thus the records that will be analysed, de pending upon a vailability, i n t his s tudy are coroner’s reports, closed pol ice f iles a nd, a nd t he doc uments w ill include bi ographical t exts, ne wspaper articles, r adio t ranscripts, t elevision doc umentaries a nd pr evious c ase s tudy r esearch, depending upon the availability of such resources. It was hoped that additional sources in the form of completed police records of serial crime would be available. This would provide accurate information and allow for a greater analysis of investigative issues. However t he us e of s econdary s ources s uch a s t he doc uments out lined a bove ha s been accepted as app ropriate m aterials f or c ase s tudy r esearch. In pa rticular, Giannangelo (1996) and K eppel a nd Birnies (2003) have pr eviously accepted t his utilisation of secondary materials in studies of serial crime. Sample It i s ai med that at l east 4 cases of s erial m urder and 2 cases of s erial r ape be considered for t his c ase s tudy r esearch. However t his w ill be de pendant upon t he availability of closed case f iles f rom N SW P olice. If a ccess t o individual cl osed case files is not available through NSW Police then the cases would be chosen based upon 357 the amount of secondary sources available. As it is aimed that incidents of serial rape will a lso be i ncluded i n this c ase s tudy r esearch to gain a f uller unde rstanding of t he investigation of s erial c rimes in general, the una vailability o f c losed case f iles w ill hinder t he ability t o adequately consider t his crime. This would be due t o the l ack of available secondary sources o f i nformation, effective cas e s tudy analysis may not be possible. The num ber of case s tudies c hosen ha s be en s upported b y p revious s tudies of serial crime, which aimed to identify the impact of specific offender characteristics on issues of interest to those researchers (Egger, 1984; Giannangelo, 1996). 2. Interviews As i nvestigation techniques and practice ar e of i nterest, it w as cons idered as essential to be able to interview members of the police force who have worked on serial crime cases. Interviewing investigators of serious crimes has been utilised in numerous studies. The study by Mouzos and Muller (2001) which aimed to identify factors which impacted the a bility of i nvestigators to solve mu rder c ases ut ilised this te chnique, in addition to using secondary resources. Many issues may not be apparent in these case study accounts and thus the interviews will enable a richer, more accurate picture of the issues. Data from interviews may also identify lacuna arising from the case studies. By conducting such a comparison between the investigator’s opinion of such practices and the recording of such practices by a third person, the accuracy and applicability of such opinions can be determined. Consent Prior to the interviews participants will receive a consent form and Information Statement de tailing th e aims of the research. P articipants ma y w ithdraw a t a ny time . Participants w ill be given ample time to read the inf ormation statement, ask the researcher any questions they may have and consider the consent form. Interviews will be c onducted at a t ime a nd pl ace c onvenient a nd s uitable f or t he purpose. T he anonymity of p articipants w ill be a ssured. Information c ollected f rom t he i nterviews 358 will be c oded a nd de -identified pr ior t o t hematic a nalysis. C onfidentiality w ill be maintained at all times. Sample Participants w ill be d rawn from th e r esearch population employed by NSW Police currently or previously as identified by NSW Police who have contributed to the investigation of serial crime. Forensic technicians employed by the NSW Police will also be recruited as participants in this research. If access to closed case files is granted this would allow for a more efficient way of determining which people could be targeted as possible participants. This would be gained by utilising the case files to determine who contributed to a specific investigation thus identifying a specific sample group. The num ber of p articipants w ill be de pendant u pon t he a vailability of s uitable participants. However it is anticipated that interviews be obtained from as many suitable participants a s pos sible ( for e xample c hief i nvestigator, s enior f orensic t echnician). A s this research is aimed at collating data that will generate a saturation of qualitative data rather than attempting to create any sort of statistical or quantitative outcomes, it is not the intention to interview large numbers of individuals. Interview Method The interviews will be semi-structured in nature, incorporating both closed and open que stions. T his form of i nterview s tructure a llows f or f urther pr obing o f i ssues that arise during t he i nterview process, when response i s unclear and/or needs further explanation. The interviews will each be conducted over a one-hour period, and will be face-to-face unl ess circumstances doe s not a llow f or t his f ormat. W hen f ace-to-face interviews are not feasible, due t o financial or ot her i ssues, t hen t elephone i nterviews will be c onducted. T he i nterviews w ill be r ecorded us ing a n a udio r ecorder a nd transcribed at a later date. 359 Analysis In-depth s emi-structured i nterviews, w hich w ill pr ovide pr imary da ta t o compare with the results obtained from the first stage of research. Both the case study information and interviews will be subject to a thematic analysis the findings of which will be compared to the information on i nvestigative techniques and practices and data gathered from t he c ase s tudies. A com puter pr ogram s pecifically d esigned to analyse qualitative data such as QSR NVivo will be used where appropriate. Confidentiality and Privacy As it is m y int ention to discover f rom N SW P olice if a ccess to closed/completed police case f iles w ill be available t o contribute t o the da ta gathered for the case study component of this study, issues relating to confidentiality and privacy will arise. In accordance with the University of Western S ydney Ethics Approval, the researcher would like to emphasise that the information that will be sought from these closed case files w ould be s pecific t o investigative pr actices onl y and conf idential information or inf ormation of c urrent op erational int erest w ill b e tr eated with the greatest r espect and c onfidentiality. T he i nclusion of c onfidential i nformation or information of current operational i nterest w ould not be nefit o r c ontribute t o t he research in any way. Only information pertaining to the use of investigative practices, techniques and technologies is of interest to this research. It is understood that access to this inf ormation will be conditional t o pe rmission be ing granted b y N SW P olice and conformity to any guidelines or protocol that they establish to ensure the sensitivity and privacy of the data is respected. Participants tha t ha ve ta ken part in semi-structured interviews w ill be a ssured that all information gathered will be treated in strictest confidence and stored securely. The r espondents w ill be m ade a ware t hat s hould t hey de cide, a t a ny s tage o f t he research, that they now longer wish to participate then they can withdraw. Respondents will be made aware that no data will be reported that compromises their anonymity. All 360 documents containing confidential information related to this research will be stored in a secure locked location. Only the researcher will have access to this information. 361 REFERENCES CITED Adderly, R. W., & Musgrove, P. (2001). Police crime recording and investigation systems: A user's view. Policing: An International Journal of Police Strategies and Management, 24(1), 100-114 BBC. (2003). Green River killer confession. Retrieved 06/11/03, from http://news.bbc.co.uk/1/hi/world/americas/3245301.stm Crimes Act 1900 (NSW), 61I (1900) Egger, S. A. (1984). A Working Definition of Serial Murder and the Reduction of Linkage Blindness. Journal of Police Science and Administration, 12(3), 348357 Egger, S. A. (1998). The Killers Amoung Us: An Examination of Serial Murder and Its Investigation. New Jersey: Prentice-Hall Fox, J. A., & Levin, J. (1998). Multiple Homicide: Patterns of Serial and Mass Murder. Crime and Justice, 23, 407-455 Giannangelo, S. J. (1996). The Psychopathology of Serial Murder: A Theory of Violence. London: Praeger Publishers Hickey, E. W. (1997). Serial Murderers and Their Victims (2nd ed.). Belmont: Wadsworth Publishing Company Hodder, I. (2003). The Interpretation of Documents and Material Culture. In N. K. Denzin & Y. S. Lincoln (Eds.), Collecting and Interpreting Qualitative Materials (2 ed., pp. 155-175). Thousand Oaks: Sage Publications Holmes, R. M., & Holmes, S. T. (1998). Serial Murder (2 ed.). Thousand Oaks: Sage Publications Keppel, R. D., & Birnes, W. J. (2003). The Psychology of Serial Killer Investigations: The Grisly Business Unit. Sydney: Academic Press Killmier, B. (1997). Tracking Violent Crime. Paper presented at the Second national outlook symposium : violent crime, property crime and public policy, Canberra Kocsis, R. N. (1999). Psychological Profiling Techniques for Australian Serial Murder Investigations. Victoria Police Association Journal, 65(10), 22-23,25 Kocsis, R. N., Cooksey, R. W., & Irwin, H. J. (2002). Psychological Profiling of Offender Characteristics From Crime Behaviors in Serial Rape Offences. International Journal of Offender Therapy and Comparative Criminology, 46(2), 144-169 Mouzos, J. (2000). Homicidal Encounters: A Study of Homicide in Australia 19891999. Canberra: Australian Institute of Criminology Mouzos, J. (2001). Investigating Homicide: New Responses For An Old Crime. Paper presented at the 4th National Outlook Synopsium on Crime in Australia- New Crimes or New Responses, Canberra Mouzos, J., & Muller, D. A. (2001). Solvability Factors of Homicide in Australia: An Exploratory Analysis. Trends and Issues in Crime and Criminal Justice, 216 362 Neuman, W. L. (2003). Social Research Methods: Qualitative and Quantitative Approaches (5th ed.). Boston: Allyn and Bacon Turvey, B. (2002). Criminal Profiling: An Introduction to Behavioral Evidence Analysis. San Diego: Academic Press 363 Research Approval- New South Wales Police --- Received from NSWP.DEVE1CHR 82813 13/09/05 10:41 ------ Dear Ms Scerra The NSW Police agrees to allow you to commence data collection for your PhD project on serial crime. Please remember that your ethics committee approval requires strict confidentiality in regard to information you become aware of in relation to any victim, suspect, witness or police officer, in the course of your work. In due course the NSW police will require, as a condition of the grant of this approval, receipt of a final version of your Thesis, along with your agreement that we are able to place the thesis in the NSW Police library, and use the information therein as we see fit, but, of course, with due respect to your moral rights as author. This will all be detailed in the formal research agreement. We will also appreciate receipt of a brief executive summary of the work, containing any observations or recommendations that you think we should consider. While the NSW Police is happy to receive recommendations, we remind you that we are under no obligation to implement them. I am organising a contact person within the Behavioural Crime Unit, who will be able to coordinate your initial contact and data collection. I hope to be able to e-mail you later today with that information. Please let me know if there is any thing else I can do to assist you. Yours Faithfully Dr Chris Devery Principal Research Officer Leadership Development Directorate 364