Файл: [D._Kim_Rossmo]_Geographic_Profiling(BookFi.org).pdf

© 2000 by CRC Press LLC

watched (see also Cromwell et al., 1990, 1991). While this search process is
more casual than determined, it is surprisingly effective, allowing offenders
to build up a mental “card file” of potential targets. When a burglar wants
to commit a break in, the various possibilities are reviewed and a decision
made. If it turns out that a particular selection is unsuitable at the time, then
the next target in the “card file” is chosen. This search process provides insight
to the target selection of those serial rapists who break into homes, many of
whom have previous burglary experience (see Schlesinger & Revitch, 1999).

In a study using routine activity and lifestyle/exposure perspectives, Warr

(1988) found that homes — and women — at greater risk for burglary are
also at greater risk for inside rape. A comparison of 1980 UCR Index offences
for 155 Standard Metropolitan Statistical Areas (SMSA) revealed rape and
burglary to be strongly correlated (

r

 = 0.79). Both crimes also showed close

and consistent similarities in their coefficients with a set of opportunity vari-
ables. These variables were grouped into three categories: (1) housing (e.g.,
low income, newer areas, rental premises, multiunit structures, etc.); (2)
female occupancy (e.g., divorced, living alone or without other adults, etc.);
and (3) combinations of the first two groups. For example, one of the strongest
combination variables was “female householders, no husband present, aged
25–34, in a renter-occupied structure with more than 50 units” (rape, 

r

  =

0.59; burglary, 

r

 = 0.54,). A summary index of similarity was also very high

(

r

2

 = 0.99). Warr concluded that home-intrusion rape (rape following an

unlawful entry of the home) is a hybrid offence — it is a violent crime with
the opportunity structure of a property crime (see also Felson & Clarke, 1998).

Police interviews with serial rapists have revealed a high awareness of

environmental cues. One offender stated he selected homes by looking for
driveways with oil spots, suggesting the absence of a vehicle. If he then saw
a woman inside the house he presumed the man was absent. Another rapist
said he avoided homes with boats parked outside as that indicated the pres-
ence of a male. A surprising number of victims leave their curtains open at
night, even after being attacked. Even a cursory look inside a home can reveal
much about who lives there, be they female, male, a couple, the elderly, or
children. Some offenders demonstrate prior knowledge of their victims,
probably gained through past surveillance, although others have been known
to search through a victim’s purse during the attack and then pretend to have
possessed such information beforehand.

The search process described by Wright and Decker can also explain the

clustering of rape sites and the occurrence of two or more attacks in a single
evening. It is not surprising that targets with suitable offence characteristics
are often situated close together. If an attack fails, a criminal simply has to
go to the next location in their mental “card file” and try again, as in the case
of the rapist with multiple preselected victims, described earlier. Because the

© 2000 by CRC Press LLC

contents of this file are strongly determined by the offender’s activity space,
the crime locations may provide a “road map” with significant investigative
implications.

8.2.2 Hunting Humans

Anatoly Onupriyenko, convicted of murdering 52 people including families
with children in Zhitomir, Ukraine, referred to himself as a hunter and his
victims as the game. “ ‘I look at it very simply. As an animal. I watched all
this as an animal would stare at a sheep’ ” (Shargorodsky, 1998).

The hunt for humans is much like the search for other criminal targets.

But there are two complicating factors for the offender: first, people move
about, and second, they have to be controlled. Criminal predators must
employ search and attack methods that address these issues. Most previous
classifications of serial murder and rape geometry have been limited to
description, ignoring the processes that underlie the outcomes. One of the
few exceptions is Petrucci’s (1997b, 1998) empirical typology of victim-
acquisition techniques used by serial sexual killers. Offenders from the U.S.
(n = 146) were selected from Hickey’s comprehensive serial murder database
and coded on the basis of a six-item protocol. Petrucci found the following
distinct groupings:

1. Abduction — The victim was attacked by surprise, and then trans-

ported to a different location where the murder occurred. This was
the most common victim-acquisition technique (36%), and is sugges-
tive of a fantasy-motivated offender. Of all serial murderers using this
method, 67% were local, 94% were Caucasian, and 31% were disor-
ganized.

2. Attacking — The victim was surprise attacked and then killed, with

no transportation. This was the least common technique (16%), and
is suggestive of an offender who is less concerned with apprehension.
Of all serial murderers using this method, 71% were local and 43%
were a minority.

3. Luring — The victim was first attracted or lured by deception, then

attacked by surprise and killed. Voluntary or involuntary transporta-
tion may be involved. This was the second most common victim
acquisition technique (25%), and is suggestive of a fantasy-motivated
offender. Of all serial murderers using this method, 41% were local
and 41% were travellers.

4. Combination — Victims were acquired through the use of multiple

tactics. This was the third most common victim acquisition technique
(23%), and is suggestive of an instrumental rather than a fantasy-

© 2000 by CRC Press LLC

driven offender. Of all serial murderers using a combination of meth-
ods, 52% were travellers and 72% were organized.

Foraging theory has been applied to the understanding of criminal pred-

ators (Canter & Hodge, 1997). While it may suggest useful insights to the
basics of hunting behaviour (Daly & Wilson, 1995), this perspective does not
provide an adequate theoretical basis. Animals must eat to survive, and every
hunt expends limited energy. Optimal foraging theory is based on a balance
between opportunity maximization and effort minimization. But none of
this is applicable to the criminal hunt.

Hunting method affects the spatial distribution of offence sites and any

effort to predict offender residence from crime locations must consider this
influence. It is therefore necessary to employ a hunting typology relevant to
the production of spatial patterns of serial predators. The construction of
the scheme now used as the standard in geographic profiling was informed
by geography of crime theory, empirical data, and investigative experience.

While murder or rape can potentially involve several different types of

crime locations, experience has shown that not all sites may be known to
police. Victim encounter or last known location (usually a close proxy of
encounter site) are often known in murder cases and always known in rape
cases. Body dump sites are known in most murder cases and victim release
sites in all rape cases, but if the murder act itself occurred in a different
location, this site will likely be known only to the offender. A rape site may
or may not be known to the victim, and hence the police. The hunting
typology is therefore concerned with offender behaviour 

vis-à-vis

 the crime

locations most probably known to police. Arson is simpler, involving station-
ary known targets and therefore only a single crime location. Consequently,
only the first three search techniques, and none of the attack methods in the
hunting typology discussed below, apply to serial arson.

8.2.3 Search and Attack

The hunting process can be broken down into two components: (1) the search
for a suitable victim; and (2) the method of attack. The former influences
selection of victim encounter sites, and the latter, body dump or victim release
sites. The hunting typology results from a combination of the search and
attack elements.

The following four victim search methods were isolated:

1. Hunter — An offender who sets out specifically to search for a victim,

basing the search from his or her residence.

2. Poacher — An offender who sets out specifically to search for a victim,

basing the search from an activity site other than his or her residence,

© 2000 by CRC Press LLC

or who commutes or travels to another city during the victim search
process.

3. Troller — An offender who, while involved in other, nonpredatory

activities, opportunistically encounters a victim.

4. Trapper — An offender who has an occupation or position where

potential victims come to him or her (e.g., nursing) or by means of
subterfuge, entices victims into a home or other location they control
(e.g., by placing want ads).

The following three victim attack methods were isolated:

1. Raptor — An offender who attacks a victim upon encounter.
2. Stalker — An offender who first follows a victim upon encounter, and

then attacks.

3. Ambusher — An offender who attacks a victim after he or she has

been enticed to a location, such as a residence or workplace, controlled
by the offender.

Hunters are those killers who specifically set out from their residence

to look for victims, searching through areas in their awareness space they
believe contain suitable targets. This is the most commonly used method
of criminal predators. Westley Allan Dodd, a serial killer executed for the
murder of three children in the state of Washington, wrote in his diary,
“Now ready for my second day of the hunt ... Will start at about 10 a.m.
and take a lunch so I don’t have to return home.” He was worried, however,
that if he murdered a child in the park through which he was searching,
he would lose his “hunting ground for up to two to three months” (Westfall,
1992, p. 59). The crimes of a hunter are generally confined to the offender’s
city of residence. Conversely, poachers travel outside of their home city, or
operate from an activity site other than their residence, in the search for
targets. The differentiation between these two types is sometimes an intri-
cate task, requiring a subjective interpretation of crime location familiarity.
The hunter and poacher categories are similar, though not identical, to the
“marauder” and “commuter” designations (Canter & Larkin, 1993),
described and discussed later.

Trollers are opportunistic killers who do not specifically search for vic-

tims, but rather encounter them during the course of other, usually routine
activities (see Eck & Weisburd, 1995b). Their crimes are often spontaneous,
but many serial killers have fantasized and planned their crimes in advance
so that they are ready and prepared when an opportunity presents itself. This
has been termed premeditated opportunism, and is related to the concept of
pattern planning (Cornish & Clarke, 1986a). In an interview, Eric Hickey

© 2000 by CRC Press LLC

(1996), an incarcerated serial murderer responsible for 12 victims, discussed
the details of one of his crimes. While it was an opportunistic attack, his
assessment of the environment, victim approach, and con to move them to
a more secluded area illustrate the killer’s rationality and adaptation.

It was the first time I ever abducted two at once, not the first time that I
murdered two in the same day. I saw them out walking across a field. I was
wandering. I was kind of in a controlled frenzy. I was certainly aware of
what I was doing, in control, but inside I was desperate, and I would not
have taken them had I not been there, anywhere else but there, I would have
let them go. There was no reason. It was a cold day, no one was around.
There was a secluded area nearby. In other words, it was a killing site, and
I was in a very remote part of town. There were houses there but there was
also fields off to one side. I had no vehicle there. (pp. 124–125)

Trappers have an occupation or position, such as a nurse or orderly in

a hospital, where potential victims come to them. They also entice victims
into their home or other location they control by means of subterfuge. This
may be done through entertaining suitors, placing want ads, or taking in
boarders. Black widows, “angels of death,” and custodial killers are all forms
of trappers, and most female serial murderers fall into this category (Hickey,
1986; Pearson, 1994, 1997; Scott, 1992; Segrave, 1992).

Raptors, upon encountering a victim, attack almost immediately. This is

the most common method used by criminal predators. Stalkers follow and
watch their targets, moving into the victim’s activity space, waiting for an
opportune moment to strike. The attack, murder, and body dump sites of
stalkers are thus strongly influenced by the victims’ activity spaces. Jon Berry
Simonis, the Ski Mask Rapist, attacked women in Florida, Georgia, North
Carolina, Ohio, Michigan, Wisconsin, Mississippi, Louisiana, Texas, Okla-
homa, and California from 1978 to 1981, becoming progressively more vio-
lent before he was arrested by the Louisiana State Police. Simon sometimes
stalked his victims, and through his work at a hospital had access to victims’
medical records including their address, marital, and work details (Michaud
& Hazelwood, 1998).

Ambushers attack victims they have brought or drawn into their “web”

— someplace where the killer has a great deal of control, most often their
home or workplace. The bodies are usually hidden somewhere on the
offender’s property.

47

 While victim encounter sites in such cases may provide

sufficient spatial information for analysis, many ambushers select marginal-

47 

Theoretically this does not have to always be so, but empirically it appears to be the

rule. This probably results from the fact that ambushers are often also trappers, and the
latter rarely exhibit significant mobility.