Chapter 1

Forensic Archaeology: Integrating

Archaeology with Criminalistics
and Criminology

W. J. Mike Groen

Abstract This chapter suggests a forensic archaeological framework that inte-

grates the archaeological process and research cycle with criminalistic and crimino-
logical knowledge. By doing so, it proposes evidential reasoning that is based on the
Bayesian inference, physical evidence recovered at crime scenes, middle-range
theories, site formation processes, forensic assemblages and mutually exclusive
hypotheses. This to analyse, interpret and reconstruct the human (criminal) behav-
iour and natural processes that have formed and modified the investigated crime
scene. The analysis and interpretation of the recovered forensic record should com-
mence during field work, at the crime scene. After completing this investigation, the
secured physical evidence, as analysed and interpreted by different fields of forensic
expertise, should be integrated into one overall synthesis. In addition, a forensic
archaeological investigation should not be an end in itself. It should also be used to
generate inductive and deductive theories and models needed for reconstructing the
human (criminal) behaviour and natural processes responsible for the formation,
distortion and destruction of the physical evidence. To be able to do this, feedback
from casework is essential, as are development and innovation. To strengthen its
base as a scientific discipline, the chapter closes with a discussion on innovation and
development within the field of forensic archaeology.

Keywords Forensic archaeology · Archaeological research cycle · Criminalistics ·

Criminology · Bayesian inference

W. J. M. Groen (*)
Netherlands Forensic Institute, The Hague, The Netherlands
Faculty of Archaeology, Leiden University, Leiden, The Netherlands
e-mail: [email protected]

© Springer International Publishing AG, part of Springer Nature 2018 1

P. M. Barone, W. J. M. Groen (eds.), Multidisciplinary Approaches to Forensic
Archaeology, Soil Forensics, https://doi.org/10.1007/978-3-319-94397-8_1
2 W. J. M. Groen

1.1 Introduction

Forensic archaeology is habitually defined as the use of archaeological methods and

techniques in a legal setting (e.g. Groen et al. 2015). Accordingly, forensic archaeo-
logical literature has a tendency to focus on the application of archaeological field
methodology in forensic and/or medicolegal settings. The archaeological main-
stream discourse on theory during the past decades (e.g. Bentley et al. 2009; Johnson
2010; Harris and Cipolla 2017) seems to have been mostly ignored within forensic
archaeology although there are noteworthy exceptions (e.g. Crossland 2013; Gupta
2013; Fondebrider 2016). In addition, most forensic archaeological advancements
seem to be methodical in nature, focusing, for example, on geographic information
systems (GIS), remote sensing and geophysics. The study of taphonomic changes
seems to be an exception to the above-mentioned. This focus on (geo- and medical)
sciences could be problematic in the long run, since one of the outcomes of the
archaeological discourse was that the analysis and interpretation of the archaeologi-
cal record, needed to reconstruct past human activities, cannot be based on method-
ological research alone (e.g. Cleland 2002, 2011; Jeffares 2008). To link the
empirical observations to inductive or deductive theories and models that are able to
clarify the studied past human activities, archaeological reconstructions should
combine empirical observations with an interpretative framework. In this way
archaeology integrates concepts from the humanities (e.g. history), the social sci-
ences (e.g. economics, cultural anthropology) and the sciences (e.g. biology, earth
sciences) into its paradigm and is thus research data- and theory-driven. Since
forensic archaeology is a form of (applied) archaeology, this framework should also
be integrated with forensic archaeology.
Although the usability of the archaeological framework in crime scene investiga-
tion has already been discussed (Groen and Berger 2017), it has not – to author’s
knowledge – been successfully incorporated in forensic archaeological casework, as
related to police cases (e.g. searches, excavations, exhumations and field recover-
ies). This chapter focuses on the applicability of the archaeological framework and
research cycle to forensic archaeological cases combined with theory and method-
ologies from criminalistics and criminology. By doing so, it builds on the probabi-
listic framework as proposed by the author and colleagues in the concluding remarks
of Groen et al. (2015). The chapter itself is divided into four parts. Part one briefly
discusses the key concepts in archaeology, criminalistics and criminology. Part two
introduces the archaeological research cycle and utilises this cycle to structure the
forensic archaeological investigation. Part three focuses on development and inno-
vation needed to aid forensic archaeology in improving its processes, probabilistic
framework and investigative methodology. In conclusion part four summarises the
main points of the chapter.
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 3

1.2 Archaeology, Criminalistics and Criminology

Archaeology can be defined as the study and reconstruction of past human activities
and related natural processes, through a systematic recovery, analysis, interpreta-
tion and reconstruction of the archaeological record. The archaeological record
covers human remains, material culture (artefacts and ecofacts), non-portable
archaeological features and environmental information. The archaeological record
is always recorded in situ and in context or in its original position and association.
Archaeological field methodology is research question driven and developed to sys-
tematically search for and reliably document the archaeological record (e.g. Hester
et al. 2008). Modern field recording systems combine spatial databases with 3D
recording strategies (e.g. Smith et al. 2015) and unmanned aerial vehicles (UAVs)
with structure from motion (SfM) photogrammetry (e.g. Howland et al. 2015) and
are fully integrated with GIS applications. Essential for understanding the archaeo-
logical record is its context, association and the site formation process or all human
and non-human (natural) processes that have formed and modified an archaeologi-
cal site (e.g. Clarke 1973; Schiffer 1976, 1983, 1987). Site formation processes can
be subdivided into non-human natural processes (n-transforms) and human natural
processes (c-transforms). Since site formation processes could also be additive,
they can introduce new features into the archaeological record. An archaeological
reconstruction of the past must therefore not only focus on the retrieved archaeo-
logical record but must also take into account the site formation processes that
(could) have modified, destroyed or masked the record that was originally depos-
ited at the research site. Since archaeologists reason retrospectively when studying
past societies, they use background knowledge and an interpretative framework to
link the obtained archaeological record to inductive and deductive theories and
models about the past.1 Archaeological background knowledge is mostly obtained
from (spatial) databases, theoretical models, material studies and actualistic experi-
ments.2 Archaeological databases are used to structure archaeological research, for
example, prior to the excavation or recovery (as background knowledge), during the
excavation or recovery (as recording strategy) and after the field work (as a recon-
struction tool). Archaeological theoretical models are often based on middle-range
theories, or MR-theories (Binford 1977, 1981; Schiffer 1976, 1988). The
MR-theories are a set of testable (and falsifiable) models or propositions about
human activity that link the spatio-temporal regularities in the archaeological
record (the empirical domain) to the more abstract high-level conceptual schemes

1
Archaeologists use different theoretical frameworks to analyse and reconstruct the past, depend-
ing on their epistemological and ontological viewpoints (e.g. Bentley et al. 2009; Johnson 2010;
Harris and Cipolla 2017).
2
For inferences about archaeological reasoning, see, for example, Sullivan (1978), Gibbon (1989),
Wylie (2002, 2011), Alberti et al. (2013) or Chapman and Wylie (2015, 2016).
4 W. J. M. Groen

or theories about human behaviour (the theoretical domain).3 Archaeological

MR-theories could be constructed from actualistic experimental studies, historical
sources, taphonomic research or ethnoarchaeological observations. Other
MR-theories are abstracted from present day economic, sociologic or cultural
anthropological theories and models. The above-mentioned means that an archaeo-
logical interpretation is not only data-driven but also theory-driven. In other words,
the archaeological collection, analysis and interpretation of data are subjected to a
selection process and are therefore interpretation driven. There is no such thing as
objective, theory-­free data collection or sampling.
Criminalistics can be defined as the scientific study concerning the recovery,
interpretation, evaluation and reporting of physical evidence obtained in criminal
cases, including the role of the forensic scientists (e.g. Saferstein 2015). For the
prediction of the forensic value of the obtained physical evidence, Locard’s exchange
principle4 and the principle of the ‘evidentiary triangle’ are used. This is in order to
predict which item or trace could be linked to the victim, the suspect and the crime.
Items or traces that can be linked to all three are, generally speaking, forensically
most relevant since they can possibly be used to identify the (unknown) offender. As
a consequence, criminalistic methodology focusses on the possible transfer, or lack
thereof, and degradation of crime-related physical evidence. Criminalistic reason-
ing is based on the Bayesian approach, meaning that the evidential value of a foren-
sic trace (to be usable in court) not only depends on the trace itself, the context of
the criminal case and the location where this trace has been found but also on the
hypotheses used to evaluate this trace. A hypothesis could be seen as the translation
of a relevant forensic question into a realistically possible answer. The criminalistic
evaluative framework is therefore probabilistic in nature and works with at least two
mutually exclusive hypotheses (or a pair of propositions) and with comparison as
the method of examination (e.g. Cook et al. 1998a, b; Evett et al. 2000; Evett 2015;
Robertson et al. 2016). A suitable hypothesis that does well for explaining the
observed results can thus be integrated into an overall police scenario concerning
what (could have) happened at the crime scene. Since the prior odds of these hypoth-
eses are considered to be the task of the police, or the court, the aim is to provide
evidence to support, or falsify, one (or more) of the hypotheses by considering the
probability of the evidence. The aim is not to test the probability of the hypotheses
themselves.5 Hypotheses, inferences and conclusions concerning the (transfer of)
physical evidence can be investigated at the source level, at the activity level and at

3
For critical discussion on the usability of MR-theories in archaeology, see, for example, Schiffer
(1976, 1988), Binford (1977, 1981), Raab and Goodyear (1984), Pierce (1989), Wylie (1989),
Salmon (1992), Maschner (1996), Tschauner (1996), Shott (1998) or Smith (2011).
4
Locard’s exchange principle states that the perpetrator of a crime could leave a physical trace on
a crime scene, or bring a trace into it and that this trace could be used as forensic evidence (Locard
1920, 1930).
5
For more information concerning the way (European) forensic laboratories value evidence and
report conclusions, see the ENFSI ‘Guideline for Evaluative Reporting in Forensic Science’
(ENFSI 2015).
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 5

the level of the offence (e.g. Cook et al. 1998a; Evett et al. 2000).6 In view of the
aforementioned, forensic archaeological reasoning ought to comply with criminal-
istic reasoning and should therefore follow the Bayesian approach when evaluating
physical evidence. In addition, forensic archaeological conclusions should be, if
possible, formed at the activity level, since an explanation at the source level alone
will not tell us how the retrieved evidence came to be deposited at the crime scene.
Criminology is the scientific study of crime, criminality, criminals, criminal
behaviour and the criminal justice and penal systems (e.g. Frailing and Harper
2015). Of importance here is that criminology also studies the causality of crime
and criminal behaviour and that GIS applications had been used to map and analyse
crime distribution and patterns (e.g. Paulsen and Robinson 2008). For example,
environmental criminology studies crime patterns within different environments
and architectural features and the impact of these patterns on humans (e.g. Bruinsma
and Johnson 2018). Another criminological discipline, offender profiling, utilises
information on geographical locations (e.g. the location where the victim lived and
the location where the victim was last seen), the spatial layout of crime scenes, the
type of crime, the psychological theories concerning the offender and the personal
history of the victim to predict the characteristics and place of residence of the
offender in order to apprehend this individual (e.g. Rossmo 2000; Ainsworth 2001;
Bartol and Bartol 2012). A number of offender profiling studies include locations of
body deposition sites – as related to the type of crime, the type of victim and the
type of offender – in their research framework7 and are therefore highly relevant to
forensic archaeology. This since such studies can be utilised to build MR-theories
about where offenders hide their victims and can provide information that could
help structure forensic archaeological search strategies involving missing persons.
Moreover, criminological studies can offer (emic) interpretations about criminal
behaviour and the spatial layout of physical evidence at crime scenes, as offered by
the offenders themselves. In doing so, criminology can provide data and informa-
tion needed to build additional MR-theories that could improve our understanding
of the retrieved physical evidence. A careful approach is needed when analysing
offenders’ testimonies since they often do not give an entirely honest account of
past events. However, given the fact that most of the academically trained forensic
archaeologists are also accustomed to work within the social sciences framework,
these limitations are not unfamiliar to forensic archaeologists.

6
The source level relates to the identification or classification of the physical evidence in relation
to its possible source. The activity level evaluates the relationship between the transferred material
and the human action that possibly resulted in this transfer. The offence level focuses on the rela-
tionship between the transferred material and the offence.
7
See, for example, Rossmo (2000), Lundrigan and Canter (2001a, b), Morton and Lord (2002),
Nethery (2004), Snook et al. (2005), Powell (2006), Häkkänen et al. (2007), Van Patten and
Delhauer (2007), Hays (2008) or Anderson (2009).
6 W. J. M. Groen

1.3 The Archaeological Research Cycle

Given the above-mentioned, a forensic archaeological site should be seen as an

accumulation of materialized patterns of past human and non-human activities,
often distorted. Additionally, we should combine our methodology with criminalis-
tics and attempt to develop a probabilistic framework – based on the Bayesian infer-
ence – to reconstruct the human activities and natural processes that took place at
the crime scene (prior to, during and after the crime has been committed). To be able
to do this, we also need to build forensically relevant MR-theories that are able to
link the retrieved forensic record (the sum of all archaeologically retrieved physical
evidence) with inductive and deductive theories and models about the behaviour
that initially formed this record. Additionally, we need to restructure forensic
archaeological investigations in accordance with the archaeological research cycle.
This cycle covers (after Swartz 1967) a sequence of nine steps: (1) preliminary
desktop research, (2) field survey, (3) final desktop research, (4) excavation or
recovery, (5) data analysis, (6) interpretation, (7) integration, (8) comparison and (9)
abstraction. The time frame of each step varies per archaeological site since some
excavation campaigns take place annually and can continue for decades. However,
most research into archaeological sites will complete this cycle within a year or so.
The sequence of steps in the cycle is not set in stone; it is possible to retreat one or
more steps, for example, when unexpected physical evidence is located or new
questions are put forward. Since archaeologists initiate data analysis and interpreta-
tion during fieldwork, some of these steps are, also, interwoven with one another,
especially steps 4 (excavation or recovery), 5 (data analysis) and 6 (interpretation).
The remaining part of this section attempts to structure a forensic archaeological
investigation according to the nine archaeological research cycle steps, starting with
preliminary desktop research and ending with abstraction of the obtained forensic
record.8

1.3.1 Preliminary Desktop Research

The preliminary desktop research (or desk-based assessment (DBA)) is the back-
bone of all archaeological research. By accessing information acquired from previ-
ous research (the background knowledge), archaeologists are able to express an
expectation of what an forthcoming excavation would bring, which deductive
research hypotheses can be addressed and which scientific questions could be
answered. These expectations also lead to the development of a preliminary excava-
tion or recovery strategy and field methodology.

8
Since the discussion of methods and techniques used during the crime scene investigation falls
outside the scope of this chapter, the reader is forward to Fisher and Fisher (2012), Gardner (2012)
or Houck et al. (2012). For discussion of forensic archaeological field methodology, see, for exam-
ple, Dupras et al. (2011), Hunter et al. (2013), Dirkmaat and Cabo (2016) or Evis (2016).
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 7

Integrating preliminary desktop research into forensic archaeological investiga-

tion would surely benefit such investigation. By assessing background knowledge,
for example, a GIS application with pedological, geomorphological and ecological
maps, the forensic archaeologist would be able to predict what the soil and ecologi-
cal conditions are at and near the to be investigated crime scene, or place of incident.
This information would enable the archaeologist to estimate which formation pro-
cesses and taphonomic transformations (e.g. taphonomic changes, soil and ecologi-
cal alterations and material degradation) could have taken place at this site and,
therefore, to predict which methods and techniques would offer the best change to
find a body, or object. The above-mentioned could be ‘easily’ done within a Bayesian
framework of reasoning, for example, by providing prior odds about the detectabil-
ity of a grave in different pedological, geomorphological and ecological contexts by
means of field walking, geophysics, remote sensing or cadaver dogs.
Additionally, background knowledge obtained from criminological MR-theories
concerning human behaviour, case review studies and experimental studies would
aid the development of deductive forensic hypotheses that should be addressed dur-
ing the investigation. Given the type of investigated crime, for example, a surface
scatter scene related to a rape and subsequent murder or a clandestine burial related
to a murder in a domestic abuse context, this offers a probabilistic framework to
reconstruct the activities that took place at the crime scene. Assessing background
knowledge, by means of archaeological predictive modelling (e.g. Verhagen and
Whitley 2012; Lieskovský et al. 2013), would also inform the forensic archaeologist
about the type of physical material that could be expected, where this material might
be located and what the evidentiary value of this material could be. Thus prior odds
concerning the spatial layout of the forensic record, the deductive hypotheses of
interest and the activities that were responsible for the spatial distribution of this
record at a given crime scene can be estimated. In this way, the preliminary desktop
research could be compared with the pre-assessment phase of the criminalistic
framework (e.g. Cook et al. 1999). The prior odds, if used during the crime scene
investigation phase, will optimise the investigative methodology and therefore the
chance of collecting the most relevant physical evidence (Groen and Berger 2017).
However, using background knowledge and ascribing prior odds before and during
the crime scene investigation could also lead to contextual biases (e.g. Ditrich 2015;
Stoel et al. 2014). In addition, confirmation bias (e.g. Kassin et al. 2013) should be
kept in mind to avoid a situation where only physical evidence that supports the
investigated deductive hypotheses is collected. Investigations should therefore also
collect evidence that could falsify the stated forensic hypotheses.
In goes without saying that introducing preliminary desktop research in forensic
archaeology will cause difficulties given the time restraints of a forensic case.
Forensic archaeologists working on police cases do not have the time to perform an
extensive preliminary desktop research, barring some cold cases. However, this
does not mean that we should omit this step in forensic archaeology. As discussed
in the development and innovation section, it all depends on configuring the back-
ground knowledge as indexed documents or spatial databases.
8 W. J. M. Groen

1.3.2 Field Survey

In forensic archaeology a field survey, if backed with the knowledge obtained from
the preliminary desktop research, will allow a limited testing of forensic hypotheses
and could be helpful for the improvement of the investigative strategy and method-
ology. Forensic archaeological field surveys generally consist of systematic surface
walkover, geophysics, remote sensing and/or subsurface testing in the vicinity of the
crime scene or place of incident, to examine local soil features, for example, by
probing, auguring or test pits. The available background knowledge and prior odds,
as obtained by means of the preliminary desktop research, should subsequently be
updated during and after the field survey. This should be done by inspecting the
crime scene, or place of incident, and by obtaining case-specific information during
a police briefing.
Additionally, a forensic archaeological field survey should be research and ques-
tion driven, integrating 3D mapping, GIS and database features into its field proce-
dures. During the survey, the crime scene should be assessed for the potential
availability of (different types of) forensic materials, given the type of crime inves-
tigated, the locally encountered conditions and the time frame of the case. In this
way, a field survey can be to a certain extent compared with the assessment phase of
the criminalistic framework (e.g. Cook et al. 1998b). Additionally, preliminary
information concerning the site formation processes and taphonomic transforma-
tions should be collected and processed. For example, to predict which types of
material evidence are, or could be, no longer present at the scene, for instance, due
to taphonomic changes or degradation. Therefore assessment will not only improve
the knowledge obtained from the preliminary desktop research but will also update
during this research obtained prior odds and therefore the investigative strategy,
forensic research questions and hypotheses that need to be considered during the
investigation. For example, the detectability of a grave will differ in different soil
conditions and be dependent on the taphonomic conservation of the to be located
human remains. Therefore a negative result given a search tool, for example, the
ground-penetrating radar (GPR), does not have to mean that the to be located body
is not present within the searched area. This result still will need to be evaluated
using the Bayesian inference. In other words, field surveys are not only important
for orientation and location of potential physical evidence. They are also important
for the updating of evidential value of the (to be expected) physical evidence and are
therefore also relevant in inductive hypotheses and scenario building.

1.3.3 Final Desktop Research

Given the above-mentioned, final desktop research should aim to formulate the final
forensic research questions and hypotheses and formulate the investigative strategy
and methodology, using the Bayesian framework of reasoning. This should ideally
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 9

be done back at the office, after finalising the field survey. However, this could also
be done at the crime scene, following the field survey and before engaging in the
excavation or recovery. It goes without saying that the formulated investigative
strategy and methodology should be in agreement with the to be answered forensic
questions. The employed methodology should be able to utilise theories and models
enabling the forensic archaeologist to recognise, analyse and interpret the forensic
record, not only after but also during the investigation of the crime scene. The inves-
tigative methodology should furthermore be case specific, be forensic hypothesis
driven, be able to optimise the allocation of search efforts and be able to maximise
the retrieval of evidence with the highest evidential value, as related to the formu-
lated hypotheses. The prediction of the type and size of the physical evidence that
could have been exchanged between the scene and the victim, between the victim
and perpetrator and between the scene and the perpetrator should also be conducted.
The formulated investigative methodology should additionally be adjusted if unex-
pected physical evidence is encountered or if the stated forensic hypotheses are
falsified. Finally, all methods employed will have to account for limitations that
could impact the retrieved forensic record, such as terrain and light conditions, the
size of the to be investigated crime scene and the available manpower, experience
and time.
Since the documentation, analysis and interpretation of the forensic record
should be centred around – in archaeological terms – assemblages or a combination
of different types of in situ physical evidence occurring in a defined space (e.g.
Lucas 2012), it is important to recognise the events and processes that have created
and subsequently affected the to be investigated crime scene. The formulated inves-
tigative methodology should therefore be able to document the encountered stratig-
raphy and contextual association in three dimensions and be able to recognise the
local site formation processes and taphonomic transformations.

1.3.4 Excavation or Recovery

A forensic excavation or recovery concentrates on the recognition, documentation,

collection and processing of the forensic record. The main goals, from the forensic
archaeological point of view, are to gather evidence that can be used to support or
falsify hypotheses, to build scenarios and to develop a crime scene formation model
that can help reconstruct the human activities and natural processes that took place
at the crime scene, prior to, during and after the crime has been committed (see next
sections). Each forensic archaeological investigation should therefore follow
archaeological and criminalistic principles.9 As discussed earlier; predictive models

9
Examples of archaeological principles are the principles of superposition, original horizontality
and lateral continuity. Examples of criminalistic principles are the Locard’s exchange principle and
the principle of the ‘evidentiary triangle’.
10 W. J. M. Groen

and MR-theories should be utilised to inform the forensic archaeologist as to where

new physical evidence could be expected not only based on what has been found so
far but also on the available background knowledge. In this way the investigation
will be hypothesis driven and will make use of abductive reasoning to proceed.
However, what type of physical evidence that will be collected and how this is done
(using standard operating procedures or SOPs) will depend on the specific national
protocols and jurisdictions and the limitation of the forensic laboratory tests (i.e.
false positives and false negatives).
It is a truism that the forensic value of material evidence studied at a forensic
laboratory is only as good as its recognition, documentation and recovery at the
crime scene. Since collection, analysis, interpretation and evaluation of physical
evidence goes hand in hand; evidence collection and evidence evaluation are inter-
dependent rather than independent. Collection and sampling of physical evidence
are therefore subject to a selection process, based on the theoretical and method-
ological aspects and the background knowledge available to the investigator. Each
forensic archaeological investigation should consequently be site plan led. Because
excavation and recovery destroys the spatial associations within an assemblage site,
plans should be created on site, preferably using objective documentation tech-
niques, for example, a total station, structure from motion (SfM) photogramme-
try, ortho-photogrammetry and/or 3D scanning. If not recorded properly, the
destructive nature of an archaeological investigation will lead to loss of evidence at
best, but it could also lead to ambiguous or erroneous interpretation of the obtained
forensic record. Consequently, mapping should not be seen as a documentation tool
only but also an assessment tool, especially if this assessment is performed within a
GIS application.10 Digital mapping of the forensic record will not only improve the
accuracy of documentation, interpretation, reconstruction and visualisation of this
record but will also save time and improve the chain of custody.

1.3.5 Data Analysis

To be able to reconstruct the sequence of human and natural processes that took
place at the archaeological site, data analysis in archaeology entails the classifica-
tion of the excavated or recovered archaeological record and the spatio-temporal
sequencing of this record into archaeological assemblages. Consequently, archae-
ologists perform a site formation analysis, by constructing and placing different
archaeological assemblages in a spatial and sequential framework, through classifi-
cation and contextual association, and by using theories and models that help
explain the presence of these assemblages at the archaeological site. This is done,
partly during the fieldwork, by means of the 3D site plans and within a GIS

10
If data and information obtained from individual sites are utilised to improve our background
knowledge (see abstraction), it is also necessary to documented all sites routinely and consistently
in the same manner, using digital methods and techniques.
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 11

application. This links the spatial locations of the mapped archaeological record to
attribute tables containing quantitative and qualitative data and information on this
record and the individual pieces of evidence that comprises this record.
It is believed that classification of the forensic record and the sequencing of this
record into forensic assemblages will help to support or exclude hypotheses, iden-
tify people present at the crime scene during the criminal act and offer preliminary
findings regarding the manner and cause of death. This will, therefore, establish
logical and defensible (preliminary) scenarios concerning what (could have) hap-
pened at the crime scene. Forensic assemblages can be constructed and analysed in
a similar manner as archaeological assemblages, by using a site plan and by assess-
ing the relevant background knowledge. Forensic archaeologists will thus be able to
build crime scene-specific site formation models and possibly sequence the physical
evidence to activities that took place before, during and after the crime was commit-
ted. This could be done by asking oneself how a grave was dug, where the spoil
from the grave was deposited, how the body was deposited in the grave, what type
of tools was used for the digging, what taphonomic alteration modified the crime
scene, etc.? Since it enables reasoning retrospectively, from the localised forensic
assemblages to the originally deposited forensic assemblages, thinking in formation
processes and forensic assemblages will help to understand how physical evidence
was originally deposited by accounting for site formation processes and taphonomic
transformations. In criminalistic terms, the use of forensic assemblages will help to
access interpretations at the activity level. However, to be able to do this, a forensic
archaeologist should also estimate the probabilities of transfer, persistence and
recovery of the individual materials and traces that combine an assemblage. This is
not necessary if the physical evidence is evaluated at the source level only. For
instance, the presence of a thin polyester sewing thread around a skeletonised lower
leg buried in a pit affected by scavenging could offer moderate support to the
hypothesis that the buried leg was originally wearing a sock made of organic fibres
and that, due to taphonomic alterations, this sock degraded leaving only the more
durable polyester fibre. Additionally, the presence of insect pupae in an area affected
by the scavenging gives – in view of the presence of the polyester fibre and the
skeletonised leg – much more support to the hypothesis that these pupae are related
to the scavenging than to the hypothesis that they are related to the original inter-
ment of the leg in the pit. In other words, by defining the pit and associated physical
evidence as a forensic assemblage and by utilising taphonomic, degradation and
ecological MR-theories, it is possible to classify and sequence (some of the) activi-
ties that took place in and near the pit.

1.3.6 Interpretation

As previously mentioned, the forensic record, as retrieved during crime scene inves-
tigation, will be a selection of the initially deposited physical materials and traces.
This is not only due to the site formation processes and taphonomic transformations
12 W. J. M. Groen

but also the investigative theoretical and methodological framework used during the
crime scene investigation. It is often believed, especially by the scene of crime offi-
cers (SoCOs), that analysis and interpretation of physical evidence can be post-
poned until after the completion of the crime scene investigation. This should be
done, according to the same officers, by collecting physical evidence at a crime
scene as completely and systematically as possible, mostly by following hypothetico-­
deductive method (HD method). In addition, it is believed that the issues of (induc-
tive) analysis and interpretation concerning the crime under investigation can be
postponed until the results from the forensic laboratory arrive. This line of reason-
ing is, from archaeological point of view, incorrect. For example, the HD method,
or the scientific method, was developed to test scientific hypotheses by observable
data in scientific disciplines such as biology, physics and chemistry. This method
works well in experimental sciences, under laboratory conditions. However, this is
not the preferred method to reconstruct past human activities from (a partially
degraded) forensic record or to help evaluate forensic evidence at the activity level.
To reiterate, analysis and interpretation of the material record should start at the
crime scene or place of incident and include a theoretical framework (e.g.
MR-theories) to link the obtained forensic record to inferences about the human
activities and natural processes that could have been responsible for the formation
and alteration of this record in the first place. To help understand and analyse spatial
and temporal patterns in the forensic record, a GIS application should ideally be
used. GIS-based analyses could visualise the spatial layout of the recovered assem-
blages, help to clarify whether data could have been missed due to site formation
processes or taphonomic transformations and help to decide whether an interpreta-
tion of the recovered record is plausible. In this way it would be possible to test
inductive and deductive hypotheses and to assign a preliminary evidentiary value to
the recovered forensic assemblage, given the hypotheses used. Hypotheses that sur-
vive falsification could be used to form scenarios about what could have happened
at the crime scene, although it should also be kept in mind that excluding one
hypothesis does not demonstrate the other to be true when there are more than two
hypotheses, as it is often the case.
To summarise, the proposed interpretative framework will be able to (1) offer an
intellectual tool for understanding the past (assessment), (2) detect assemblages in
the forensic record against which evidential claims can be evaluated (analysis and
interpretation) and (3) offer frames of reference against which observations of the
criminal record may be contrasted to gain understanding (reconstruction). Of impor-
tance is that hypothesis testing at crime scenes will also expose ambiguities. This is
essential since it will force the forensic archaeologist to consider alternative hypoth-
eses or will point him/her towards new hypotheses (and therefore new scenarios)
and consequently initiate a search for alternative physical evidence. In other words
to update the current investigative methodology with new data or information and –
if feasible – to modify this methodology, there should be a feedback loop between
interpretation and final desktop analysis.
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 13

1.3.7 Integration

In archaeology, all individual lines of research, for example, physical anthropology,

zooarchaeology, archaeobotany and ceramic analysis, will eventually be combined
in one overall synthesis or excavation/recovery report. Most of the time finalising
this report will be the responsibility of the archaeologist who was responsible for
the excavation or recovery. Integration of data and information, as obtained during
a forensic archaeological investigation of a police case, is currently not considered
part of forensic archaeology. However, a forensic archaeological analysis and inter-
pretation is sometimes combined with analyses and interpretations obtained from
forensic anthropology (e.g. taphonomic analysis) and forensic ecology (e.g. post
mortem interval (PMI) estimation), to offer a multidisciplinary evaluation of the
encountered evidence and the most likely reconstruction. This constitutes a method
of working that clearly benefits forensic investigation and the subsequent court
proceedings.
Given its interdisciplinary framework, the archaeological process could be help-
ful in combining the separately carried out forensic evaluations – at the crime scene
and in the laboratory – into one or more crime scene reconstruction(s). Such a
reconstruction could best be performed within a Bayesian investigative framework,
for example, using Bayesian networks (e.g. Dawid and Evett 1997) and GIS-related
spatial statistics. An integration of all observations, hypotheses and background
knowledge will surely offer a better reconstruction than the separate areas of foren-
sic expertise can on their own.

1.3.8 Comparison

An archaeological report will compare the results obtained from the finalised
excavation/recovery with results acquired from past excavations/recoveries that
took place in the vicinity of the studied archaeological site. This is important since
identification of (dis)similarities with previous archaeological studies can aid in
further testing of hypotheses, offer new research questions and help the evaluation
of the archaeological research strategy and field methodology.
Case comparison is also useful during a forensic archaeological investigation. It
is relatively easy to accomplish if a GIS-based documentation system was used dur-
ing the investigation. The use of GIS will not only allow a fast and reliable analysis
of the investigated crime scene; it will also provide a fast comparison with previously
investigated crime scenes by means of spatial statistics. In this way information
obtained from different sites can be compared and combined to investigate cases at a
supra-site level. For example, to state how common a tool mark encountered on the
wall of the investigated pit is or to state if the dimensions of the burial pit deviate
from the mean. Comparison with previous casework could be also relevant to formu-
late new forensic hypotheses, to see if possible physical evidence could have been
14 W. J. M. Groen

missed and to evaluate the investigative methodology used during the investigated
case at hand. Additionally, case comparison will also aid in an intelligence-­led
framework of policing and provide additional crime intelligence (e.g. Ribaux et al.
2010a, b).
Finally, after comparison, a (peer reviewed) forensic archaeological report will
be ready to be submitted to the police or the court, depending on the jurisdiction.

1.3.9 Abstraction

Abstraction is the last step in the archaeological research cycle. By formulating new
research ideas and hypotheses about the archaeological past, by updating the exist-
ing background knowledge and by helping to adjust future excavation and recovery
strategies, this step offers the much needed feedback for future preliminary desktop
research. Most of the time archaeological abstraction takes place by means of data-
base updating, scientific publications, workshops and symposia.
Abstraction is also useful in forensic archaeology. New investigations will,
undoubtedly, expose physical evidence finding the current theories and models defi-
cient. Furthermore, it will now be possible to formally update the existing back-
ground knowledge to help formulate new research questions and forensic hypotheses
and to help adjust and therefore improve our investigative methodology. Such
abstractions must fall within the ethical and legal possibilities and limitations, for
example, rules concerning the storing of personal data. Thus, forensic archaeologi-
cal investigation should not be an end in itself. It should also be applied to supra-­
case level and used to generate new theories and models and to improve their
predictive value.

1.4 Development and Innovation

To strengthen its scientific base, forensic archaeological research should also focus
on the refinement of theories, models, methodologies and methods from both the
natural and social sciences. To improve its operational field methodology, forensic
archaeology, as a science, should focus on the pedological, ecological, biological
and criminalistic theories and models that are needed to understand site formation
processes and taphonomic transformations. New theories and models to predict
where certain types of physical evidence may be encountered, given the type of the
crime, and to assign (preliminary) evidential value to the obtained forensic record
should be developed. Such knowledge would not only aid the development of scien-
tific MR-theories but could also help explain the presence or absence of certain
1 Forensic Archaeology: Integrating Archaeology with Criminalistics and Criminology 15

types of material evidence within the retrieved forensic record. Additionally, foren-
sic archaeologists should keep up to date with new research and development within
3D documentation, predictive modelling, GIS science, geophysics and remote sens-
ing and incorporate these innovations into their investigative framework. By inte-
grating criminalistic and criminological theories and models into its interpretative
framework and by using the archaeological research cycle, forensic archaeology, as
a social science, should try to interpret and reconstruct human activities that took
place at the crime scene or place of incident. The to be developed criminological
MR-theories could be built by studying the location and distribution of relevant
forensic assemblages resulting from different types of criminal activities. Depending
on the questions to be addressed and the forensic hypotheses to be tested, such theo-
ries could be inductive (linking the empirical reconstructions to higher schemes or
theories) or deductive (creating testable forensic hypotheses derived from higher
schemes or theories). Of importance here is that forensic archaeology, in contrast to
main stream archaeology, has direct access to information about past human behav-
iour (specifically information regarding criminal behaviour in the form of court
testimonies, perpetrator confessions and trial rulings) and can therefore evaluate our
interpretations and reconstructions of the past, something that is not possible in
mainstream archaeology.
Four primary types of research could provide background knowledge, as
described above: (1) case review studies concerning evidence patterns, evidence
quantities, assigned evidentiary value, etc., as encountered at different types of
crime scenes during past casework, (2) criminalistic studies concerning evidence
patterns based on actualistic experiments using simulated crime scenes, (3) crimi-
nological studies concerning (offender’s and witness’s) testimonies and behaviours
that could help explain the evidence patterns encountered at crime scenes during
regular casework and (4) review of jurisprudence concerning the use of hypotheses
and scenarios, as addressed by the court.
Of importance is here that this knowledge should be stored in indexed documents
(by tagging or associating relevant information) and in quantitative and qualitative
databases and that it should be linked to geographical and spatial locations, prefer-
ably within a GIS application. Such indexed documents and databases could be
used as predictive tools during the different forensic archaeological investigative
phases. During the preliminary desktop research, they could, for example, be used
to predict where clandestine burials could be located and what the taphonomic con-
dition of such sites might be. During the field investigation, the GIS-related data-
bases could be used to map and document the encountered forensic record, to
provide up-to-date view of this record and to test hypotheses. Post-excavation
indexed documents and databases could be used as analytical tools, to analyse and
reconstruct the obtained forensic record and to test additional forensic hypotheses.
To ensure relevancy all databases, if ever developed, should constantly be updated
with new data and information from new forensic archaeological investigations.
16 W. J. M. Groen

1.5 Conclusion

Forensic archaeology can be redefined as an archaeological discipline that com-

bines the archaeological framework with the theory and methodology of criminalis-
tics and criminology, all this within the context of (criminal) law. The aims of
forensic archaeology should not only be to preform investigations in casework but
also to conduct research. Forensic archaeology can only evolve scientifically and
methodologically if it is broadened beyond the general boundaries of a technical
investigation. If forensic archaeological practice is redefined as a cyclic scientific
process, it will not only be able to generate new forensic knowledge and new ways
to look at the physical evidence, but it will also be able to improve future casework
by means of case abstraction.
The introduction of the archaeological research cycle, crime scene formation
models, taphonomic transformations and the concept of background knowledge and
forensic assemblages in forensic archaeology will without a doubt aid in the loca-
tion of physical evidence that is capable of discriminating between forensically rel-
evant inductive and deductive hypotheses. Such an approach is strengthened by
linking the obtained forensic record to human activities that took place before, dur-
ing and after the crime was committed. To help consider where to look for certain
types of physical evidence and what its evidentiary value, given the forensic hypoth-
eses addressed, could be and to evaluate why certain types of physical evidence
were not discovered, the proposed framework should be integrated with the Bayesian
paradigm.

Acknowledgements The author would like to express his gratitude to Coen Nienaber (Netherlands
Forensic Institute) for his helpful and constructive comments on the earlier version of this chapter
that contributed to its final version.