Police Photo

Police Photography : Visual Representation, Scientific Evidence
and the Construction of the Criminal
by
Jonathan M. Finn
Submitted in Partial Fulfillment
of the
Requirements for the Degree
Doctor of Philosophy ^
Supervised By
Professor Lisa Cartwright
Program in Visual and Cultural Studies
Department of Art and Art History
The College
Arts and Sciences
University of Rochester
Rochester, New York
May 2003
R ep ro d u ced with p erm ission o f th e copyright ow ner. Further reproduction prohibited w ithout perm ission.
UMI Number: 3102272
UMI
UMI Microform 3102272
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company

300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Curriculum Vitae
Jonathan Finn was bom in Ottawa, Ontario on July 14, 1972. He attended
McMaster University from 1990 to 1995 and graduated summa cum laude with a
Bachelor o f Arts in Art History in 1995. While at McMaster Jonathan pursued a year
of study in the Art History Department of Leeds University, England. Jonathan
entered the graduate program in Art History at York University in 1995, earning his
Masters o f Arts degree in 1997. He began the Visual and Cultural Studies Program at
the University of Rochester in 1997. Jonathan pursued his research in forensic
photography under the direction of Professor Lisa Cartwright and received the Master
of Arts degree from the University of Rochester in 2001.
Abstract
This dissertation, “Police Photography: Visual Representation, Scientific Evidence
and the Construction of the Criminal,” examines the use of photographic
representation and photographic technologies in modem and contemporary American
and Canadian law enforcement and criminal identification practices. I combine
approaches from visual studies, particularly the history of photography, and from
science and technology studies to position the photograph as an inscription, as a
material representation at work in the constmction of scientific evidence. I argue that
the production, archiving and exchange of photographic representations in law
enforcement and criminal identification practices function in the constmction of
identity.
The dissertation addresses three identification practices fundamental to modem
and contemporary law enforcement: fingerprinting, DNA analysis, and information
databases. I stress that each is dependent on photographic representation. I argue that
the specific collaboration of photography and the theory of fingerprint identification
at the close of the nineteenth century gave rise to law enforcement practices based on
the collection and use of latent identification data. This transformed the subject of
law enforcement from the individual, physical body to a larger, aggregate social
body. I trace the development of DNA analysis in law enforcement at the close of
the twentieth century to examine the development of power and authority associated
with the control o f inscriptions. I argue that the FBI constructed a dominant position
in the field of DNA analysis through the development of control over the visual
representation of DNA, the autoradiograph.
The dissertation concludes with an examination of two digital archives used in
current law enforcement and criminal identification practices, the National Crime
Information Center 2000 and the Combined DNA Index System. The theoretical
possibilities associated with the collection and use of latent identification data
announced at the close of the nineteenth century have become a reality in 2003. I
suggest that contemporary law enforcement practices are driven by the collection and
archiving of identification data and argue that this practice functions in the
redefinition of criminality and the body. I argue that the body in the twenty-first
century digital archive is constructed as a site of potential criminality.
V
Table of Contents
Curriculum Vitae ii
Abstract iii
Table o f Contents v
List of Figures vi
Foreword ix
Chapter One Introduction 1
Chapter Two Picturing the Criminal: Photography and 19
Criminality in the Nineteenth Century
Chapter Three Photographing Fingerprints: Data, 60
Evidence and Latent Identification
Chapter Four DNA Analysis: Competition, 102
Standardization and the Control of
Inscriptions
Chapter Five The Body in the Digital Archive: 145
Redefining Crime and Potential
Criminality
Conclusion 177
Works Cited 183
vi
List of Figures
Figure Title Page
Photographic representations of the criminal: a 1
mugshot, fingerprint and autoradiograph.
2.1 The Inspector’s Model. 21
2.2 Two examples of the visibly deviant body used as 34
illustrations in Lombroso’s work. “Italian Criminal. A
Case of Alcoholism,” (L) and “Anton Otto Krauser,
Apache” (R).
2.3 Tools used by Lombroso for measuring the “signs” of 37
criminality. The algometer (L) for measuring “general
sensibility and sensibility to pain,” and the campimeter
(R) for measuring field of vision.
2.4 “Physiognomy of Russian Female Offenders.” 40
2.5 Galton composite photographs showing the “criminal 44
type.”
2.6 Galton composite photographs showing the “Jewish 44
type.”
2.7 A page of images depicting male criminals from 50
Thomas Byrnes’ Professional Criminals o f America.
vii
2.8 A page of images depicting female criminals from 51
Thomas Byrnes’ Professional Criminals o f America.
2.9 Measuring the width of the skull for anthropometric 54
data.
2.10 Bertillon’s “synoptic table of facial expressions.” 55
2.11 A Bertillon identification card showing the recording of 56
anthropometric measurements, physical description and
full face and profile photographs.
3.1 The ridge characteristics of the fingerprint. 91
3.2 Fictional crime scene: the first photograph positions the 92
dinner plate within the context of the crime scene.
3.3 Fictional crime scene: the second photograph positions 92
the fingerprint on the dinner plate.
3.4 Fictional crime scene: the third photograph isolates the 93
found fingerprint.
3.5 Fictional crime scene: the print from the plate is lifted 93
onto a sheet of acetate, enabling it to be scanned into
the AFIS database.
3.6 Fictional crime scene: the criminal record matching the 94
identified print is collected.
3.7 Fictional crime scene: a negative is made from the 95
found print.
viil
3.8 Fictional crime scene: a negative is made of the latent 95
print from the fingerprint record.
3.9 Fictional crime scene: the fingerprint chart with its 96
accompanying narrative.
4.1 The DNA autoradiograph. 109
5.1 NEC’s LS21™ Live Scan Workstation. 162
5.2 A latent fingerprint workstation by NEC. 163
5.3 Nanogen’s NanoChip® portable DNA scanner. 163
Foreword
The writing of this dissertation took form largely according to the numerical order
of the chapters. Wherever possible I have left my discussion and analysis of law
enforcement practices and initiatives in their original tense. Thus, in Chapter Two I
refer to the proposed plan of United States Attorney General John Ashcroft to
fingerprint, and photograph visitors to the United States and in Chapter Four I discuss
this same topic, the National Security Entry-Exit Registration System, as it is now
legal and in full operation. I have refrained from revising the earlier chapters of the
text to reflect the specific time within which this dissertation was written. The events
of September 11, 2001 in which passenger airplanes were hijacked and crashed into
the World Trade Center, the Pentagon and in a field in rural Pennsylvania, and, more
specifically, the response to them by law enforcement organizations and agencies
have had significant influence on the subject of this dissertation. Under the rubric of
a “war on terror” law enforcement agencies in the United States have been invested
with significant power and have received substantial funding to improve and
implement identification equipment and programs. Subsequently, concerns
surrounding law enforcement and criminal identification practices that were largely
theoretical in 2000 have become a reality in 2003.
1
CHAPTER ONE
INTRODUCTION
ij s i f i !*
III ! ; I S ! l l i
isi i " ...... .
SI
The images above function as visual substitutes for the human body within
practices of criminal identification. The image on the left is of James Allen, a “hotel
thief and confidence man,” arrested, photographed and included in Thomas Byrnes’
1886 text, Professional Criminals o f America. The middle image represents a
fingerprint identified and photographed by the Metropolitan Toronto Police in a
fictional crime scene of 1996. The final image is an autoradiograph depicting DNA
analysis from a rape-murder case and printed in the National Research Council’s
1992 report, DNA Technology in Forensic Science}
1DNA Technology in Forensic Science (Washington, D.C.: National Research Council,

1992); Thomas Byrnes, Professional Criminals o f America (New York: Cassell and Company, 1886).
2
Through these representations the criminal subject is recorded, studied, followed
and ultimately identified. In addition to their roles as evidentiary statements, these
images also participate in important debates on privacy, surveillance and power. Far
from being mere static representations, these photographs are engaged in a
Foucauldian “micro-physics o f power:” they function across juridical, medical, legal
and other discourses in the construction o f identity and its attendant rights and
limitations. As sites o f tremendous social importance, to produce and interpret these
representations is to participate in cultural constructions of normalcy, deviance,
gender, race, and in moralistic debates of right and wrong. As such, the need to
examine this type of representation critically is paramount. This dissertation, “Police
Photography: Visual Representation, Scientific Evidence and the Construction of the
Criminal,” is an attempt to address this need.
Continued scientific and technological innovations have greatly altered the ways
in which the criminal body has been represented visually within law enforcement and
criminal identification practices. Over the past two decades a steady flow of
technological tools has become available, enabling new ways of representing the
crim inal body. Infrared and ultra-violet lasers, chemical baths, DNA analysis,
computer databases, and digital imagery have yielded several new visual substitutes
for the criminal. Despite the specific nature of such changes, visual representation
has remained an important means through which the criminal subject is understood
The fictional crime scene was produced during an apprenticeship held with the Metropolitan Toronto
Police's Forensic Identification Services in August, 1996.
3
and defined. The act of representing, or “making the criminal visible,” is central to
contemporary law enforcement and criminal identification practices and is the focus
of my dissertation. The criminal is most obviously represented through the traditional
police photograph, or mug shot. However, the criminal is also represented in more
disparate and subtle ways, as in the fingerprint, the autoradiograph, and other forms
of visual and textual evidence. Reading these images as objective documents or
records presupposes the existence of the criminal body. In contrast to this, by taking
the criminal as a social construct, I aim to locate and examine the production and
interpretation of photographic representation as it functions in the process of
construction. Further, through my dissertation I will show that contemporary visual
technologies at work in criminal identification are directly dependent on scientific
and social scientific discourses established in conjunction with the development of
police photography in the nineteenth century.
By the close of the nineteenth century, photographic representation of the criminal
body was enmeshed in a socially defined binary of normal and deviant and in
questions of power, surveillance and privacy. Now at the beginning of the twenty-
first century, we are caught in an interesting paradox. Technological developments
have advanced the abilities o f individuals and groups, both public and private, to
record and study the human body down to its minutiae, often without its consent or
acknowledgement. However, and partly in response to this development, this activity
is taking place within and by a society that is increasingly concerned with such
intrusive practices and that prides itself on its democracy. This tension is nowhere
4
more clear than in the recent proposal of United States Attorney General John
Ashcroft. Under the rubric of the “war on terror,” the proposal calls for photographs
and fingerprints to be taken o f all visitors to, and foreign visa holders living in, the
United States who originate from Middle Eastern countries.2 Ashcroft’s
announcement attests to the tremendous power and authority vested in the
photograph. Proposed uses of the medium in this way reinforce its highly
problematic nature as a mode of representation.
What must be addressed is the changing nature of visual representation in
practices of criminal identification from the nineteenth century to the present day. To
what extent are contemporary methods of visual representation merely new
manifestations of nineteenth-century practices? What new issues have arisen and
what have disappeared? And, perhaps most importantly, what is at stake in these new
methods of representation, as they assist in defining the criminal subject?
In addressing the role of photographic representation within the construction of the
criminal, an interdisciplinary approach is essential. Visual representations of the
criminal body emerge from continual negotiations between humans, technology and
the social networks through which they interact. The naming, regulating,
classification and identification of the criminal body have roots in the intersections of
scientific and visual practices. Adding to the complexity is the unique status of
photography and the extensive debates about the medium’s subjectivity and
2 Ashcroft’s announcement was made on June 5,2002.
5
objectivity. To approach this topic, I combine analyses from the field of visual
studies, particularly art history, and from the field of science and technology studies.
Within the still limited body of academic work that addresses the photographic
representation of the criminal, that of John Tagg and Alan Sekula remains
unparalleled. Tagg’s The Burden o f Representation and Sekula’s “The Body and the
Archive” remain the most important texts on the topic.3 Working from the ideas set
forth in Discipline and Punish, Tagg addresses the ubiquity of photographic
representation within Foucault's conception of the carceral network.4 To this end he
investigates the medium as it functioned throughout a series of nineteenth and early
twentieth-century practices. His analysis includes discussions on the rise of
photographic portraiture, the establishment of the photograph as legal document, the
use of photography in medical and police practices, and the federally sanctioned
documentary work of the Farm Security Administration. In all cases, Tagg refutes
the possibility of the universality of the photograph, stressing the historical specificity
of its value and meaning. He writes: "the photograph is not a magical 'emanation' but
3 Allan Sekula, “The Body and the Archive,” October 39 (1986): 3-64; John Tagg, The
Burden o f Representation: Essays on Photographies and Histories (Basingstoke: Macmillan, 1988).
Other articles and books have addressed the topic. See: David Green, “Classified Subjects,” Ten. 8, no.
14 (1984): 30-37; David Green, “On Foucault: Disciplinary Power and Photography,” Camerawork 32
(1985): 6-9; David Green, “Veins of Resemblance: Photography and Eugenics,” The Oxford Art
Journal 7, no. 2 (1985): 3-16; Tom Gunning, “Tracing the Individual Body: Photography, Detectives,
and Early Cinema,” in Cinema and the Invention o f Modern Life, ed. Leo Chamey and Vanessa R.
Schwartz (Berkeley: University of California, 1995), 15-45; Suren Lalvani, Photography, Vision, and
the Production o f Modern Bodies (Albany: State University of New York, 1996); Sandra Phillips,
Mark Haworth-Booth, and Carol Squires, Police Pictures: The Photograph as Evidence (San
Francisco: Chronicle Books, 1997).
4 Michel Foucault, Discipline and Punish: the Birth o f the Prison, trans. Alan Sheridan (New
York: Vintage Books, 1979).
6
a material product of a material apparatus set to work in specific contexts, by specific
forces, for more or less defined purposes."5 This leads Tagg to refute any single
history o f photography, referring to it instead as "a flickering across institutional
spaces.
Whereas Tagg focuses on the use of photography as it was employed by state
institutions, Sekula de-emphasizes the specificity of the medium, positioning it within
a larger "bureaucratic-clerical-statistical system..." or, in his now famous term, an
archive.7 Sekula describes the archive as a particularly visual space where the good,
the heroic and the celebrated of society coexist with and are defined alongside the
criminal, the deviant and the poor. Within the developing archival mode of the
nineteenth century, practices of physiognomy, phrenology, eugenics, and social
statistics grew in conjunction with the new field of criminology and provided a
scientifically legitimized discourse of the criminal body. Culminating in the work of
the Italian criminologist, Cesare Lombroso, criminality could be read directly from
the body and, by extension, captured in its visual representation.
In addressing contemporary practices of criminal identification the work of Tagg
and Sekula leaves space for continuing critical investigation. The myriad of new
visual and information technologies at work in criminal identification calls for a
rethinking o f their ideas, which refer almost exclusively to nineteenth-century
5 Tagg, Burden, 3.
6 Tagg, Burden, 118.
7 Sekula, "The Body," 16.
7
practices. In addition, in its emphasis on a more abstract or theoretical field of visual
representation, the work of Tagg and Sekula does not address the concrete production
and use of these images within scientific practice.
To address these questions and to complement work in visual studies, my project
draws from the field of science and technology studies, where it focuses on the social
construction of scientific facts.8 A metaphor often used in social studies of science
and technology is that of the black box. In Science in Action Bruno Latour uses the
metaphor to trace the construction of "scientific fact and technical artefacts" through
scientific practice.9 For Latour, a central feature of scientific practice is the ability to
black box, or to make given, certain concepts, practices and presuppositions. In this
way, the complex negotiations that take place in the construction of scientific facts
are enclosed by the black box and rendered invisible in the final product. Once
closed, the black box can serve as a site to be used in future scientific practice (so that
any black box can contain within it a series of other black boxes). Because of this
process, and key to the metaphor, Latour emphasizes that analyses of scientific
practice should not take the final product as their object of study. Instead, social
studies of science should begin at the planning stages in an attempt to "be there before
8 There is a wealth of material dealing with the social construction o f scientific facts. Some of
the central texts informing this dissertation include: David Bloor, Knowledge and Social Imagery
(London; Boston: Routledge & K. Paul, 1976); H. M. Collins, Changing Order: Replication and
Induction in Scientific Practice (London; Beverly Hills: Sage Publications, 1985); Thomas S. Kuhn,
The Structure o f Scientific Revolutions, 2nd, enl. ed. (Chicago: University of Chicago Press, 1970);
Bruno Latour, Science in Action: How to Follow Scientists and Engineers Through Society
(Cambridge, Mass.: Harvard University Press, 1987); Bruno Latour and Steve Woolgar, Laboratory
Life: The Construction o f Scientific Facts, 2nd ed. (Princeton: Princeton University Press, 1979).
9 Bruno Latour, Science, 21. The metaphor is used outside science studies as well. I use
Latour's account of it here as I find it the clearest explanation.
8
the box closes and becomes black."10 In its ultimate manifestation Latour's method
positions the researcher alongside the scientists, documenting every move in the
creation o f scientific facts.
I want to suggest that, in addressing the photographic representation of the
criminal, work in the social study of science allows us to avoid the traps of this
metaphor. That is, instead of beginning with the final product (the photograph of the
criminal), we should start with the "complex negotiations" in the process of
production.11 In doing so here I want to complement the work of Tagg, Sekula and
other photographic theorists and historians by addressing the complex negotiations in
the production and use of photographs of the criminal body: to "be there before the
box closes and becomes black."
The Photograph as Immutable Mobile
Throughout this dissertation I position the photograph as an inscription, or
immutable mobile, within the social construction of scientific knowledge. In my use
of the terms I draw primarily from the work of Bruno Latour and Steve Woolgar. In
Laboratory Life, the authors trace the construction of scientific facts by examining the
10 Latour, Science, 21.
11 This could also be said of much work in the history o f photography and is hinted at by
Patrick Maynard in The Engine o f Visualization. Maynard suggests that too often "photographs are
considered a puzzling sort of thing, to be addressed in terms o f their relationship to other things:
notably the relationship o f being a photograph o f other things." Patrick Maynard, The Engine o f
Visualization: Thinking Through Photography (Ithaca, N.Y.: Cornell University Press, 1997), 10.
9
daily activities within a specific endocrinology laboratory.12 The book is based on
research conducted by Latour during a 1975-77 study of laboratory work at the Salk
Institute for Biological Studies. In writing the text, Latour and Woolgar begin with
the description of a fictitious character through which their text will take form. “The
observer” is conceived of as a newcomer to the laboratory. He has no specific
knowledge about the activities taking place in the lab; he is an anthropologist, trying
to understand its practices. Latour and Woolgar use this fictional character to set out
the central question of their book. The observer asks: “How is it that the costly
apparatus, animals, chemicals, and activities of the bench space combine to produce a
11
written document, and why are these documents so highly valued by participants?”
Throughout their text, Latour and Woolgar demonstrate the central importance of
documents, or inscriptions, in the production of scientific facts. As diagrams, curves,
pictures, lists, tables, or textual statements, inscriptions are the basic currency of
scientific practice. They are produced in the lab by inscription devices, defined as
“any item of apparatus or particular configuration of such items which can transform
a material substance into a figure or diagram which is directly usable by one of the
members of the office space.”14 Crucially, the labour and procedures involved in the
production of inscriptions are lost in the final representation. Latour and Woolgar
point to two consequences of this action. First, “inscriptions are seen as direct
12 Latour, Laboratory.
13 Latour, Laboratory, 48.
10
indicators o f the substance under study.” 15 Second, the process gives way to a
“tendency to think of the inscription in terms of confirmation, or evidence for or
against, particular ideas, concepts, or theories.”16
For Latour and Woolgar, the importance of inscriptions and inscription devices
w ithin the laboratory cannot be overstated. They are central in identifying, defining
and even creating the object of study. They write: “it is not simply that phenomena
depend on certain material instrumentation; rather, the phenomena are thoroughly
constituted by the material setting of the laboratory.”17 In a circular manner,
inscriptions build on inscriptions, simplifying increasingly complex data. Through a
process Latour calls a “cascade of ever simplified inscriptions,” diverse, real-time
events are continually reduced and transformed into two-dimensional
representations. 18
•
In his essay, “Drawing Things Together,” Latour again investigates the role of
inscriptions in the construction of scientific facts, this time focusing specifically on
visual representation.19 Of central importance is the transformation from live object
to two-dimensional representation. For Latour, scientific practice is driven by the
creation and use of images. He writes:
18 Bruno Latour, “Drawing Things Together,” in Representation in Scientific Practice, ed.

Michael Lynch and Steve Woolgar (Cambridge: MIT Press, 1990), 19-68; 40.
19 Latour, “Drawing.”
11
Scientists start seeing something once they stop looking at nature and
look exclusively at prints and flat inscriptions. In the debates around
perception, what is always forgotten is this simple drift from watching
confusing three-dimensional objects, to inspecting two-dimensional
images which have been made less confusing.
It is Latour's emphasis on images that are made less confusing that is so interesting.
Latour’s account allows us to see the specific importance of visual representation in
the construction and black-boxing o f scientific fact. Crucially, as inscriptions, visual
representations are not simply points of reference or illustrations used during
scientific practice. They are the carefully produced, dynamic sites where meaning
and knowledge are negotiated and scientific facts are created.
In “Drawing Things Together” Latour expands his discussion of inscriptions,
labeling the representations used in scientific practice as “immutable mobiles.” Five
features define these objects. They are mobile, immutable, presentable, readable and
combinable with one another. To understand these features it is important to note that
Latour opposes the notion of science as an isolated, individual or hermetic practice.
Rather, scientific practice functions within a diverse network of social relations.
Scientific practice proceeds through continued dialogue between individual and
groups of practitioners, between laboratories, companies and institutions and through
professional channels such as conferences and journals. Therefore, in order to be
20 Latour, "Drawing," 39.
12
effective, an inscription produced in any specific laboratory must be able to be
exchanged, interpreted and reinterpreted throughout this network of individuals,
contexts and locations and it must do so without the loss of essential information.
The inscription must be mobile and the information it contains must be immutable.
As the basic currency of scientific practice, the construction and deployment of
immutable mobiles is crucial. Put simply, in the process of establishing scientific fact
the group that produces, employs and controls the best immutable mobiles wins out
over others. To this end, diverse, real-time events are continually reduced and
transformed into two-dimensional representations, or immutable mobiles, and made
less confusing.
The notion of inscription and the concept of the immutable mobile are particularly
effective in studying modem and contemporary criminal identification practices. The
reason for this is two-fold. First, I argue that the modem conception of the criminal
body, developed in the nineteenth century, was tied to contemporaneous
developments in the social sciences that sought ways of classifying, systematizing,
storing and sharing information. One result of this confluence of activity was an
increased need for visual representation within social, institutional and scientific
contexts. Visual representation, or inscription, allowed for the substitution of the
individual with the group and for the study of large numbers of bodies and objects.
Second, a defining feature of the modem criminal, developed in the latter half of the
nineteenth century, was his mobility. Law enforcement practices of the time reflected
a need to communicate identification information between specific locales. In order
13
for this communication to be effective, information identifying the criminal’s body
had to be mobile and immutable. Inscriptions of the criminal’s body had to move
seamlessly between contexts and locations without the loss or alteration of the
necessary information. Inscriptions were helpful in reducing the complex three-
dimensional body into a two-dimensional representation; however, to be effective
within law enforcement practices that were increasingly reliant on communication,
the inscription of the criminal body had to be made into an immutable mobile.
The camera developed in conjunction with this confluence of scientific and social
scientific activity of the nineteenth century. I stress that the supposed objectivity,
cost-effectiveness and near universal acceptance of the photograph assured its central
position as an inscription across social, institutional and scientific practices. In
addition, the mechanical nature of the camera guaranteed the immutability of the
information within the frame o f the image.
In “The Body and the Archive,” Sekula positions the photograph as a site where
the disparate and diverse activities of the live world are reduced to a standardized
two-dimensional format and made comparable. Like Sekula, I argue that the
photograph, as an inscription, helps to translate information about the body contained
in blood, hair and in fingerprints into comparable, two-dimensional representations.
Law enforcement agencies use the camera to transform the body, with all its complex
idiosyncrasies, into images that are less confusing. Just as in the endocrinology
laboratory, the resultant representations are taken to be “direct indicators” of the
subject of study. However, rather than indicating molecules, the inscriptions
14
produced and used by law enforcement agencies indicate the criminal body. Within
law enforcement and criminal identification practices, images of faces, fingerprints
and DNA are routinely collected and compiled into networks of similar
representations, to be reproduced, read, and exchanged as necessary.
Importantly, and again drawing from the work of Sekula, I stress that photography
and photographic technologies are only a part of the diversity of scientific,
technological, social, economic and cultural practices at work in law enforcement and
criminal identification. My treatment of photography and photographic
representation in this dissertation is purposefully and necessarily broad. I address the
concrete production and use of photographic technologies in law enforcement, while
also using the image as a point of departure to investigate more abstract and
theoretical issues associated with law enforcement and criminal identification such as
the construction of identity and issues of power, surveillance and privacy.
This dissertation is chiefly concerned with three specific facets of modem and
contemporary criminal identification: fingerprinting, DNA analysis, and information
databases. Each is fundamental to current identification practices and each is
dependent on photographic representation. In addition, and to fully understand the
nature of photography within these specific practices I also outline the
contemporaneous development of photographic representation and the criminal body
in the nineteenth century.
Chapter Two provides the necessary historical and contextual framework for the
dissertation. I outline the developing relationship between photography and criminal
15
identification in the nineteenth century and survey the existing literature on the
subject. I show that by the close of the nineteenth century the body had been defined
as a text throughout a series o f scientific and social scientific practices. Culminating
in Cesare Lombroso’s 1876 text, L ’Uomo Delinquente, the new field of criminology
promoted the ability to read normalcy and deviance directly from the body via the
visual signs it was believed to offer.
The chapter argues that the camera is an essential component in the construction of
the modem criminal subject. Photographic technology developed in conjunction with
the newly emerging social institutions of the nineteenth century and served as a
fundamental tool in recording the identity of both normal and deviant bodies. I stress
that, as it developed for use within law enforcement and criminal identification
practices, the camera originally served as a tool of representation. However, by the
close o f the nineteenth century the photograph as static representation was mobilized
and transformed into an inscription through the classificatory system of Alphonse
Bertillon.
Chapter Three examines fingerprinting as a method of criminal identification. In
contrast to existing literature on the subject which positions fingerprinting within a
teleology of identification practices I argue that the specific collaboration of
fingerprint identification and photography announced a new mode of law
enforcement, defined by a new role of the camera. Fingerprinting gave rise to
practices of latent identification within which the camera functioned as a tool of data
collection. During the latter half of the nineteenth century and prior to
16
fingerprinting’s introduction to law enforcement, the camera was used to represent
and record the captured criminal’s body. The identity of the body as normal or
deviant often preceded the use of the camera. In contrast to this, the ability to
photograph fingerprints independent of the presence and knowledge of the physical
body gave rise to the collection of latent identification data. Latent data represents an
unknown physical body, its identity to be determined within the confines of the police
laboratory. I argue that the specific intersection of photography and fingerprint
identification greatly extended the power and reach of law enforcement in public life.
The chapter provides extensive primary source research on fingerprint
identification. Specifically, I draw on research conducted during an apprenticeship
position with the Metropolitan Toronto Police’s Forensic Identification Services
Division. I use a fictional crime scene produced during this research to show the
centrality of photography within fingerprint identification. In the chapter I argue for
the emergence of a new archive within early twentieth-century law enforcement
agencies, one that included the collection and storage of latent identification data.
Whereas Chapter Three focuses on the concrete production and interpretation of
inscriptions within criminal identification, Chapter Four addresses the subject of the
institutional control of inscriptions. The chapter traces the development of DNA
analysis in law enforcement from the 1980s through to its formalization in the DNA
Identification Act of 1993. I use Latour’s concept of “dominance” over inscriptions
to analyze the FBI’s construction of authority over DNA analysis. I argue that by
17
controlling the inscription of DNA, the autoradiograph, the FBI has achieved a
monopoly over DNA analysis in law enforcement.
The 1993 DNA Identification Act, enacted in 1994, legalized the FBI as the
authority over DNA analysis standards in law enforcement. I stress that this brought
a premature end to the productive, critical debate surrounding the technique and that
it served as a necessary precondition for the development of “dominance” over DNA
analysis. I argue that the DNA Identification Act enabled the FBI to achieve
dominance over DNA analysis in three primary ways: it legalized the Bureau’s
control over the procedures for analysis; it constructed a workable boundary for the
FBI within the larger field of DNA analysis; and it provided the FBI with substantial
funding to disseminate their standards and procedures to law enforcement agencies
throughout the United States.
Chapter Five returns to the notion of the archive and investigates contemporary
information databases in law enforcement. I argue that, in contrast to Sekula’s
nineteenth-century archive which was limited by its physicality, the twenty-first
century digital archive exists largely free of these physical constraints. The
theoretical possibilities associated with the collection of latent identification data,
announced by the intersection of fingerprinting and photography as addressed in
Chapter Three, have become a reality in the twenty-first century. Subsequently, law
enforcement agencies routinely collect and store vast sums of identification data for
its potential use. I argue that these practices of data collection betray any clear
rationale and function in the redefinition of crime, criminality and the body.
18
In the chapter I analyze two specific digital archives in contemporary law
enforcement: the National Crime Information Center 2000 and the Combined DNA
Index System. In each case I question the rationale governing the collection of
identification data and its storage in these databases and raise concerns regarding its
potential misuse. Specifically, and drawing from the discussions of Chapters Three
and Four, I argue that the data contained in these databases are archived as immutable
mobiles, and that these inscriptions can then be re-deployed according to the
changing needs of law enforcement and to mutable conceptions of crime and
criminality. I argue that the body in the digital archive is defined as a site of potential
criminality.
This dissertation contributes to the limited body of work that addresses the
intersections of photography, law enforcement and criminal identification. I provide
extensive primary source research on fingerprinting, DNA analysis and contemporary
information databases used by law enforcement agencies. This research is combined
with a critical analysis o f law enforcement and criminal identification practices that
stresses the importance o f visual representation in the construction of identity. In
addition, by combining perspectives from visual studies and from science and
technology studies, this dissertation presents an approach to critically addressing
photography and photographic technologies that acknowledges, reflects and is
directly drawn from the complexity of the specific subject of study.
19
CHAPTER TWO
PICTURING THE CRIMINAL: PHOTOGRAPHY AND CRIMINALITY IN THE
NINETEENTH CENTURY
The police mug shot has become an icon in contemporary visual culture. The
pose, framing and formal conventions of the image are features easily recognized
throughout the general public. The mug shot forms the basis of television shows like
America’s Most Wanted and continually surfaces throughout the myriad of police
dramas on television and in film. It has been the subject of an art exhibition at the
San Francisco Museum of Modem Art;1 and it has formed the seat of controversy as
in the picture o f O.J. Simpson used by Time and Newsweek for their covers of June 27
1994.2 Most recently, as outlined by United States Attorney General John Ashcroft,
the mug shot might become the newest technological tool adopted for use in the “war
on terror.”3 In this last example, mug shots would be required of all visitors to the
United States and current foreign visa holders who originate from Middle Eastern
countries. Ashcroft’s announcement raises deep and immediate concerns about racial
profiling, stereotyping and social policy in the contemporary United States. These
1“Police Pictures: the Photograph as Evidence,” San Francisco Museum of Modem Art,
October 17, 1997 - January 20, 1998. Curated by Sandra S. Phillips. Catalogue: Sandra Phillips,
Mark Haworth-Booth, and Carol Squires, “Police Pictures: The Photograph as Evidence,” (San
Francisco: Chronicle Books, 1997).
2 Time and Newsweek used the same police photograph o f Simpson; however, Time
heightened the contrast o f the image making Simpson’s skin appear darker.
20
same issues have been present since the origin of the mug shot but have only become
clear with the benefit of historical distance. Ashcroft’s proposal highlights the
tremendous authority attributed to the image and the subsequent importance of
critically addressing this mode of representation.
The mug shot has its roots in photographic portraiture of the mid-to-late nineteenth
century. Here the formal conventions of the image were defined: frontal, head and
shoulders, clear, and, where possible, no facial expression. The goal was the same
despite the nature of the portrait: to capture all the significant details of the sitter’s
face. The face was taken as representative of the individual’s character. Criminals
subjected to the camera were often depicted attempting to twist and turn and change
their facial expressions to avoid identification (Figure 2.1).
Contemporaneous with the rise of photographic portraiture and the police mug
shot was the development of the new field of criminology. This work promulgated
direct links between the physical body and criminal propensity and was championed
in the work of Cesare Lombroso.4 Lombroso pioneered the field of criminal
anthropology and his work is central to the development of modem, Western
disciplinary practices. He was the first to apply positivist methods to the study of
J Ashcroft’s announcement was made on June 5,2002.
4 Cesare Lombroso and William Ferrero, The Female Offender, ed. M.A. W. Douglas
Morrison, The Criminology Series (New York: D. Appleton and Company, 1895); Cesare Lombroso,
Crime: Its Causes and Remedies, trans. Henry P. Horton, M.A., Patterson Smith Reprint Series in
Criminology, Law Enforcement, and Social Problems (Montclair: Patterson Smith, 1968 (c. 1911));
Gina Lombroso-Ferrero, Criminal Man: According to the Classification o f Cesare Lombroso,
Patterson Smith Reprint Series in Criminology, Law Enforcement, and Social Problems (Montclair:
Patterson Smith, 1972 (c. 1911)).
21
crime, describing the “bom criminal” and linking criminal behaviour to atavism.
Under Lombroso, criminality was something that could be read directly from the
body and, by extension, captured in its visual representation.5
Figure 2.1: “The Inspector’s Model.” From Thomas Byrnes’ 1886, Professional
Criminals o f America.
5 On the development of criminal anthropology both in Europe and in the United States see:
Nicole Hahn Rafter, “Criminal Anthropology in the United States,” Criminology 30, no. 4 (1992): 525-
545. For general discussions on Criminology and the work o f Lombroso see: Israel L. Barak-Glantz
and C. Ronald Huff, ed., The Mad, the Bad, and the Different: Essays in Honor o f Simon Dinitz
(Lexington: Lexington Books, 1981); Piers Beime, ed., The Origins and Growth o f Criminology:
Essays on Intellectual History, 1760-1945 (Aldershot: Dartmouth, 1994); David G. Horn, Social
Bodies: Science, Reproduction, and Italian Modernity, Princeton studies in culture/power/history
(Princeton, N.J.: Princeton University Press, 1994); David G. Horn, “This Norm Which is Not One:
Reading the Female Body in Lombroso's Anthropology,” in Deviant Bodies: Critical Perspectives on
Difference in Science and Popular Culture, ed. Jennifer Terry and Jacqueline Urla, Race, Gender, and
Science (Bloomington: Indiana University Press, 1995), 109-128; C. Ray Jeffery, Criminology: An
Interdisciplinary Approach (Englewood Cliffs: Prentice-Hall, 1990).
22
This chapter examines the development of the modem criminal subject through the
work of Lombroso alongside the contemporaneous development of photography. It is
my contention that our contemporary understanding of the criminal subject is based in
social and scientific discourses established in conjunction with a new conception of
the criminal body in the nineteenth century. I argue that a central feature of the
modem criminal is his ability to be represented visually and I locate photography as a
premier mode of representation. Developments in photomechanical reproduction
during the nineteenth century allowed for the relatively quick and easy documentation
of bodies. This was essential to law enforcement agencies that faced increasing
numbers of increasingly mobile criminals.
This chapter is not an exhaustive history of visual representation within criminal
identification. Instead, by re-investigating the key issues and ideas and by focusing
on the work of Lombroso, I emphasize the centrality of photography in the
construction of the modem criminal body. I agree with Simon Cole’s contention that
“criminal identification techniques were crucial to the constmction o f “the criminal”
and “the recidivist”.”6 I stress that, as it developed for use within law enforcement
and criminal identification practices, the camera originally served as a tool of
representation. However, by the close of the nineteenth century Alphonse Bertillon
6 Simon Ablon Cole, “Manufacturing Identity: A History of Criminal Identification

Techniques From Photography Through Fingerprinting” (Ph.D. diss., Cornell, 1998), 28-29.
23
was successful in mobilizing the image through a system of classification. The
photograph as static representation gave way to the image as inscription.
Within work addressing modem penal and judicial systems, Michel Foucault’s
Discipline and Punish remains paramount.7 In this work Foucault examines the
growth o f a new disciplinary society in the late eighteenth and nineteenth centuries.
The new system was based in a normalizing discourse that sought to reform, and not
harm, the physical body and, by extension, the soul. This system operated through
the continued production of knowledge of the human body, and the exercise of power
based on that knowledge. David Green summarizes Foucault’s project:
The emergence of this new form of power is traced through the
changes which occur in the concept of punishment from a system of
retribution to a system of reform. It involves the gradual replacement
of types of punishment which are spectacular, ritualistic and
manifestly violent and in which the body of the offender was the direct
and immediate target of retribution, by a more modest, discrete and
subtle means o f correction and training whose target was no longer the
body of the offender but his ‘soul,’ understood in terms of subjectivity,
consciousness and personality.8
7 Michel Foucault, Discipline and Punish: the Birth o f the Prison, trams. Alan Sheridan (New
York: Vintage Books, 1979).
8 David Green, “On Foucault: Disciplinary Power and Photography,” Camerawork 32 (1985),
7. It is interesting to note that the recent execution o f Timothy McVeigh easily fulfills the criterion of
being “spectacular, ritualistic and manifestly violent.”
24
As Green’s remarks show, an essential feature of the new disciplinary society was
the way in which power, in the form of discipline and punishment, was exercised. Its
application had to go unnoticed and unchallenged to be effective. Paralleling the
transformation in punishment, power had to move from the public into the private
domain. As Foucault has shown this was accomplished through the growth o f the
institution. Through the subtle workings of what he has called the carceral network
(the body of institutions including prisons, schools, hospitals and reformatories) a
new modality of power arose, a “micro-physics” of power. This new modality
existed at the level of the everyday and in the form of continued surveillance, an
essential component of which was the camera. As a technology that was easily allied
with the sciences, the camera provided the perfect tool for the documentation,
classification and regulation of the body.
Working from the ideas set forth in Discipline and Punish, John Tagg and Allan
Sekula have examined the photograph’s role in Foucault’s carceral network.9 Tagg
highlights economic and political practices as they informed the creation,
9 Other sources that parallel and/or summarize the discussion of Tagg and Sekula include:
David Green, “Classified Subjects,” Ten. 8, no. 14 (1984): 30-37; David Green, “On Foucault:
Disciplinary Power and Photography,” Camerawork 32 (1985): 6-9; David Green, “Veins of
Resemblance: Photography and Eugenics,” The Oxford Art Journal 7, no. 2 (1985): 3-16; Tom
Gunning, “Tracing the Individual Body: Photography, Detectives, and Early Cinema,” in Cinema and
the Invention o f Modern Life, ed. Leo Charney and Vanessa R. Schwartz (Berkeley: University of
California, 1995), 15-45; Peter Hamilton and Roger Hargreaves, The Beautiful and the Damned: The
Creation o f Identity in Nineteenth Century Photography (London: Lund Humphries, 2001); Suren
Lalvani, Photography, Vision, and the Production o f Modem Bodies (Albany: State University o f New
York, 1996); Sandra S. Phillips, “Identifying the Criminal,” in Police Pictures: The Photograph as
Evidence, ed. Sandra S. Phillips, Mark Haworth-Booth and Carol Squires (San Francisco: San
Francisco Museum of Modem Art, Chronicle Books, 1997), 11-33. For a discussion of the role of
photography within the asylum and psychiatric hospital see: Sander L. Gilman, Hugh Welch Diamond,
and John Conolly, The Face o f Madness: Hugh W. Diamond and the Origin o f Psychiatric
Photography (New York: Brunner/Mazel, 1976).
25
dissemination and interpretation of photographs within disciplinary institutions.
Sekula places less emphasis on the photograph itself by highlighting its role within a
larger “bureaucratic-clerical-statistical s y s t e m . o r , in his now famous term, an
archive.10 He gives equal weight to discussions on physiognomy, phrenology, social
statistics, anthropometry and the then emerging fields o f eugenics and criminal
anthropology.
My treatment of photography in this chapter and in the dissertation follows more
closely that of Sekula than Tagg. The conception of the archive is particularly helpful
in understanding the ways in which the production and interpretation of photographs
function in the construction of identity. Sekula begins "The Body and the Archive"
by discussing the double system of representation that characterized photographic
practice in the nineteenth century. Under this system the photograph was both
honorific and repressive. He uses the example of photographic portraiture to explain
that
Every portrait implicitly took its place within a social and moral
hierarchy. The private moment of sentimental individuation, the look
at the frozen gaze-of-the-loved-one, was shadowed by two other more
public looks: a look up, at one's "betters," and a look down, at one's
“inferiors.”11
10 Sekula, “The Body,” 16.
26
In this way the photograph served a classificatory function. It provided a static,
visual means to identify oneself among others in society and, in turn, to be identified
from others.
The double role o f the image points to a key feature in the conception of the
archive. For Sekula, the archive exists as a complicated visual arena in which all
groups o f society co-exist, simultaneously reinforcing their mutual exclusivity. This
is how the photographic portrait is at once both honorific and repressive.12 By
emphasizing this double role of the image Sekula is able to show that in the
nineteenth-century construction of a deviant or criminal body, a more generalized
social body was also being produced. Within the visual space of the archive deviance
and normalcy helped to define one another.
Sekula locates the archive’s origin in a single, dominant “hermeneutic paradigm”
that had developed by the mid-nineteenth century.13 He locates work in physiognomy
by Johan Kaspar Lavater and work in phrenology by Franz-Joseph Gall at the base of
this paradigm. As he writes, “both [practices] shared the belief that the surface of the
body, and especially the face and head, bore the outwards signs of inner character.”14
Under this paradigm, the portrait, including the mug shot, took on specific
importance: it stood in for the body and character of the individual it represented.
12 This is similar in concept to Tagg’s notion of the “burden” o f representation.
14 Sekula, “The Body,” 11; Johan Caspar Lavater, Essays on Physiognomy, Designed to
Promote the Knowledge and the Love o f Mankind (London: T. Bensley, 1810).
27
The interest in the external signs of the body was paralleled by an emerging
interest in the field of social statistics and its application to crime. Here, the work of
Adolphe Quetelet is central, as he was the first to study crime and criminality through
the use o f statistical models. In The Rise o f Statistical Thinking: 1820-1900,
Theodore M. Porter puts forth a detailed account of the emergence of this new social
science as it was tied to practices of social policy.15 Porter begins the discussion in
the seventeenth century and practices of political arithmetic where records of
populations and birth and death rates were compiled in support of a society’s self
regulation. Throughout the nineteenth century a greater number of socio-economic
events became subject to statistical analysis and, by the late nineteenth to early
twentieth centuries, our current conception o f the field had developed. Where
Quetelet had the strongest impact was in the application of this emerging social
science to the study of crime.16
Quetelet’s work on crime began in response to studies produced by the French
government during the 1820s.17 These studies documented criminal activity and
showed that it had little variance from year to year. This meant, among other things,
15 Theodore M. Porter, The Rise o f Statistical Thinking: 1820 -1900 (Princeton: Princeton
University Press, 1986). Porter does not draw on the work of Foucault but it is clear that, through his
account, the development of social statistics could be read in support of Foucault’s major claims
regarding the growth of the institution in society.
16 Porter is considerably less complimentary o f Quetelet than is Sekula, stating “[his]

importance as a social scientist seems modest, for his social physics was and remained completely
unworkable” (54). He does however acknowledge Quetelet’s contribution as being more abstract, as in
his fight for statistics as a valid scientific practice and in his application of statistics to the study of
human activity.
17 Porter, Statistical Thinking, 49.
28
that if its causes could be illuminated, crime could be reduced. Quetelet began to
apply statistical methods to the study of crime toward this goal.
Quetelet’s work was unique in that he came to the subject from astronomy and
promoted a single statistical model, based in mathematical probability, to be used
regardless of context.18 As Porter shows, Quetelet’s work was rooted in error theory;
he believed in a natural order to the universe in which man, through conscious action,
created deviations. His task in studying deviations from the natural order was “to
bring order to apparent social chaos.”19
Quetelet’s crowning achievement was his conception of “I’homme moyen” or
“average man.” Using data collected on both the physical and moral characteristics
of man, he plotted a statistical distribution in the form of a bell curve.20 Sekula
summarizes the result:
In an extraordinary metaphoric conflation o f individual difference with
mathematical error, Quetelet defined the central portion of the curve,
that large number of measurements clustered around the mean, as a
zone of normality. Divergent measurements tended toward darker
regions of monstrosity and biosocial pathology.
18Porter, Statistical Thinking, 44.
19Porter, Statistical Thinking, 42.
20 Porter takes up the issue of the difficulty in measuring the moral character o f man on page
53.
29
It is important to reiterate that Quetelet’s work presupposed a natural order to the
universe. Thus, as Sekula describes the metaphor, human difference, both physical
and moral, was perceived in terms of a deviation from a natural average. However,
what is absent in Sekula's account and what must be stressed here is that Quetelet
treated deviation as both positive and negative. Deviation, or change, caused by the
conscious decisions of man could be productive if it was of benefit to society, if it
contributed to what Quetelet conceived of as man’s natural, evolutionary

99
progression. To return to the distribution, there exists, on one side, deviations that
stunt progress and, on the other, deviations that promote it. In the middle rests the
majority of actions (and the citizens who produce them), neither harming nor helping
progress. It is this feature of Quetelet’s work that is of most importance in relation to
Lombroso’s theories on crime.
Through the creation of I ’homme moyen, Quetelet provided a statistical basis for
understanding criminal activity. He accomplished this by substituting analysis of the
individual for that o f the group. As Porter notes:
Quetelet maintained that this abstract being [the average man], defined
in terms of all human attributes in a given country, could be treated as
the “type” of the nation, the representative of a society in social

93
science comparable to the center of gravity in physics.
30
Working according to data believed to be representative of social groups, Quetelet
was able to provide scientific legitimacy to existing social hierarchies. Criminals
were now shown to be unproductive and in fact, counter-productive to what was
understood as the natural progress of society. In opposition to this, and across the
ocean of average citizens, individuals such as Quetelet worked toward the progression
of mankind. Quetelet’s work fit perfectly within an emerging evolutionary discourse
that privileged certain bodies over others.24
Quetelet’s work is important to this discussion in four ways. First, his distribution
provided a way to explain human behavior by dividing it into a binary of normal and
deviant. Second, and as a corollary to this, the social scientist using Quetelet’s work
gained a distinct level of authority and control over his subject matter (the social
world).25 Third, because his work was based in mathematics and error theory,
deviance was taken to be both positive and negative. In this way Quetelet defined
three groups of bodies: counter-productive, average, and productive. Finally, as it
was based in mathematics, Quetelet’s work had scientific legitimacy, enabling it to be
allied with other scientific practices including the newly emerging field of criminal
anthropology.26
24 Darwin's Origin o f the Species was published in 1859 and Descent o f Man in 1871.
25 David Horn shows the extent to which Lombroso did the same through his work on female
criminality. See: David G. Horn, “This Norm.”
26 Theodore M. Porter provides an excellent discussion of the relationship between objectivity

and quantification in scientific practice (through the use o f numbers). See: Trust in Numbers: The
Pursuit of Objectivity in Science and Public Life (Princeton: Princeton University Press, 1995).
31
The importance of inscription underlies all of the above features. By substituting
the individual with the group and by reducing human behavior to a sequence of
numbers, Quetelet attempted to demonstrate, in the form o f a single, static and two-
dimensional bell curve, the aggregate behavior and characteristics of complete
societies. Lombroso would combine this work with the theoretical claims of
physiognomy and phrenology in the development of criminal anthropology. The
result was an individual and social body that were interchangeable and that were
readily understandable through the interpretation of their external, bodily signs.
Lombroso and the Criminal’s Body
In his introduction to the English translation o f Lombroso's Crime: Its Causes and
Remedies, Maurice Parmelee claims that Lombroso "led in the great movement
towards the application of the positive, inductive methods of modem science to the
problem of c r i m e . a n d that he "stimulated, more than any other man, the
development of the new science of criminology."27 Prior to Lombroso, and under the
direction of the Italian criminologist Cesare Beccaria, the Classical School of
criminology reacted against the then overarching discretionary power of the judge in
cases of criminal justice. Beccaria and the Classical School called for the equal
treatment of all under the law, and suggested that punishments should be matched to
the crime committed.
27 Lombroso, Crime, xxxii.
32
Where the Classical School emphasized the universal application of the law,
Lombroso, with his contemporaries Enrico Ferri and Raffaele Garofalo, detailed a
system in which punishment was based more on the individual character of the
offender than on the specific crime committed. Lombroso called it a shift in focus
from the study o f crime to the study of the criminal body.28 Under the leadership of
Lombroso, these three formed the Italian School of criminology, also called the
Modem, Positivist and Anthropological School, and sought to identify a distinct
anthropological criminal type.29 Crucially, through outlining different types of
criminal behaviour, this group also promoted the ability to reform certain types of
offenders. Under the Classical School, criminals were incarcerated for a definite and
pre-determined period of time. Under the Modem School, criminals were often given
an indefinite sentence, based on their specific anthropological type and their response
to treatment, or reform.
The central text to emerge from the Italian School was Lombroso's L'Uomo
Delinquente, first published in 1876. Here Lombroso outlined his theory of
criminality; defining his notion of the "bom criminal" and arguing that criminal
behaviour was essentially atavistic. Crime was something that certain bodies were
predisposed to, social influence was almost exclusively denied. Leonard Savitz
summarizes Lombroso's bom criminal noting that, "physically, emotionally and
28 Lombroso-Ferrero, Criminal Man.
29 Lombroso-Ferrero, Criminal Man.
33
behaviouraliy, he was very similar to primitive races and to children, all of whom
were hedonistic, non-intellectual, curious, cruel and cowardly."30
This idea met with considerable criticism and, over the next several years,
Lombroso altered his theory to allow for greater social influence in criminal
behavior.31 By the time Crime: Its Causes and Remedies was published (originally
published in Paris in 1899 and translated into English by 1911) Lombroso had
constructed an elaborate theory taking account of various "types" of criminality. In
addition to the bom criminal, Lombroso defined the passionate, political, accidental
and occasional criminal, often with several sub-categories to each. Remaining true to
the original positivist goal, punishment, or treatment, was linked to the type of
criminal rather than to the crime committed.
What emerged from Lombroso's work was a statistical distribution of criminality
remarkably similar to Quetelet’s I ’homme moyen. To the left was the bom criminal, a
habitual offender who was biologically predisposed to crime. In opposition to this,
the accidental or occasional criminal was an otherwise "normal" citizen who, in a
random and isolated moment, committed a criminal act. The bom criminal was
beyond reform. He existed in a repressed state, parallel to that of children, savages
and animals. The potential for reform increased as one moved further away from this
pole. Nicole Hahn Rafter describes the hierarchical order:
MLombroso-Ferrero, Criminal Man, xi.
31 In opposition to the Italian School, Gabriel Tarde and Alexandre Lacassagne formed the
French School of criminal anthropology, attributing primary significance to social causes. Followers
of the French School were among the largest opponents o f Lombroso. The differences between
schools can be found in the criminology sources outlined in footnote 5.
34
At the bottom of the scale is the bom criminal, rough in appearance
and manners, a foreigner or Negro uneducated, of poor
background, a drinker. At the top stands ... [the] gentlemanly normal
offender, anomaly free, a product of not heredity but environment,
intelligent and skilled, conscience stricken and refomaable.32
Figure 2.2: Two examples of the visibly deviant body used as illustrations in
Lombroso’s work. “Italian Criminal. A Case of Alcoholism,” (L) and “Anton Otto
Krauser, Apache” (R). From Lombroso-Ferrero, Criminal Man.
32 Rafter, “Criminal Anthropology,” 539. A chart detailing Lombroso's scheme matching

punishment to criminal type can be found in Lombroso-Ferrero, Criminal Man, 210-212.
35
Rafter's reference in the above passage to the "anomaly free" offender is
important. In reacting to the Classical School of criminology, Lombroso not only
established a new pseudo-scientific method in the study of criminal behavior; he also
established a new kind of criminal. With Lombroso's positive, inductive method,
criminal behavior could be traced to, and read from, the body. In Lombroso's theory,
the number of "signs" of criminality (tattoos, cranial and bodily anomalies, and other
distinguishing physical marks) was directly related to the severity of the case (Figure
2.2). In other words, those criminals who displayed several and obvious signs of
criminality were likely to be biologically predisposed to crime (the bom criminal),
were likely to be habitual offenders, and were less amenable to reform. Conversely,
those subjects who displayed few or no signs were increasingly less likely to exhibit
criminal behaviour and had the highest potential for reform. Criminal behaviour was
reduced to a simple, positivist correlation between physical, bodily "signs" and the
propensity to commit crime. Within this scheme the scarred and tattooed body of
Anton Otto Krauser, depicted in Figure 2.2, represented a body more prone to
recidivism and less likely to be reformed than that o f the alcoholic, also depicted in
Figure 2.2.
Rafter has noted that the biologization of crime by Lombroso and his
contemporaries was a feature fundamental in their claim to scientific legitimacy.
Discussing the originators of criminal anthropology in Europe she writes:
[They] all considered their work “scientific,” by which they meant that
unlike earlier commentators on crime, who had included God and free
36
will in the causational picture, they would be materialistic, examining
only phenomena anchored in the natural world of matter.33
Toward this goal criminal anthropologists set out measuring and recording data on
crime in a way reminiscent of Quetelet. Unlike the statistician, criminal
anthropologists turned to the criminal body for their data; physical, bodily anomalies
were taken as proof of criminal disposition. Commenting on Lombroso’s work,
Rafter notes that, “nature had made the investigator’s task relatively simple: To detect
bom criminals, one needed only the appropriate apparatus.”34
Developing and employing the “appropriate apparatus” became a central part of
Lombroso’s work. He borrowed, altered and used an array of instruments in
measuring the “signs” of criminality. To measure the level of affection in subjects
(whether or not they have feelings toward others) he tested their blood pressure using
the plethysmograph and the hydrosphygmograph invented by Francis Frank. The
phonograph would measure speech patterns and Edison’s electric pen could be used
for handwriting analysis. Anfonsi’s tachyanthropometer and craniograph could
measure the subject’s height and arm length and the form and capacity of the skull.
Tactile sensibility was measured with Weber’s esthesiometer. Du Bois-Reymond’s
algometer measured general sensibility and sensitivity to pain and Nothnagel’s
thermo-esthesiometer measured sensibility to temperature. Vision was tested by
using Landolt’s apparatus; sense of smell was measured with Ottolenghi’s
33 Rafter, “Criminal Anthropology,” 534.
34 Rafter, “Criminal Anthropology,” 535.
37
osmometer; and strength with the dynamometer.35 Using these tools and others
Lombroso filled his texts with representations and inscriptions: tables, charts,
statistics, photographs, and drawings, all attesting to the criminal nature of the body
(Figure 2.3).
Figure 2.3: Tools used by Lombroso for measuring the “signs” o f criminality. The
algometer (L) for measuring “general sensibility and sensitivity to pain,” and the
campimeter (R) for measuring field of vision. Both illustrated in Lombroso-Ferrero,
Criminal Man.
That Lombroso’s “science” was problematic is clear. Like Quetelet’s, Lombroso’s
work on the criminal was tied directly to evolutionary and, later, eugenicist practices
which privileged certain types of bodies. The bom criminal had parallels to
“savages”and “lower races” and was clearly bound to racial hierarchies.36 The female
35 A more detailed discussion of these instruments, including illustrations, can be found in the
first chapter of Part Three of Lombroso-Ferrero, Criminal Man.
36 Rafter, “Criminal Anthropology.”
38
was continually defined as a threat or benefit to society in terms of procreation. Her
existence as prostitute was so heinous as to draw comparison to the male bom
criminal.38 Throughout his work, Lombroso continually defined the deviant body in
opposition to its white, male counterpart.
A chief result of the work of Lombroso and contemporary criminologists was to
provide a scientific basis for existing theories on society and social structure. Using
the statistical methods of Quetelet and the theories of physiognomy and phrenology,
Lombroso wrote criminality onto the body. Commenting on Lombroso’s
methodology, Rafter notes that his criteria at first seem “putatively derived from
biology” but that they “are in fact derived from social class and then attributed back
to biology.”39 This is nowhere more clear than in Lombroso’s treatment of the female
criminal.
As their titles suggest, Quetelet’s / ’homme moyen and Lombroso’s L ’Uomo
Delinquente concerned themselves with the study of male criminality. Lombroso also
produced work extensively on the female criminal, culminating in his 1895 La Donna
Delinquentz, or Female Offender, co-authored by William Ferrero.40 In this text
j7 Horn, “This Norm.”
38 Summarizing his study of female criminality, Lombroso writes: “the true bom criminal
exists among women only in the form o f the prostitute, who already finds in her lamentable calling a
substitute for crime,” Crime, 406. On the pathologization of the female body and, in particular, the
prostitute see: Sander L. Gilman, Difference and Pathology: Stereotypes o f Sexuality, Race, and
Madness (Ithaca: Cornell University Press, 1985).
j9 Rafter, “Criminal Anthropology,” 539.
40 Lombroso, The Female Offender.
39
Lombroso again turns to detailed written and photographic accounts of physical
bodies as the material source for his theoretical endeavors (Figure 2.4). The basis for
this text was a study of twenty-six skulls and five skeletons from prostitutes as well as
case studies of seventy-seven criminals who died at prisons in Turin and in Rome.
The text relies heavily on visual representation and inscription. It is filled with charts,
diagrams, statistics and photographs, amalgamated into case studies detailing the
physical anomalies o f female criminal subjects.
David Horn has produced a detailed and insightful reading of Lombroso’s work on
female criminality.41 He notes that an early problem facing Lombroso was that
female criminal bodies displayed fewer anomalies than their male counterparts. As
this did not support his existing theory, Lombroso had to reformulate it and account
for the new findings.42 Lombroso accomplished this, Horn suggests, by returning to a
concept o f the statistical distribution. He explained the lack of difference and
anomalies in female bodies by claiming that diversity among species was an asset
(even in the form o f physical signs of criminal propensity), thus reinforcing the
primacy of the male body. Horn explains, “the relative rarity of signs of degeneration
on the female body, which had seemed potentially a sign of women’s superiority,
could now be linked to a lesser degree of variability, itself a sign of inferiority and
weakness.”43
41 Horn, “This Norm.”
42 This is not specific to Female Offender. Lombroso’s many texts on male criminality can all
be read as continued re-workings of his original theory on the bom criminal.
43 Horn, “This Norm,” 117.
40
Figure 2.4: “Physiognomy o f Russian Female Offenders,” from Lombroso, Female
Offender. One o f the many plates in the text that served as “data.”
In both Female Offender and Crime: Its Causes and Remedies the female criminal
is continually described in a way that positions her as inferior to her male counterpart.
Females neither match the savagery or anim alistic nature of the male bom criminal
41
(with the exceptional case of the prostitute), nor are they capable of the more
intellectual and strategic crimes of the male accidental criminal. In other words,
female criminality falls at the median of the classificatory and reformative curve. By
using the same statistical and pseudo-scientific method of L ’Uomo Delinquente,
Lombroso defined the female criminal, and with it the female body, as ineffectual.
Whether studying the male or female body, Lombroso relied on the body itself for
his data. As Rafter has suggested these visual “signs” were often read, or discovered,
in support of existing social hierarchies. The result was a new conception of the
criminal subject. Non-white, non-Western, sexually deviant and animalistic, his or
her criminality was readily apparent on the skin. This new, visibly criminal body,
and Lombroso’s reverse, biologizing method is summarized in what Lombroso
describes as a pivotal case in his work. After being called to perform an autopsy on
the Italian criminal Vilella, Lombroso opens the skull and finds:
The problem o f the nature of the criminal - an atavistic being who
reproduces in his person the ferocious instincts of primitive humanity
and the inferior animals. Thus were explained anatomically the
enormous jaws, high cheek-bones, prominent supercilliary arches,
solitary lines in the palms, extreme size of the orbits, handle-shaped or
sessile ears found in criminals, savages, and apes, insensibility to pain,
extremely acute sight, tattooing, excessive idleness, love of orgies, and
the irresistible craving for evil for its own sake, the desire not only to
42
extinguish life in the victim, but to mutilate the corpse, tear its flesh,
and drink its blood.44
Photo-mechanical Reproduction in Law Enforcement
The development of the modem criminal body, with its visual “signs” of deviance,
was contemporaneous with rapidly expanding innovations in photographic
technology. Photo-mechanical reproduction developed over a half-century from the
experiments in the 1820s by Joseph Nicephore Niepce and Jacques Louis Mande
Daguerre with the camera obscura to the 1888 release of the first Kodak camera,
giving rise to what Tagg has called “the era of throwaway images.”45 The
development of the half-tone process in the 1880s allowed for “economical and
limitless reproduction of photographs in books, magazines, advertisements, and
especially news papers.”46 By the closing years of the nineteenth century, nearly
anyone could, and many did, produce photographs.
Police departments, penal, medical and other social institutions quickly seized the
benefits of the camera as a relatively quick and easy method of documenting
individuals. Within this context, and as Tagg, Sekula and other historians of
photography have shown, the camera was co-opted to capture the physical
pathologies o f the body toward evolutionary, eugenicist, and medico-legal goals. The
44 Lombroso-Ferrero, Criminal Man, xxv.
43
camera allowed for the study of large numbers of bodies and provided a text, which
could not resist its interpretation. The intersections of photographic technology and
the new conception o f the criminal culminated in the work o f Francis Galton and his
photographic composites. Perhaps more than any of his contemporaries, Galton
promoted the tremendous potential in utilizing photographic technology in studying
the human body.
Galton continually worked on and tested various theories of evolution but is most
famous for his work on eugenics. He defined and introduced this topic in his 1883
Inquiries into Human Faculty.*1 In the introduction to the text he summarizes his
project:
My general object has been to take note of the varied hereditary faculties of
different men, and of the great differences in different families and races, to
learn how far history may have shown the practicability of supplanting
Inefficient human stock by better strains, and to consider whether it might not
be our duty to do so by such efforts as may be reasonable.48
Like Lombroso, Galton treated the body as a series of signs to be examined and
interpreted. To this end, he traced heredity through the visible signs of the body. By
producing composite images of faces and bodies he would make general
pronouncements on race, gender, sexuality, and criminality. Further, based on his
47 Francis Galton, Inquiries Into Human Faculty and its Development (London: J.M. Dent and
Sons, 1907).
48 Galton, Inquiries, 1.
44
findings, and through the eugenic practice of controlled breeding, society could rid
itself of its lowest forms, that of the “residuum.
Figure 2.5: Galton composite photographs showing the “criminal type,” c. 1870s.
From Philips, Police Pictures,
Figure 2.6 Galton composite photographs showing the “Jewish type,” c. 1870s. From
Phillips, Police Pictures.
45
The composite image is one of multiple exposures on a single frame of film
(Figures 2.5 & 2.6). Gallon’s first composites were made from photographs he
received from Pentonville and Millbank prisons. Sir Edmund Du Cane, then H.M.
Inspector o f Prisons, asked Galton to examine the photos and see if he could identify
features specific to certain types of criminality.49 Galton divided the photos into three
classes: those convicted of murder and manslaughter, those convicted of felonies and
those convicted of sexual offences.50 Galton then took composite photos of the
individual classes by dividing total exposure length by the number of cases. If an
exposure was to last ten seconds and ten criminals were to be photographed then each
criminal would receive 1/10* or one second of exposure. Galton defines the rationale
behind the process as follows:
to bring into evidence all the traits in which there is agreement, and to leave
but a ghost of a trace of individual peculiarities. There are so many traits in
common in all faces that the composite picture when made from many
components is far from being a blur; it has altogether the look of an ideal
composition.51
As they were part of studies of heredity, Galton produced composites of both
“good” and “bad” stock. He created composites of Alexander the Great (using coins
depicting him), sisters, brothers, families (across sexes), horses (to show the heredity
49 Francis Galton, Memories o f My Life (London: Methuen, 1908), 259.
50 Green, “Veins,” 11.
51 Galton, Inquiries, 7.
46
of successful race horses), criminals (divided according to crime), persons suffering
from consumptive disorders, phthisical patients, Westminster schoolboys, lunatics,
members o f the Royal Engineers, members of the Jewish race, and others. Under
Gabon’s theory of eugenics and its link to inheritance any and all bodies were subject
to composite portraits. In the case of Alexander the Great the composite was created
to show the ultimate height of human kind, this was then contrasted with other
composites to show those groups that might be avoided through controlled breeding.
Gabon’s composite images stand as the premier example of nineteenth-century
scientific and social scientific activity that sought to identify and understand the
deviant body through its visual representation. His composites were used by
Lombroso for illustration in the 1895 French and 1896-1897 Italian editions of
Criminal Man.52 They were also used for the cover of the 1890 text, The Criminal,
written by a follower of Lombroso’s, Havelock Ellis.53 Gabon’s composites can be
seen as a photographic manifestation of Quetelet’s statistical distribution and
Lombroso’s theory on crime. Large numbers of bodies were quickly reduced to
aggregate ‘types’ and stood to represent entire populations.
From its inception, the photograph’s potential for documenting criminals was
acknowledged by law enforcement agencies. As early as 1841 the French police
53 Havelock Ellis, The Criminal (London: Walter Scott, 1890). Nicole Hahn Rafter identifies
Ellis’ text as a key entry point for Americans to the Italian School of criminal anthropology. She calls
Ellis "far and away the most influential" of the scholars responsible for exposing the work o f the
originators, including Lombroso, to the United States. “Criminal Anthropology,” 530.
47
began producing daguerreotypes of prisoners.54 In the United States, the San
Francisco Police department began daguerreotyping prisoners in 1854, followed by
departments in New York in 1858, Cleveland in the late 1860s and Chicago in 1870.5:>
And in Canada, the Toronto Police collected photographs of criminals beginning in
1874.56 Early articles discussing the topic appeared in the 1872 and 1875 editions of
The British Journal Photographic Almanac by the noted photographers, Oscar G.
Rej lander and W.B. Woodbury respectively.57 By the last decades of the nineteenth
century police departments in major metropolitan areas across the world were
collecting photographs of criminals.
Describing the use of daguerreotypes by the San Francisco Police, Harris Tuttle
writes: “the pictures were mounted in leather books, five to six hundred per book.
Supplementing these photo albums were record books with brief criminal histories
and descriptions of the criminals.”58 The resultant albums, called “rogue’s galleries,”
were collected by major police departments across the world. As crime rates rose, the
galleries grew in size and, as such, presented a central problem for police. Use of the
galleries was dependent on an officer’s ability to memorize the faces, names and
54 Harris B. Tuttle, “History of Photography in Law Enforcement,” Finger Print and

Identification Magazine 43, no. 4 (1961): 3-28,4.
53 Cole, “Manufacturing,” 51; Phillips, “Identifying the Criminal,” 19; Tuttle, “History,” 16.
56 H.S. Cooper, “The Evolution of Canadian Police,” in The Police Function in Canada, ed.
William T. McGrath and Michael P. Mitchell (Toronto: Methuen, 1981), 37-52.
57 O. G. Rej lander, “Hints Concerning the Photographing of Criminals,” The British Journal
Photographic Almanac (1872): 116-117; W. B. Woodbury, “Photographing Criminals,” The British
Journal Photographic Almanac (1875): 128.
58 Tuttle, “History,” 16.
48
information compiled in the books. To exemplify the problem, Berlin’s gallery grew
to such an extent that, by 1877, it was subdivided into ten crime-based groups and
required three attendants. In 1880 the Berlin gallery housed 2135 images and within
three years this had increased to 3459.59
One of the most famous of the rogue’s galleries, and one which highlights the
problematic nature of their use, is Thomas Byrnes’ 1886 Professional Criminals o f
America .60 Byrnes, then Chief Detective of the New York Police, created a text with
204 frontal portraits, or mug shots, of criminals divided six per page. As with other
galleries, the images are interspersed throughout text which describes each offender
including his or her name, alias, biographical and physical description and criminal
history. Figures 2.7 and 2.8 show typical pages of Byrnes’ book. The entry on James
White, #94, reads as follows:
James White, alias Pop White, alias Allen, alias Doctor Long.
Hotel Thief and Confidence Man
Description:
Seventy years old in 1886. Bom in Delaware. Painter by trade. Very
slim. Single. Height, 5 feet 8 1/2 inches. Weight, about 135 pounds.
Gray hair, dark-blue eyes, sallow complexion, very wrinkled face.
Looks like a well-to-do farmer.
59 Tuttle, “History,” 8.
60 Thomas Byrnes, Professional Criminals o f America (New York: Cassell and Company,
1886).
49
Record:
OLD POP WHITE, or “DOC” LONG, is the oldest criminal in his
line in America. Over one-third o f his life has been in State prisons
and penitentiaries. He has turned his hand to almost everything, from
stealing a pair of shoes to fifty thousand dollars. He was well known
when younger as a clever bank sneak, hotel man and confidence
worker. He is an old man now, and most of his early companions are
dead. He worked along the river fronts of New York and Boston for
years, with George, alias “Kid” Affleck (56), and old “Hod” Bacon,
and was arrested time and time again. One of their victims, whom
they robbed in the Pennsylvania Railroad depot at Philadelphia in 1883
of $7,000, died of grief shortly after.
Old White was discharged from Trenton, N.J., State prison on
December 19, 1885, after serving a term for grand larceny. He was
arrested again in New York City the day after for stealing a pair of
shoes from a store. He pleaded guilty, and was sentenced to five
months in the penitentiary on Blackwell’s Island, in the Court of
Special Sessions, on December 22, 1885.
Pop White’s picture is a good one, taken in July, 1875.
ja s s s s g a m y . c h a k ls s h a so s, n n n me,
*U *S <SG j l i i CA LtY , A..IAS i?OSTOS> f.M W .C .Y LM iKO C E N T S * . T T " C
8AM CC A fiD P s C K P O C T B ', P IC K P O C K E T AMD 8A N C 0. 8A N C 0.
94 95 96
j'a?j£$ Alien, ./£?sfw lewis, m um johnsoh,

ASJAS -'T'r YAHijrE aaT i.OPG. :■ JO PICKPOCKET.
HO TEL *N 0 C O N F ID E N C E W AN. BANCO.
Figure 2.7: A page of images depicting male criminals from Byrnes’ Professional
Criminals.
51
121 It# 129
ANNIE REILLY, S O P H IE LYONS, K A T E RY AN,

ALiA?» ANAiE, ALUS LEW , PICKPOCKET,
DISHONEST SERVANT P IC K P O C K E T A N D B L A C K M A IL E R .
13® 131 132
A N N I E MACK, LOUISA JOURDAN, 0,4T H A R 'lr E ARMSTROfVG,

AUAS U V TU . LOU'Sc., AiJAS MAftV ANN DO'-VD — O i'J .Q N .
sw eak and sh o p lif te r P IC K P O C K E T AN D S H O P L IF T E R . PICKPOCKET.
Figure 2.8: A page o f images depicting female criminals from Byrnes’ Professional
Criminals.
52
In the preface of the text Bymes notes that his mission is to make this rogue’s
gallery available for public knowledge and use. He writes:
While the photographs o f burglars, forgers, sneak thieves, and robbers
of lesser degree are kept in police albums, many offenders are still able
to operate successfully. But with their likenesses within reach of all,
their vocation would soon become risky and unprofitable.61
This statement points to Byrnes’ attempt to address the central difficulty of rogue’s
galleries. The ease of documenting the body announced by developments in photo
mechanical reproduction enabled police departments to amass tremendous sums of
mug shots. However, as these repositories of images grew, their utility became
increasingly problematic. Memorizing the 204 offenders in Byrnes’ gallery, let alone
the 3459 in the Berlin gallery, would be impossible for even the best police officers.
Because o f this Bymes attempted to enlist the public in practices of criminal
identification. If the public could familiarize themselves with the faces in the gallery,
they could assist police in identifying the criminals among them.
Another attempt to address the shortcomings of rogue’s galleries was developed in
the work of Alphonse Bertillon.62 After taking a job with the Paris Prefecture of
Police in 1879, Bertillon identified that the central problem plaguing the department’s
identification practices was the vast number of photographs in its collection. More
specifically, the department had accumulated tremendous sums of images but lacked
61 Byrnes, Professional, Preface.
62 For a thorough account of Bertillon’s work, see: Henry T. F. Rhodes, Alphonse Bertillon,
Father o f Scientific Detection (London: Harrap, 1956).
53
any real way to utilize them. To alleviate this problem, Bertillon developed his
signaletic system of identification.63
Bertillon’s system involved recording specific identification data onto a
standardized form. Law enforcement officers would first record detailed
anthropometric measurements of eleven body parts: height, head length, head
breadth, arm span, sitting height, left middle finger length, left little finger length, left
foot length, left forearm length, right ear length and cheek width (Figure 2.9). These
measurements were supplemented with the physical description of the body, but
particularly of the features of the face and head. For this task, Bertillon devised a
short-hand language describing the minute details of the body in tremendous detail
(Figure 2.10). The final aspect of Bertillon’s signaletic system was the recording of
one full face and one profile photograph. The resultant identification card, or portrait
parle, was a comprehensive data-sheet, detailing the criminal subject (Figure 2.11).
63 Bertillon’s methods are detailed through several of his own texts, including: Alphonse
Bertillon, La Photographie Judiciaire avec un Appendice sur la Classification et L'identification
Anthropometriques (Paris: Gauthier-Villars, 1890); Identification Anthropometrique: Instructions
Signaletiques - Album (Melun: Imprimerie Administrative, 1893); Alphonse Bertillon's Instructions
for taking descriptions for the identification of criminals, and others by the means of anthropometric
indications (New York: AMS Press, 1975).
54
Figure 2.9: Measuring the width of the skull for anthropometric data, c. 1900. From
lies, Chrissie, and Russell Roberts, ed. In Visible Light: Photography and
Classification in Art, Science and The Everyday (Oxford: Museum of Modem Art,
1997).
55
TABLEAU SYKOPTIQUE DES TRAITS PHTSIO1O11Q0K

p o u r y e rv if au re io v * 'Jt. 'i >t(l
S‘aw sms*
w .
< w * « u k u a w * w tjg a
_ _ y Q - ^ g ntfffuinv-iwaf
**'"l" ,,.IV,->.>.<■_ f.V

I|
J O I C S
ngasagg
- - J - ’...... “■"'^ ■■■■■■n»rt»i;#1ftiin....■im.wiam.iiii
t'54«t**«fc jpWli4H| * ttism

'u * P S O T f lT O S iSLAJSd <r«H6t«stf &eAt
i#riSIBIa wSHBlB
- w rm m m .
'*■•*ja a ta-ghSS ,: 'iiS'ale w*™* *
r 4 V 9 W N O a V - s j i.M
•*•■' '— s UkP?S*
Oj-. !JI« .it »!'.■ «8fH*4. t»»*»*><.» . •• «S'v»t»
m s m m e k e e js e je u t l K t B
>^W
SwflB»'
H E E ggE g BKBEKKK
I E E ®
. _
B i ^ g a z * i ' ....... ._l

S S f is a is a d 1 — ,
e p iiiK W S S S is s r i c a s a i E s
Figure 2.10: Bertillon’s “synoptic table o f facial expressions.” From Phillips, Police
Pictures.
R ep ro d u ced with p erm ission o f th e copyright ow ner. Further reproduction prohibited without perm ission.
56
af 'JC *r. . V - W V* , ^ fc rTTT rK

v.H>? " *1
ffeitewt ftttew d ’CflBiHal ItetlfieaiicM feW-f
W^WRC*, “&0.
SsiiSlilf dUZ~L ,«**
IB
, . „ ............................. ............................................ Xksm&4.............
rt*v .4ci- f'iW .» /Wj. . . M*f...............:
l^»i........... ai.'-»................... -
SmU*t........................................ J>M.................. M g ’..............
-;—
m
, „, s ....................................
, .............. ^ W .’ - - : » , l X ^ . . . . - « . .. . ~ ........................... ..
aA tt«' $ $M
«*46+*kiwx*>.<$***%*£4iSfi*t
. ^ . sU hI SsLaÂ..2&^„4a^4a*l■:«**. p i |p : p # ^ |l | tolplilill
iSiiiBIII
Vj#.,fe~* 3Pt»*A*i*j, SfevA^.*, rî * gliiMipiiiiiiiiii
>l^sm <■■'■* *^M«w*'...* .................. ......... v—.... - ^
mw Jfflttl sm.il' CCOiHa^Lfl&KWWIJi
Figure 2.11: A Bertillon identification card showing the recording of anthropometric
measurements (top right), physical description (bottom right) and full face and profile
photographs, c. 1908. From Cole, Suspect Identities.
Through the systematic application of anthropometric measurements and detailed
physical descriptions (using the short-hand language devised by Bertillon), the body
could be translated into a series of words and numbers and could be classified
according to its visual signs. Bertillon identification cards were classified by sub
dividing them according to the eleven anthropometric measurements. Each of these
was further subdivided according to the categories “small,” “medium,” and “large.”
The resulting categories o f cards were divided by eye colour and finally filed
according to ear length.64 Searches of the filing system operated according to the
64 Cole, “Manufacturing,” 79.
57
same procedure; eliminating categories of cards until a matching identification card
was found.
The essential difference in Bertillon’s method was that he provided the
classificatory basis necessary for a totalized system of identification. As Simon Cole
suggests, Bertillon was the first to grasp the fundamental concept of criminal
identification: classification. Cole writes,
The portrait parle marked the shift from a visual culture to a literate
one based on specialized language, bureaucratic files, government
records. Bertillonage was a system of translations: the body might be
translated into numbers and words, encoded further into abbreviations,
and finally translated back into a visual image.65
The importance of Bertillon’s classificatory system is acknowledged throughout the
accounts of Cole, Sekula, Tagg, and Henry Rhodes and other authors writing on the
topic. What I want to stress here is the particular effect it had on the photograph.
Bertillon’s classificatory system mobilized the photograph. With anthropometry
and the potrait parle, the image became an inscription. Rather than merely represent a
body as in Bymes’ rogue’s gallery, the photograph could now participate in a system
of visual and textual representations. Combined with the textual data of the
identification card, the image could be indexed, sorted, accessed and exchanged
within and between law enforcement agencies. The potential of Bertillon’s system
particularly over rogue’s galleries, is highlighted by Lombroso. He writes:
58
If a good police commissary in Italy wants to put his hand upon the
unknown author of some crime, he has recourse to his memory, to
photographs, and also to the clumsy criminal register instituted a few
years ago. But in a kingdom as large as Italy, with such rapid means
of communication, thousands of individuals escape observation. The
best memory would be not much help. Delinquents easily success in
eluding the police by changing their names, or, if arrested, give them a
false idea of their antecedents by taking the name of some respectable
person. From this one sees how necessary it is to have means of
identifying accused persons with scientific accuracy; and of all the
systems proposed for this purpose that of Bertillon is undoubtedly the
best.66
Once officially accepted for use by the Paris police in 1883 Bertillon’s signaletic
system was quickly adopted across Europe, Canada and the United States. As has
been addressed in great detail by Simon Cole, and as will be discussed in Chapter
Three, Bertillon’s system was not entirely successful and would eventually be
replaced by fingerprinting as the dominant mode of criminal identification.
Nonetheless, it is important to note that Bertillon’s work did have an impact on the
use of photography in law enforcement and criminal identification practices.
66 Lombroso, Crime, 252.
59
Quetelet’s statistical distribution and Lombroso’s anthropological criminal type
were part of a confluence of scientific and social scientific activity in the nineteenth
century that produced a new conception of the criminal. The physical body of the
criminal was replaced by its representation and inscription in the form of text, curves,
charts and photographs. Indeed the ability to be represented must be seen as the
defining feature of the modem criminal subject. Individual, physical bodies present
innumerable difficulties for law enforcement and criminal identification practices.
The relative ease and inexpense of photography, combined with its believed
objectivity, enabled the substitution of the photograph for the live body. Law
enforcement agencies throughout the world adopted the camera quickly and amassed
large repositories of mug shots. However, as the resultant rogue’s galleries grew the
utility o f these representations decreased. Bertillon’s classificatory system mobilized
the photographic representation of the criminal. The static representation of the
rogue’s gallery became an inscription at work in the production of scientific
knowledge and the construction of a criminal, and with it, a normal body.
60
CHAPTER THREE
PHOTOGRAPHING FINGERPRINTS: DATA, EVIDENCE, AND LATENT
IDENTIFICATION
Speaking at the 1926 International Association of Chiefs of Police convention, I.
Edgar Hoover, then the newly appointed director of the Federal Bureau of
Investigation, announced that the Identification Division had passed the million mark
in its collection of fingerprint records.1 During the fiscal year 2001 the same agency
received 15.4 million fingerprint submissions.2 Currently, the FBI receives
approximately 50,000 fingerprint records a day. Half of these are related to criminal
investigation and half relate to civilian use, such as background checks for
employment purposes.3 The collection of personal information on such a vast scale
raises distinct concerns regarding the control and use of this information and in the
role of law enforcement within public life.
1Donald C. Dilworth, ed., Identification Wanted: Development o f the American Criminal

Identification System 1893 - 1943, Police History Series (Gaithersburg: International Association of
Chiefs of Police, 1977), 166.
2 Thomas Carey, Assistant Special Agent in Charge, Washington Field Office, FBI,
“Communication with the Law Enforcement Community,” Congressional Statement, November 13,
2001. Accessed September 2002, available from
http://www.fbi.gov/congress/congress01/careylll301.htm.
3 Federal Bureau of Investigation, “Inauguration of the Integrated Automated Fingerprint

Identification System (IAFIS),” (Press Release, August 10,1999). Accessed November 2002,
available from http://www.fbi.gov/pressrel/pressrel99/iafis.htm.
61
These issues have taken on tremendous importance following the events of
September 11,2001. Now, within the United States and under the guise of the “war
on terrorism,5’ previously disparate law enforcement databases and information
networks are being combined into a centralized resource of identification information.
Mobilized by discourses centered on notions of fear and national threat, these
heightened communication and identification capabilities are being employed in
overtly discriminatory practices. These include racial profiling and the mandatory
fingerprinting and photographing of visitors to the country from nations deemed a
national threat. The collection, interpretation and control of personal information are
fundamental in this effort and its construction of individual and national identities.
This chapter traces the origin of these practices to the development of fingerprinting
during the turn of the nineteenth and twentieth centuries.
Bertillonage, the system of criminal identification developed by Alphonse
Bertillon at the close o f the nineteenth century, was a necessary response to
increasing populations in urban areas and to increasing crime rates. The impetus
driving Bertillon’s work in Paris was to make a vast photographic archive useful to
police in cases of identification. The resultant reduction of bodies and their histories
into identification cards enabled police to keep track of growing numbers of crimes
and criminals. This is a central point of Allan Sekula’s analysis of the nineteenth-
century visual archive, that the defining feature of criminal identification at the close
of that century was the filing cabinet and not the photograph.4
4 Allan Sekula, “The Body and the Archive,” October 39 (1986): 3-64: 16.
62
Bertillonage remained in use well into the 1920s; however, by the close of the
nineteenth century the practice was continually being called into question.5 As the
use of Bertillonage increased, two central problems emerged. First, Bertillonage was
time-intensive: identification cards took significant time to produce and to search.
The system could not keep pace with the needs of law enforcement which faced
rapidly increasing crime rates. Second, Bertillonage was prone to discrepancies
between individual practitioners. The system depended on exact measurements
obtained through the universal application of procedures, apparatuses, and Bertillon’s
specialized language. As more agencies and individuals practiced Bertillonage,
greater discrepancies arose.
Dr. Henry Faulds and Sir William Herschel introduced fingerprinting as a method
o f identification during the 1870s and 1880s. From its outset, fingerprinting appeared
to avert the problems of Bertillonage. It was faster in both the collection and retrieval
of information, and removed much of the human element responsible for
discrepancies in application. Working from this position, most histories dealing with
criminal identification describe fingerprinting as a natural extension or expansion of
the ideas behind Bertillonage. The technique is described as a more accurate and
efficient means of classifying and identifying criminals. In this chapter I stress the
5 Simon Cole, Suspect Identities: A History o f Criminal Identification and Fingerprinting

(Cambridge: Harvard University Press, 2001); Simon Cole, “Manufacturing Identity: A History of
Criminal Identification Techniques From Photography Through Fingerprinting,” (Ph.D. diss., Cornell
University, 1998); Donald C. Dilworth, ed., The Blue and the Brass, American Policing: 1890-1910,
Police History Series (Gaithersburg: International Association of Chiefs of Police, 1976); Donald C.
Dilworth, ed., Identification Wanted: Development o f the American Criminal Identification System
1893 - 1943, Police History Series (Gaithersburg: International Association of Chiefs of Police, 1977).
63
significant changes fingerprinting introduced into identification practices and discuss
the broader ramifications of these changes. Fingerprinting, I argue, announced new
questions regarding the role and ieve! of law enforcement in public life. More
specifically, fingerprinting raised questions about the ethics and ramifications of law
enforcement practices based on the collection, control and use of personal
information, issues at the forefront of discussions about criminal identification
practices in the twenty-first century.
This chapter argues that identification by fingerprint is largely a photographic
process. The camera is present from the crime scene to the courtroom. It is the
central means through which prints are collected, examined, identified and shared.
To speak o f a fingerprint is to speak of Its photograph. The individual, physical print
holds little value in comparison to its photographic representation which can traverse
physical barriers and can be at work in any number o f locations and contexts
simultaneously. As I show, the collaboration of the theory of fingerprint
identification and photographic technology produced coherent, unified and highly
effective evidentiary statements that function as Latour’s immutable mobiles. I argue
that the collaboration of fingerprinting and photography in the production and
deployment o f immutable mobiles was essential in transforming practices of criminal
identification.
64
The rich and diverse history of the fingerprint as a method of identification has
been addressed in significant depth.6 This chapter, therefore, will not provide an
historical, overview of the subject. Instead I want to highlight some of the central
features within the history of fingerprinting In order to foreground discussion of the
photographic nature of fingerprint Identification. I conclude with a fictional crime
scene involving identification by fingerprint.7 The fictional scene demonstrates the
primacy and problematic nature of photography within fingerprint Identification, from
data collection, through the examination and production of evidence and into the
courtroom.
6 Sources offering a detailed historical overview include: Harrison C. Allison, Personal

Identification (Boston: Holbrook Press, 1973); Frederick R. Cherrill, The Finger Print System at
Scotland Yard (London: Her Majesty's Stationary Office, 1954); Cole, “Manufacturing;” Cole,
Suspect, Dilworth, Identification; Chief Inspector J.H. Duncan, An Introduction to Fingerprints
(London: Butterworth and Co., 1942); Bert Wentworth and Harris Hawthorne Wilder, Personal
Identification: Methods for the Identification o f Individuals, Living or Dead (Chicago: T.G. Cooke,
1932); George Wilton Wilton, Fingerprints: History, Law and Romance (London: William Hodge and
Company, 1938).
7 As part of the MA degree in Art History at York University, Toronto I established an

apprenticeship with the Forensics Identification Services Division of the Metropolitan Toronto Police.
The apprenticeship took place during August of 1996. Throughout the month I worked with Scenes of
Crime and Identification Officers to examine the use of photography within practices of criminal
identification. Much of the information on fingerprinting was taken from this apprenticeship. This
included access to various manuals and publications used in the training of Officers. Chief among the
materials was the Ontario Police College's Seem s o f Crime Manual and the following articles
produced by the Metropolitan Toronto Police Identification Bureau: "History of Fingerprints;"
"History of Identification by Fingerprints;" "Fingerprints as a Means of Identification;" "A Capsule
Chronological History of Fingerprint Identification;" and "Structure of the Skin."
65
Fingerprints as Immutable Mobiles
As a method of identification, fingerprinting is usually attributed to the work of
Dr. Henry Faulds and Sir William Hersche!.8 In the late 1870s Faulds, then a
physician at Tsukiji Hospital in Japan, was asked to examine finger impressions left
in ancient Japanese pottery. This project inspired Mm to collect and study the prints
of humans from various nationalities with hopes of finding racial or familial
similarities. Faulds outlines Ms work in an 1880 article in Nature'. “On the Skin-
furrows of the Hand.”9 He notes that Ms attempts to trace heredity through the
examination of fingerprints were inconclusive but that fingerprinting presented
several other potential uses. CMef among these was the ability to identify criminals.
Faulds writes: “When bloody finger-marks or impressions on clay, glass, &c., exist,
they may lead to the scientific identification o f criminals.”10 He then refers to two
cases in which he was able to use fingerprinting successfully to aid in criminal
investigations and concludes “there can be no doubt as to the advantage of having,
besides their photographs, a nature-copy of the for-ever-unchangeable finger-forms of
important criminals.” 11
8 Simon Cole also stresses the importance of another individual, Thomas Taylor, who founded
the first division of Microscopy at the U.S. Department of Agriculture. Taylor delivered a lecture in
1877 on the use of fingerprints to identify criminals that became the first published work on the subject
in the West. See Cole, “Manufacturing,” p. 128.
9 Henry Faulds, “On the Skin-Furrows of the Hand,” Nature 22, Oct. 28 (1880): 605.
10Faulds, “On the Skin-Furrows.”
11Faulds, “On the Skin-Furrows.”
66
William Herschel, an employee for the British East India Company, made a
subsequent contribution to Nature: “Skin Furrows of the Hand,”12 Herschel’s work
on fingerprints began in the 1850s when he was working in Bengal and used
fingerprints in place of written signatures on contracts. Like Fauids, Herschel tried to
trace heredity through fingerprints and was also unsuccessful. He writes:
My inspection of several thousands has not led me to think that it will
ever be practically safe to say of any single person’s signature
[fingerprint] that it Is a woman’s or a Hindoo’s, or not a male

13
European’s .... In single families I find myself the widest varieties.
Like Fauids, Herschel saw the potential for fingerprints to be used in criminal justice
practices, proposing that prisoners entering jail be fingerprinted for potential
identification needs. Despite their differences or the order in which their work
appeared, Fauids, Herschel, and other early investigators approached fingerprints
similarly in that they investigated them as unique identifiers.14 The central concern of
this work was to prove that no two fingerprints could ever be the same.
As described by Fauids and Herschel, the fingerprint was prone to the problems
facing other modes of identification. As an inscription, as a piece of data, the
fingerprint’s value lay in its potential to participate in a larger system of criminal
12 W. I. Herschel, “Skin Furrows of the Hand,” Nature 23, Nov 25 (1880): 76.
13 Herschel, “Skin Furrows.”
14Throughout the literature dealing with fingerprinting there has been continued debate as to
who founded the technique. The most recent example is that of Colin Beavan who pleads that Henry
Fauids should be acknowledged as the central figure in the technique’s history. See: Fingerprints: The
67
identification. The fingerprint had to be organized and classified. Contemporaries of
Fauids and Herschel sought ways to mobilize the fingerprint and transform it into the
basis for a complete system of identification. The work of Francis Gallon and Dr.
Juan Vucetich was of significant importance in this regard. Galton took up Fauids’
work through the prompting of his cousin, Charles Darwin, ultimately seeking
evidence of heredity through the examination of fingerprints. A necessary corollary
to Ms scientific investigations was the construction of a system for ordering and
classifying his data. Although unsuccessful in his quest to prove heredity through
fingerprints, Galton did produce essays and a detailed text on the topic in wMch he
described a system to classify prints.15 A year prior to Gabon’s 1892 text Dr. Juan
Vucetich established the Identification Bureau in Argentina and developed a system
for classifying fingerprints into a large, workable file. Vucetich’s system was
employed throughout Latin America. Simon Cole calls Vucetich’s “the first
classification system wMch rendered finger prints a practical means of indexing a
large criminal identification file.”16
The transformation in fingerprint identification came with the work of Sir Edward
Henry and his assistant, Azizul Haque. Like Herschel, Henry worked under
colonialist governments in India. Henry devoted specific attention to creating a
Origins o f Crime Detection and the Murder Case that Launched Forensic Science (New York:
Hyperion, 2001).
15 Francis Galton, “Personal Identification and Description: I f Nature 38, June 21 (1888):
173-77; Francis Galton “Personal Identification and Description: 2,” Nature 38, June 28 (1888): 201-
02; Francis Galton, Finger Prints (London: Macmillan, 1892).
16Cole, “Manufacturing,” 230.
68
detailed system for collecting, sorting and searching fingerprints. He divided
fingerprints into four basic types: arches, loops, whorls, and composites. These types
were further subdivided according to distinction in the ridge characteristics of the
print.17 Using these categories, Henry devised a complete classificatory scheme. His
1900 text, Classification and Uses o f Finger Prints, quickly became the central text
on the topic.18 The resultant “Henry system” remains at the foundation of current
fingerprinting practices.
Henry’s system provided a means for identifying, classifying, filing and searching
fingerprints. This enabled law enforcement agencies to collect fingerprint files
against which they could potentially identify individuals brought into police custody.
Under Henry’s system, fingerprints were analyzed according to four classificatory
divisions.19 The first division, the primary classification, was based on the
impressions o f the ten fingers of the suspect or criminal. The prints were assigned
numbers based on the presence of specific features, such as loops, whorls and arches.
The specific sequence of numbers representing the ten prints of the hand was
expressed as a fraction and served as the file number for that particular fingerprint
17Most sources agree that the majority of work attributed to Henry as regards classifying
fingerprints was actually produced by Haque.
18Sir E.R. Henry, Classification and Uses o f Finger Prints, 7th [1934] ed. (London: His
Majesty’s Stationary Office, 1900).
19There is a wide array of sources that address the procedures of fingerprint identification.
Three particularly clear sources are, Charles Edward Chapel, Fingerprinting: A Manual of
Identification (New York: Coward McCann, 1941); James Holt, Finger Prints Simplified: A Handbook
o f the Science o f Finger Print Identification (Chicago: Frederick J. Drake and Co., 1920); Frederick
Kuhne, The Finger Print Instructor (New York: Muibj and Company, 1916).
69
record. The second division, the sub-classification, distinguished prints with the same
primary classification through an analysis of the patterns of the index fingers. The
third division, the second sub-classification, divided prints having the same primary
and sub-classification according to the ridges of the index and middle fingers. The
fourth division, the final classification, used the ridges of the little finger to separate
prints having the same primary, sub-classification and second sub-classification.
Once reduced through these four stages, prints were filed in cabinets or in a pigeon
hole system organized according to the classificatory scheme.
To check a suspect against the fingerprint file, Ms or her prints would be subject to
the same classification process. Once the suspect’s prints were categorized through
the four divisions they would be compared against the records in the corresponding
pigeon hole o f the file. In a 1920 fingerprint manual, James Holt parallels the process
to searching an encyclopedia. He writes:
The finger print file has been likened to an encyclopedia. In fact it is
an encyclopedia of prints, the primary classification or first division is
the index; the sub-classification or second division the page; the
second sub-classification or third division the paragraph, the final
classification or fourth division the line in the paragraph.20
Three years prior to the publication of Henry’s Classification and Uses o f Finger-
Prints, the British Home Office officially accepted fingerprinting over Bertillonage in
20Holt, Finger Prints Simplified, 112.
70
cases of identification.21 Although it was taken up in various locations and contexts
throughout the closing decades of the nineteenth century, fingerprinting was not
officially introduced to North America until the 1904 World’s Fair in St. Louis.
Representing Scotland Yard at the Fair, John K. Ferrier provided demonstrations of
the technique to prominent law enforcement officials from Canada and the United
States. After the close of the Fair, Ferrier remained in St. Louis to teach the
technique to interested persons.
During the closing years o f the nineteenth century and the first decade o f the
twentieth, fingerprinting and anthropometry coexisted as methods o f identification
throughout Western Europe and in North America. Law enforcement organizations
experimented with both systems, often combining them. Bertillon eventually
provided space on his identification cards for fingerprints. Faced with increasing
crime rates and increasingly mobile criminals, law enforcement practices o f the time
were preoccupied with the need for large-scale information exchange. The situation
was compounded in North America by the fragmentation of its police practices. As
Donald Dilworth notes, policing within the United States developed sporadically and
yy
at the local level, and was marked by competition among locales. The same issue
existed in Canada, although to a lesser degree as the country had a significant federal
22Dilworth, Identification, 1.
71
level police force.23 These were in contrast to the European model, which was
centralized and more easily facilitated communication and cooperation.
Initially, identification cards were seen as a productive step toward developing
communication between local police departments. Information was shared by
departments through the mail in the form of “wanted” circulars. These circulars
included photographs, Bertillon measurements and a description of the offender’s
history. The effort was aided, in the early 1890s, by the journal, The Detective, which
published information on criminals in a manner reminiscent of Rogue’s galleries.24
The need for communication between law enforcement organizations culminated
in the formation o f the International Chiefs of Police Union in 1893. Comprising law
enforcement officers from across Canada and the United States, the Union was
renamed the International Association of Chiefs of Police (IACP) in 1902.25 A
primary goal of the Association was to establish a centralized resource for housing
and sharing information on criminals. This took form in the National Bureau of
Criminal Identification, which opened in Chicago in 1897 and moved to Washington,
D.C. in 1902. The promise o f a staffed, centralized resource was that it could more
effectively distribute identification information among police departments.
23 For a detailed account of the development of municipal, provincial and federal police in
Canada, see: William Kelly and Nora Kelly, Policing in Canada (Toronto: Macmillan, 1976); William
T. McGrath, and Michael P. Mitchell, ed., The Police Function in Canada (Toronto: Methuen, 1981);
C.K. Talbot, C.H.S. Jayewardene, and TJ. Juliani, ed., Canada's Constables: The Historical
Development of Policing in Canada (Ottawa: Crimcare, 1985).
24 Dilworth, Identification, 33.
25 See Dilworth, Identification and The Blue and the Brass.
72
At its outset, the National Bureau promoted the use of Bertillonage for
identification, It was hoped that through the use of a common language, Bertillonage,
departments across Canada and the United States could share information on
criminals. However, two central problems quickly surfaced: the time-intensive nature
of the technique and the discrepancies in application between practitioners. By the
opening years of the twentieth century, police and law enforcement personnel openly
criticized the Bertillon system. At the 1903 International Association of Chiefs of
Police convention, George M. Porteus, Superintendent of the National Bureau of
Criminal Identification, commented:
... the unreliability of photographs as well as the inability of most
people to identify by this means is generally conceded. The time
occupied in attempting an identification in this way makes it decidedly
unsatisfactory, if not entirely impracticable. . ,.26
Proceedings of the International Association of Chiefs of Police conventions from
the years surrounding the turn o f the century attest to the general frustration with the
Bertillon system.27 Even if the National Bureau could facilitate better communication
between departments, it could not guarantee the reliability of identification
Information, which still varied between practitioners. Through the Association’s
conventions, law enforcement officers continually expressed interest in standardizing
identification practices. A separate meeting of Bertillonage and fingerprint specialists
26 Quoted in Dilworth, Identification, 5.
27 Dilworth, Identification and The Blue and the Brass.
73
took place at the 1914 convention and gave rise to the International Association for
Criminal Identification in 1915, re-named the International Association for
Identification in 1918.28 The primary task o f this group was to universalize
identification procedures.
Allied with photography, fingerprinting produced identification information that
was concise, compact and readily transferable. The resultant documents served the
needs o f law enforcement, which faced increasing crime rates and increasingly
mobile criminals. This effort was aided by the development of fax technology, first
demonstrated in 1902,29 and the ability to transmit photographs over radio, described
in 1920.30 The potential use of these communication technologies for criminal
identification was immediately acknowledged. A 1922 Scientific American article,
“Finger-Prints via Radio,” described this benefit of the new technology:
The latest obstacle thus placed in the path o f the wrong-doer is the
long arm of the radio, which not only serves to ensure rapid
intercommunication of news and notes between the various police
forces of the world, but also transmits photographs and finger-prints of
28Dilworth, Identification, 109. The International Association o f Identification continues to

operate today. Information on this organization and its history is available through the Association’s
web-address, available from http://www.theiai.org.
29 On the development of fax technology see: Jonathan Coopersmith, “Facsimile's False

Starts,” IEEE Spectrum, February (1993): 46-49; Daniel M. Costigan, Electronic Delivery of
Documents and Graphics (New York: Van Nostrand Reinhold Company, 1978).
30 This process was developed by Eduard Belin of Paris. See: Austin C. Lescarboura,
“Sending Photographs Over Wires,” Scientific American 123, (1920): 474-84.
74
the wrong-doer for the purposes of instantaneous and positive long
distance identification.33
Interestingly, it was not just law enforcement officials but also communications
industry representatives who lauded the benefits of the new technique. At the 1925
International Association o f Chiefs of Police convention, JJ. Carty, Vice President of
the American Telephone and Telegraph Company spoke on the benefits of the
telephotograph and the ability to transmit fingerprints over telephone lines.32 The
ability to send fingerprints over radio and telephone positioned fingerprinting within a
large communications system. This increased not only the mobility of the fingerprint
but also the reach of law enforcement.
The relative ease with which the fingerprint could be produced and shared assured
the technique’s prominent position within law enforcement. Centralized resources
housing fingerprint files took form in the establishment of the Canadian Criminal
Identification Bureau in 1911 and the Identification Division of the Federal Bureau o f
Investigation in 1924. In both cases, fingerprint records and other identification
information were amalgamated into national archives. During this same period, the
technique’s weight as a legal and evidentiary document was being established in
founding court cases.33 By the 1930s fingerprinting served as the basis for a complete
31 “Finger-Prints via Radio,” Scientific American 127, December (1922): 386.
32 Dilworth, Identification, 186.
33 Colin Beavan and Simon Cole address the founding cases in Britain and the United States
respectively. See: Beavan, Fingerprints; Simon Cole, “Witnessing Identification: Latent
Fingerprinting Evidence and Expert Knowledge,” Social Studies o f Science 28, no. 5-6 (1998): 687-
712.
75
system of identification within the Western world,34 national bureaus were
established which housed and shared fingerprint records; and the technique had
gained M l acceptance as scientifically and legally valid. Developments in
communication technologies such as fax and radio connected law enforcement
practices across the Atlantic Ocean, transforming criminal identification into an
increasingly international practice.35
Simon Cole, Donald Dilworth and others have addressed the transition from
Bertillonage to fingerprinting in detail. In this chapter I stress that the success of
fingerprinting is bound to the construction of the fingerprint as an immutable mobile.
Compared to Bertillonage, fingerprints provided the necessary information (for
identification) in a more condensed physical space. Further, in contrast to the
complexity o f Bertillon’s photographic, textual and numerical descriptors,
fingerprints offered a single, visual statement. Finally, although fingerprints and
Bertillonage measurements were both derived directly from the criminal body,
fingerprints afforded a more direct representation of the biological body than did
photographic portraits or anthropometric measurements. More than a representation,
they presented a physical trace of the criminal. Writing an early history of the
technique, George W. Wilton highlights this biologically based authority: “But, while
34The United States, Canada and the United Kingdom employed variations of the Henry
system; Argentina used the Vucetich system and countries throughout Europe employed individual
systems that were largely based on Henry’s and Vucetich’s. Within each classification system prints
were analyzed according to the basic patterns and ridge details outlined by Henry and his
contemporaries, enabling prints to be shared between law enforcement departments.
76
photographic prints or slides and recorded voices are artificial, the evidence of finger-
impressions is the natural inference of presence at some particular spot of the
maker.”36
While this “natural inference of presence” guaranteed the weight of fingerprint
evidence as the direct manifestation of a criminal’s body, the mechanically objective
photograph transferred that presence between locales. I refer specifically to Lorraine
Daston’s and Peter Galison’s discussion of the historical specificity of mechanical
objectivity and its relationship to the natural sciences.37 Galison discusses the
physical, medical and biological sciences o f the nineteenth and early twentieth
centuries as relying on the mechanically objective camera to represent what was
accepted as the inherent truth of the natural world. Using the scientific atlas as his
example, Galison describes this relationship:
On the mechanical-objective view, realism, accuracy, and reliability
all were identified with the photographic. Nature reproduces itself in
the proceduraily produced image; objectivity is the automatic,
sequenced production of homomorphic images from the object of
inquiry to the atlas plate.38
35 The tremendous advances in computer and database technologies in the 1960s and 1970s
greatly extended the depth and power of this system, and will be addressed in Chapter Five.
36Wilton, Fingerprints.
j7 Lorraine Daston and Peter Galison, “The Image of Objectivity,” Representations, no. 40
(1992): 81-128; Peter Galison, “Judgment Against Objectivity,” in Picturing Science Producing Art,
ed. Caroline A. Jones and Peter Galison (New York: Routledge, 1998), 327-359.
38 Galison, “Judgment,” 334.
77
This combination of mechanical objectivity and the “natural inference of presence” of
a body, I argue, constructed the fingerprint as a particularly effective immutable
mobile. The compact visual space of the fingerprint could be reproduced and
transferred between locations and contexts indefinitely, each time reasserting the
supposed natural presence of the criminal5s body. I now turn to this chapter’s main
argument: that the success of fingerprint identification must be seen as inextricably
bound to its collaboration with photography.
Photographing Fingerprints
Fingerprint identification emerged within the same confluence of activity
discussed in Chapter Two. Like other scientific and social scientific practices of the
time, fingerprinting benefited from the rapid developments in photomechanical
reproduction. As an individual, physical object, the fingerprint is of little value. Its
weight as an evidentiary statement depends on its ability to reference a criminal body
from the crime scene to the courtroom. This is achieved through its mobilization in
the photograph.
Crucially, in accounts o f fingerprinting, both historical and critical, the fingerprint
and its photograph are rarely differentiated. As a result, what is routinely referred to
as “the fingerprint” is, more accurately, its photograph. This is nowhere more clear
than in the fingerprint’s use in court. Within the courtroom, “the fingerprint” is
actually a fingerprint chart: a material representation that includes photographic
enlargements of fingerprints, and is produced specifically for display in court.
78
Yet throughout literature dealing with criminal identification, and in accounts of
court cases and other judicial and penal practices, recognition of the photograph’s
role in fingerprinting Is rare. Explicit references to the medium In this sense are
confined to fingerprint instruction manuals, which detail the proper methods of
photographing prints from crime scenes. The conscious separation of the fingerprint
and its photographic representation would betray the authority of the technique. The
continued reference to “the fingerprint” throughout judicial and law enforcement
practices illustrates the level to which this separation has been removed from view.
Differentiating between the fingerprint and its photographic representation is not
to suggest that the latter is an inaccurate representation o f the former. Rather, it is to
stress the extent to which the latter functions as an unacknowledged representation of
the former. Fingerprint identification benefited from the communications capabilities
associated with photomechanical reproduction. The photograph made the fingerprint
mobile. However, the camera also rendered what was perceived as the “inherent
value” of the fingerprint, its “natural inference of presence,” immutable by
positioning it within a discourse of mechanical objectivity.
The collaboration of the theory of fingerprint identification and photography
signaled a distinct transformation in identification practices. Like physiognomy,
phrenology, and anthropometry before it, fingerprinting yielded data understood to be
derived directly from the physical body. Prints were collected directly from the
bodies of known criminals. However, unlike these earlier forms of identification,
fingerprinting enabled the collection and use of a new type of data: that derived
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
indirectly from the body, lifted from the crime scene. Using a camera, officers could
collect fingerprints long after a criminal had disappeared. Fingerprinting, I argue,
transformed identification practices into a system based on latent identification.
The procedure of this new method of identification and its promise were described
by Superintendent Arthur Hare at the 1904 International Association of Chiefs of
Police convention. He notes:
In cases of murder, the marks left by the blood stained fingers have
been photographed and compared with the finger prints of the habitual
criminal on file in the bureau, and by reason of the simple
classification, is sometimes identified and the crime brought home to
an individual who would never have been suspected but for the clue
which nature has provided.39
Superintendent Hare’s comment references the mobilization of the camera to link the
direct and indirect presence of the criminal. The criminal’s presence at the crime
scene, via the “blood stained fingers,” is tied to his presence in the bureau’s file by
the deployment o f the photograph. The criminal exists in both locations
independently and is only known through the use of the camera.
The specific importance of photography within this new system of identification is
perhaps best exemplified in a technique proposed by F.H. DePue of the San Francisco
Police Department during the opening years of the twentieth century. While Ms
efforts were ultimately rejected, DePue’s work provides a clear illustration of the
39 Quoted in Dilworth, Identification, 60.
80
unique features that photography afforded law enforcement and which were
ultimately manifest in fingerprint identification. DePue’s system worked by
comparing photographic projections of suspected and wanted persons. The enlarged,
projected images were dissected into a grid and anthropometric measurements were
taken and compared. DePue’s system was unique in the way he promoted the use of
the camera. He explains:
By the DePue system you can take a flash photo anywhere of your
suspected party, do your own measuring at your own sweet pleasure in
the secrecy of your office, without your subject having the slightest
knowledge of what is going on.40
DePue’s system highlights the three features fundamental to fingerprint identification
and its collaboration with photography. First, the camera enabled the collection of
information on the criminal in a subtle and non-physical way. Second, the resultant
images could be examined and re-examined infinitely and independent of the physical
body. Finally, this entire process could be performed without the suspect’s
knowledge.
With physiognomy, phrenology and anthropometry, measurements were taken
from a body that was physical present and often known to be normal or deviant. In
contrast to this, fingerprinting gave rise to the production and accumulation of data on
unknown bodies. Fingerprints were devoid o f the “obvious” bodily connotations of
40Dilworth, Identification, 50. DePue presented his system to the International Association of
Chiefs of Police convention in 1902.
81
race, gender and class, and the various “signs” of criminality or deviance that served
as the data for prior methods o f identification. Fauids, Herschel and Galton had all
concluded that fingerprints did not reveal themselves as belonging to any specific
type of body. The search for criminality shifted from application (as in the
application of instruments to the criminal’s body) to interpretation (as in the
interpretation of the fingerprint).
The results of this shift in identification practices were significant. The camera
transported data from the complexity of the ever-changing live environment to the
controlled setting of the police laboratory. This, in turn, gave way to the increased
importance o f the act of interpreting, or reading, images and, ultimately, to the
construction of a distinct, professional group of fingerprint examiners.41
Identification was no longer the purview of entire departments (or the general public
as was Thomas Byrnes’ hope) but of a specialized group o f trained and certified
fingerprint examiners. Identification was determined through the production and
interpretation of photographs of fingerprints and the separation of evidence from data.
The development of the half-tone process and celluloid roll film gave rise not only
to an expansive and fragmented photographic industry, but also to the separation of
practices within the criminal identification process. Officers in the field were no
longer charged with identifying criminals. Equipped with cameras, their specialized
41 Simon Cole has addressed the professionalization of fingerprint examination. See,

“Witnessing.”
82
task was data collection.42 Within the laboratory, the trained eye of another expert
separated the evidence from the data and identified the criminal body. Prints
collected from crime scenes were done so with the promise of identification, not
because they displayed obvious associations with criminality. In this sense,
fingerprints are always and only sites of latent or potential identification.
The ramifications of this shift in Identification practices are best understood by
returning to a central feature of Latour’s discussion of immutable mobiles. As Latour
explains, the creation, control and deployment of immutable mobiles within any
scientific practice is bound to positions of power and authority. Cole’s work has
addressed fingerprinting in this context. Specifically, Cole highlights the power
fingerprinting afforded Its practitioners as it originated and developed in colonial
India. He shows that fingerprinting was viewed as a way to differentiate among faces
that were perceived as visually homogeneous and that its function was primarily tied
to the recording, tracking and control of colony populations. In addition, Cole shows
that the technique’s adoption into Western police practices was similarly invested
with the tracking and control of bodies. Because fingerprinting was an inexpensive
and efficient method of identification, Cole argues, the technique “offered a way to
bring the poor, the ill, the addicted into the custody of the state under the guise of
petty crimes, such as prostitution, vagrancy, or intoxication.”43 He suggests that the
421 do not mean to suggest that this resulted in a total shift in criminal justice practices.
Officers in the field still participated in criminal identification; however, within fingerprint
identification their role was as data collectors.
43 Cole, Manufacturing, 276.
83
net effect was “to give increasing numbers of people a criminal record which could be
used against them at a later date.”44
In Cole’s account, power is accrued through the arrest of greater numbers of
bodies. By casting a wider net, more criminals were caught, and more were subject to
practices of identification. I suggest yet another means through which fingerprinting
afforded greater power. By giving rise to a system based in latent identification,
fingerprinting enabled a new and far-reaching mode of state surveillance. The
aggregates o f data, in the form of photographs of fingerprints, that could be and
would be produced through fingerprinting brought increasing numbers of individuals
into the police laboratory. Centralized fingerprint archives such as those of the
Canadian Criminal Identification Bureau and the Identification Division of the
Federal Bureau o f Investigation housed data on criminals and non-criminals alike.
Criminality was increasingly found in the interpretation of these immutable mobiles
and not on the body.
I do not want to suggest that, from its inception, fingerprints were being collected
randomly in the hopes o f finding criminals. Early uses o f the technique mirrored
prior identification practices: it was used to record the criminal body. However, as it
was allied with the camera, fingerprinting enabled the collection of aggregates of data
representing the criminal, the non-criminal and the potential criminal The camera
mobilized law enforcement and identification practices so that their subject was no
longer the individual body but a larger, aggregate social body, within which any
84
specific body was potentially criminal. Rather than elaborate on this process in the
abstract, I turn to a concrete examination of identification by fingerprint.
Photography from the Crime Scene to the Courtroom
Superintendent Hare’s comment at the 1904 International Association of Chiefs of
Police convention highlights the central role of photography in fingerprint
identification. Importantly, cases like that described by Hare would have been
relatively uncommon during that time as fingerprinting operated according to the ten-
print Henry system. The ability to identify a single or an isolated few latent
fingerprints found at a crime scene against a fingerprint file based on the ten-print
system was not impossible but was time-consuming.45 In response to this individual
police departments collected single fingerprint files, organizing the prints according
to their specific idiosyncratic features.46 Numerous single print identification systems
developed through the late 1910s and the 1920s in Europe and North America and
were commonplace by the end o f the 1930s.47
The theory and practice o f fingerprint identification remained surprisingly
unchanged during the middle decades of the twentieth century. Law enforcement
agencies continued to collect fingerprint records, filing them according to their own
45 Chapel, Fingerprinting, 234.
46Chapel, Fingerprinting, 52.
47 For a detailed discussion of single finger print systems see, Chapel, Fingerprinting, and
Cole, “Manufacturing.”
85
specific variant of the systems developed by Henry and Vucetich. As these
fingerprint archives grew their usefulness declined. Filing systems based on Henry’s
or Vucetich’s classifications became inadequate in the face of the vast accumulation
of fingerprint records. The development of computer technology in the 1960s and
1970s offered a solution to this problem. Converting the live fingerprint to an
electronic code enabled law enforcement agencies to collect, store and search
fingerprints in ways far beyond the limits of the earlier manual systems.
Experiments to record fingerprints electronically began in the late 1960s and early
1970s under the direction of companies including McDonnell Douglas Corporation,

AQ
KMS Technology Center, Sperry Rand, Rockwell International and NEC. Despite
their specific approaches, each company developed computer programs for reading
and searching fingerprints based on the same principles governing Henry’s
classification system. Law enforcement agencies adopted the technology quickly.
Among the first to employ automatic fingerprint identification equipment were the
New York State Identification and Intelligence System in 1968, the Atlanta Police
Department in the early 1970s, the Kansas City Police Department in 1974 and the
Nassau County Police Department in 1975.49 At the federal level, the FBI installed
its first fingerprint scanner and began digitizing its extensive fingerprint collection in
48 On the development of Automated Fingerprint Identification Systems see: Bruce I.

Brotman and Rhonda K. Pavel, “Identification: A Move Toward the Future,” Law Enforcement
Bulletin , no. July (1991); Cole, Suspect, Chapter 10: Digital Digits; Bill Reed, “Automated Fingerprint
Identification: From Will West to Minnesota Nine-Fingers and Beyond,” Journal o f Police Science
and Administration 9, no. 3 (1981): 317-326; Thomas F. Wilson, “Automated Fingerprint
Identification Systems,” Law Enforcement Technology, August / September (1986): 17-46.
49 Reed, “Automated,” 320-321.
86
1972.50 A year later the Royal Canadian Mounted Police installed its automated
system.51 By the late 1970s and early 1980s Rockwell International and NEC had
developed complete electronic fingerprinting units for commercial sale.32 These
units, referred to as Automated Fingerprint Identification Systems (AFIS) were
capable o f scanning, classifying, filing, storing and searching fingerprints
electronically and are now the mainstay of law enforcement agencies.
The original AFIS units developed by Rockwell International and NEC scanned
pre-existing inked impressions, transforming them into digital code. Law
enforcement agencies could scan existing, inked fingerprint records or could
fingerprint a person brought to the police station and then scan the record into the
system. By the late 1980s and early 1990s paperless and inkless “live scan” systems
were developed. Today numerous companies produce and sell a variety of AFIS
hardware, software, peripheral and accessory products and law enforcement agencies
are moving towards a paperless environment.53
50Cole, Suspect, 252.
51 Reed, “Automated,” 323.
52 Cole, Suspect, 253.
53 Three central companies providing AFIS equipment and products include: Lockheed
Martin, see http://www.lockheedmartin.com/lmis/level3/fingiden.html; NEC, see
http://www.necsolutions-am.com/idsolutions; and Printrak International (a Motorola Company), see
http://www.printrakintemational.com. An on-line source with extensive lists of AFIS products and
companies is maintained by a member of the U.S. Army crime laboratory. See,
http://onin.com/lp/afis/afis.html, accessed May 24,2003.
87
Automated Fingerprint Identification Systems have two primary features: the
ability to scan a fingerprint and record its minutiae, and the ability to search existing
fingerprints according to the recorded information. AFIS creates a binary code for a
fingerprint based on the spatial relationship of the print’s minutiae. The binary code
and digitized image of the print are stored in the program’s database. AFIS searches
will yield a number of potential matches, predefined by the operating personnel,
based on similar binary codes found in the database. Searches will return the images
of the prints with similar code for a side-by-side comparison against the print in
question. Importantly, the fingerprint examiner must ultimately decide on
identification, AFIS functions to replace the manual tasks of filing and searching the
fingerprint archive.
AFIS units archive two distinct types of prints: complete 10-print fingerprint
records collected directly from the physical body and, single, partial or multiple latent
prints found at crime scenes. Searches function across three levels. First, a complete
10-print record can be searched against the database of known offenders. This is
primarily beneficial for establishing identity when the person being printed may
falsify their name. Second, a latent print can be searched against the database of
known offenders. This search would identify individuals present at a crime scene,
otherwise unknown to police. Third, a new 10-print record can be searched against
the latent database. This final search could link the individual who is being printed to
past and unsolved crimes.
88
The development and growth of computer technology from the 1960s through to
the present has revolutionized the fingerprinting capabilities of law enforcement. The
majority of law enforcement agencies in the United States and Canada currently use a
combination of the traditional, ink-based system and the new, paperless and inkless
live-scaa systems. However, I stress that the basic classificatory and searching
mechanisms remain remarkably similar to those developed at the turn of the
nineteenth and twentieth century.
Despite the continued advances in systems for classifying and searching
fingerprints the camera’s role has remained relatively constant. Fingerprinting
manuals from the early 1900s to the present describe a similar process. The camera is
used initially to document the crime scene prior to investigation. Officers enter the
scene, search for fingerprints and dust, illuminate or otherwise prepare prints to be
photographed. Photographic enlargements are made from the collected prints away
from the crime scene. Prints are classified according to the existing system, from the
ten-print Henry system of the early 1900s to more current automated digital systems,
and searched against a fingerprint file. Once a match is made photographic
enlargements of the found and known prints are produced. The identifying features
of the prints are marked on their photographs and described in an accompanying text.
To better understand the role o f photography in fingerprint identification I turn to a
fictional crime scene produced during an apprenticeship with the Forensics
Identification Services Division of the Metropolitan Toronto Police.54 The case is
54 See note 7.
89
real in application and procedure but fictional in its goal. The purpose of the fictional
scene was to trace the use of photographic representation in fingerprint identification
from the collection of crime scene materials through investigation and into its use in
the courtroom.
The fictional scene proceeds as follows: Scenes of Crime and Identification
Officers examine and photograph a crime scene, its contents and its setting.55 Within
the scene, sites that might reveal fingerprints (window sills, doors, glasses etc.) are
dusted, illuminated through ultra-violet or infra-red light, or manipulated in another
way to make the prints visible.56 These prints are then either photographed, lifted (a
copy is taken by applying adhesive tape or another material to the dusted print and
“lifting” it off of a surface), or both depending on the nature of the print. In this
specific case, a dinner plate located at the crime scene is examined and dusted and
fingerprints are found. All prints from the scene that can be lifted or photographed
are taken to the police department where they are scanned into the Automatic
Fingerprint Identification System. Using a computer mouse, an officer marks the
idiosyncratic features of each print, their "ridge characteristics," and searches the
database for similar markings. AFIS presents a series of possible matches, each of
which is compared against the found prints. If a match is not made the print under
55Within the Metropolitan Toronto Police, Scenes of Crime Officers receive general training
on evidence collection, including basic notes on photography. Identification Officers receive more
specialized training, particularly in reference to the collection and photographing of crime scene
materials.
90
examination remains in the database as a potential source for future Investigations. If
a match is made, the fingerprint and any accompanying information will be collected
by the examining officer. In our case a print from the dinner plate draws a match to a
print in the database and both are prepared for further examination.
Identification by fingerprint takes place through matching the Idiosyncratic
features of found (from the crime scene) and known (on file with the police) prints.
The first step is to locate the hand and finger from which the print originated (i.e. the
left or right hand and the middle, index, etc. finger). Next the print's general pattern
is noted (whether it swirls or moves in a given direction). The most significant part
of the process is the Identification of the ridge characteristics, sometimes still referred
to as the “Henry details” or the “Galton details.” The various combinations of these
features, as dots, lakes, bifurcations and ridge endings are unique to each print (Figure
3.1). Using the ridge characteristics as signposts, the examining officer constructs a
narrative that links the known and found print.
Although simplified, the scene described is representative of the general
investigative procedure. The fictional scene demonstrates the central role of
photography throughout the investigation. A first photograph is taken that positions
the dinner plate within the context of the crime scene (Figure 3.2). The second
photograph locates the fingerprint on the plate (Figure 3,3). The third is a close-up of
the print (Figure 3.4). Once this general continuity is established the print is lifted
56 It is important to note that the collection of fingerprints is only one part of the larger
investigative process. Other forms of potential evidence, such as objects at the scene are also collected
and are photographed.
91
onto acetate, enabling it to be scanned into APIS and to be used for further
investigation (Figure 3.5). In our fictional scene, AFIS produced a match to a latent
print on file. The matching criminal record is obtained and an examination takes
place (Figure 3.6).
For the examination the Identification Officer uses the third photograph from the
series, which was the close-up image o f the print. Both the close-up and the latent
print (from the fingerprint chart) are re-photographed and enlarged to 8/12 by 11
(Figures 3.7 and 3.8). These two photographs are then read by the examining officer
and prepared for use in court. The resultant fingerprint chart is accompanied by a
narrative account of how the evidence is to be read (Figure 3.9). This material is
combined with a detailed list of the examining officer’s credentials and presented as
final evidence.
Figure 3.1: The ridge characteristics of the fingerprint.
92
Figure 3.2: The first photograph positions the dinner plate within the context of the
crime scene.
Figure 3.3: The second photograph positions the fingerprint on the dinner plate.
93
Figure 3.4: The third photograph isolates the found fingerprint.
Figure 3.5: The print from the plate is lifted onto a sheet of acetate, enabling it to be
scanned into the APIS database.
94
L o c k -u p No
s&efurcnsr
wmm
j
I# METROPOLITAN TORONTO
POLICE
i
TO' HI
.A
SAM..NQUE,<JNEF£gl''S§ftg
APS?tT£2gT
iNT . *O
£W
NE
OTRU
ROER
CO
QR
UD
tyE-D.EN
G!C
VA
5SRE
A
U'ASM
OP
NU-TA
<=AW
lN.uT
TiC aT
PE E
SD
.O ,R
FOD
MEA
FO
.TRM
IO
NEDOR
OU05m8ifii-3
«.ESSU,R
G
EI.VED
DO
NA
fT EF!IADATg
rig
FOURSNCERSTAKENTOO£TH6R-JM PR£S$-.ONSM m.TANE£OSSQUATSCQOfGTS
*0%
III!
«4
-‘fljiva p o occ c a j o h , sva-j* r,
| V-A
-u« ; <«■
S s ^ r t u r e e» a re a< i-S i^ a n S S o e ta f^ S w w
S ” D f |: y g L ?.e L
iî^m
"dcSs-dTcT w.'"” ’ ’'"'’’"""IfflftfrKSir-rj
**
'fHaceof Aprov>ra»t~Uauaeniissaneai’.
c(NOK-Nwnalasjross®tSup!t»e««Sf®
Figure 3.6: The criminal record matching the identified print is collected.
95
Figure 3.7: A negative is made from the found print to be enlarged for final
identification and presentation in the fingerprint chart.
Figure 3.8: A negative is made of the latent print from the fingerprint record to be
enlarged for final identification and presentation in the fingerprint chart.
96
FOUND PRINT m KNOWN PRINT
Explanation of Illustrated Characteristics
A. Ridge ending downward.

Move left one ridge to:
B. Bifurcation opening downward.
Move left into the furrow (space between the ridges) and follow that down to:
C. Ridge ending upward.
Left into the furrow and follow that down to:
D. Ridge ending upward.
Move up and to the right seven ridges to:
E. Bifurcation opening downward.
Move left one ridge and follow that up to:
F. Bifurcation opening downward.
Figure 3.9: The fingerprint chart with its accompanying narrative.
97
The fictional crime scene shows the centrality of photography in fingerprint
identification. The photographs establish a continuity of presence; a photographic
paper trail linking the criminal to the crime scene. Throughout the examination the
camera is used exhaustively in order to isolate and make the criminal visible. His
presence, manifest in the ridge characteristics of Ms fingerprint, drives the
investigation. All else is excluded. What we see in the final photographic
representations used in the fingerprint chart are complicated visual texts that have
been constructed to look uncomplicated. Their clarity and precise appearance are
taken to announce their clear and precise meaning.
The fictional scene also shows the inseparability and interchangeability of the
fingerprint and its photographic representation in the construction of evidence.
Photographs o f the fingerprint function as immutable mobiles throughout the
investigation. The mechanically objective camera reproduces, represents and
guarantees the “natural inference of presence” of the criminal from the crime scene to
the courtroom. His presence on the dinner plate is tied to his presence in the
Department’s file through the production and interpretation o f photographs. Despite
their various manifestations these images continually display the presence of the
criminal through the ridge characteristics o f the fingerprint. The variables of time,
location, purpose and context of production are rendered invisible in the fingerprint
chart, wWch shows only the presence of a specific body.
The fictional scene focused on the collection and examination of a single
fingerprint. This was necessary to trace the production o f evidence and the use of
98
photography throughout that process. It is important to note that the final print is one
among a much larger number collected from the crime scene. Each of the
fingerprints collected is archived into the APIS database. The prints not useful to the
fictional scene can be linked to past crimes or can wait for identification in future
cases. Which is to say that an individual present at the fictional crime scene, although
not ultimately identified or arrested in relation to that case, could be linked to the
fictional scene if arrested, and printed at a later date. The prints remain archived as
sites o f latent identification: as data to be transformed into evidence.
Conclusion
The point of differentiating between the fingerprint and its photographic
representation is not to call for the more accurate use of language in law enforcement
and legal and penal practices. Rather, it is to acknowledge the fingerprint’s status as
an immutable mobile. In this way “fingerprinting” is more accurately understood as
an identification system that emerged from, and relies upon, the collaboration of
photography and a theory of identification (that fingerprints are unique identifiers).
This allows us to examine the intersections of photography and the fingerprint in the
production of evidence. As an immutable mobile, the relative ease, efficiency and
accuracy of the fingerprint assured its success over competing methods of
identification. Law enforcement practices required identification information in a
compact, concise format that could be easily produced and interpreted and transferred
between locales quickly and without misinterpretation or the loss of information.
99
Photography and the fingerprint combined to produce the immutable mobiles
necessary to fulfill these needs. In this sense, fingerprinting merely extended existing
law enforcement practices which sought to keep pace with increasingly mobile
criminals. However, as I argue, fingerprinting raised distinctly new questions
regarding the ethics associated with the collection, control and use of personal
information by law enforcement. This chapter concludes with a discussion of these
issues.
Simon Cole describes the development and use of fingerprinting within law
enforcement as it was tied to existing conceptions of race, gender and class.
Specifically, he positions fingerprinting as a tool used in the control of “marginalized
individuals.”57 In Cole’s account fingerprinting afforded law enforcement distinct
power as it worked in concert with existing criminal codes to bring more people “into
the system.” I agree that fingerprinting greatly assisted in this effort; however, I
stress that the technique asserted a greater power as it transformed practices of
criminal identification into a system based on latent identification.
Within this new system of criminal identification, photographs of fingerprints
offer tremendous power to law enforcement as they can be deployed infinitely and
immutably between contexts and locations. Fingerprint files attest to the “natural
inference of presence” of innumerable bodies in innumerable locations throughout
time. Suspects can be linked to past crimes through their “natural Inference of
100
presence” on file with the police. The collection and use of latent information in this
regard has proved to be an invaluable tool for law enforcement in tracking and
arresting recidivists. However, law enforcement practices based on the use of latent
identification have clear and far-reaching social ramifications. Specifically,
aggregates of data used in latent identification can be interpreted and re-interpreted
according to changing conceptions of criminality.
A fingerprint, whether live or photographed, is immutable in the sense that it
attests to an individual’s presence at a specific time and location. In contrast to this,
criminality is a highly mutable social construct. Throughout the development of the
modem criminal subject, outlined in Chapter Two, criminality was bound to
conceptions of race, class, and gender. These notions continue to influence
contemporary discussions o f crime and criminality. I argue that law enforcement
practices based on latent identification threaten to discriminate against individuals
according to notions of race, class, gender, sexual orientation and other features.
Changing conceptions of criminality based on these characteristics yield new
parameters for searching and interpreting the fingerprint file.
I do not want to suggest that fingerprint data is collected for deliberate and
malicious use; however, as it is collected for potential use fingerprint data threatens to
discriminate against individuals, groups and nations according to mutable conceptions
of criminality. The aggregates of immutable identification information stored in
fingerprint files exist as sites of potential or latent identification. The data is
mobilized and identified through processes of investigation and interpretation that
101
reflect historically specific conceptions of criminality. Within this process, variables
including race, class, gender and sexuality as well as moralistic values of right and
wrong predetermine the use and interpretation of this data. At stake are the
construction of individual, group and national identities with their attendant rights and
limitations.
Paul Rabinow uses the phrase “Galton’s Regret” to title an essay in which he
compares fingerprinting and DNA analysis as methods of identification.58 He notes
that Gallon regretted the inability to use fingerprints to track and study hereditary
relationships. In Ms studies o f fingerprints, Gallon, like Ms predecessors and
successors, found variability across families, races and nationalities. The fingerprint
offered no information beyond the individual body it represented. Rabinow uses tMs
regret to position Ms discussion o f the potential social ramifications o f DNA analysis.
I employ it to differentiate between the fingerprint and its photographic
representation. Fingerprints are immutable references to Individual bodies. They
attest only to an individual’s physical presence at a distinct time and location. In
contrast to this, photographs of fingerprints mobilize those bodies within mutable
conceptions of criminality. Photographing fingerprints produces aggregates of data
within wMch any individual is potentially criminal.
58Paul Rabinow, “Gabon's Regret: Of Types and Individuals,” in DNA on Trial: Genetic
Identification and Criminal Justice, ed. Paul R. Billings (Plainview: Cold Spring Harbor Laboratory,
1992), 5-18.
102
CHAPTER FOUR
DNA ANALYSIS: COMPETITION, STANDARDIZATION AND THE CONTROL
OF INSCRIPTIONS
In “Gallon’s Regret: O f Types and Individuals,” Paul Rabinow points to a central
distinction between fingerprinting and DNA analysis.1 Whereas the former is a
phenotypic process, the latter is genotypic. Fingerprinting is based on the external,
physical signs of the body, and DNA identification works at the cellular level.
Proponents of DNA analysis describe it as the most powerful and effective method of
criminal identification precisely because it operates at this primary biological level.
As described in a 1996 National Research Council report on DNA evidence, “these
newer molecular techniques permit the study o f human variability at the most basic
level, that of the genetic material itself, DNA.”2 DNA analysis suggests the natural
inference of presence of the criminal at a level beyond that of fingerprinting. Instead
of being the tactile trace o f a body, DNA is understood to represent the biological
constitution of that body.
Despite being in only in its second decade of use within law enforcement
practices, DNA analysis has become an essential and powerful tool in criminal
1Rabinow’s essay appears in Paul R. Billings, ed., DNA on Trial: Genetic Identification and
Criminal Justice (Plainview: Cold Springs Harbor Laboratory, 1992), 5-18.
2 The Evaluation o f Forensic DNA Evidence (Washington, D.C.: National Research Council,
1996), 9.
103
identification. It is particularly useful in cases involving rape and murder where
biological material is left and where other forms of evidence are often absent. DNA
analysis has enabled the capture of large numbers of repeat violent and sexual
offenders and has been used to exonerate the falsely convicted. In addition, DNA
technology has several applications beyond law enforcement. It is used by
biotechnology corporations in paternity testing and in genetic research, by the
military for the future identification of soldiers killed in battle or missing in action, by
employment, insurance and health organizations in determining clients’ contracts and
policies and by research laboratories for studies in disease, health and heredity. The
potential use and misuse o f genetic information within these contexts has been a
source o f continued discussion since the technique’s inception.
By 1994, only seven years after its initial use for criminal identification in the
United States, DNA analysis within law enforcement was standardized under the
direction o f the FBI. The DNA Identification Act, written in 1993 and enacted in
1994, gave the FBI authority over the technique’s procedures and protocols. This
chapter investigates the construction of DNA analysis as a method o f criminal
identification in the United States from its inception in 1987 to the enactment of the
DNA Identification Act in 1994. I locate the control of inscriptions as fundamental in
this process. I argue that, by legalizing the FBI as arbiter of how DNA evidence is
produced, the DNA Identification Act gave rise to a position of “dominance” over its
inscriptions.
104
Latour uses the term “dominance” to describe a central, authoritative position
within scientific practice. In the construction of scientific facts and artefacts,
individuals and groups compete for dominance within their shared field of practice.
Latour suggests that this competition takes place through the production and control
of inscriptions. The group that produces and employs the best inscriptions wins out
over others and establishes dominance in the field. Latour compares the process to an
arms race, noting “once one competitor starts building up harder facts [through the
production and deployment of inscriptions], the others have to do the same or else
submit.”3 Further, once one competitor establishes a position of dominance it
becomes increasingly difficult to challenge that position and its inscriptions. He
explains: “although in principle any interpretation can be opposed to any text and
image, in practice this is far from being the case; the cost o f dissenting increases with
each new collection, each new labeling, each new redrawing.”4 Competitors unable
to challenge the dominant position are forced to accept that position and its
inscriptions.
In this chapter I argue that the FBI has achieved dominance, as described by
Latour, over DNA analysis in law enforcement. I trace the development of this
position through the gradual control of the inscription of DNA, the autoradiograph;
more specifically, through the control and dissemination of procedures for producing
J Bruno Latour, “Drawing Things Together,” in Representation in Scientific Practice, ed.

Michael Lynch and Steve Woolgar (Cambridge, Mass.: MIT Press, 1990), 19-68; 35.
4 Latour, “Drawing,” 42.
105
autoradiographs. I argue that the DNA Identification Act concretized the FBI’s
dominant position in three primary ways. First, It legalized the Bureau to develop,
monitor and control the standards for producing autoradiographs. Second, it
delimited the FBI’s authority over DNA analysis to those practices specifically
related to law enforcement. The establishment of this boundary was necessary to
assure the FBI’s position within the highly competitive market of DNA analysis.
Third, the DNA Identification Act provided the FBI with the necessary funding to
implement a National DNA Index, enabling the Bureau to disseminate Its procedures
for DNA analysis throughout local and state law enforcement agencies. This further
strengthened their control over the technique and also gave them access to the genetic
identification data housed in law enforcement agencies throughout the United States.
Because o f its unique dependence on inscriptions DNA analysis provides an
excellent arena through which to address the importance of visual representations in
scientific practice and the power associated with their control However, whereas
Chapter Three focused on the concrete production of inscriptions and immutable
mobiles, this chapter examines the more abstract field o f the institutional control of
material representations. I use the inscription of DNA, the autoradiograph, as a point
of departure to investigate the construction o f authority over a specific scientific
practice. The chapter addresses the significant and far-reaching effects this dominant
position has on future uses and applications o f DNA within law enforcement
practices.
106
In contrast to fingerprinting, DNA analysis has received significant critical
attention throughout specialized and popular presses. This chapter draws
significantly from tins work, using a variety of sources from scientific, legal,
academic and professional fields. I stress that DNA analysis is largely a visual
process. Within existing work on the technique, material addressing its visual nature
is limited. In their essay, “The Fixation of (Visual) Evidence,” K. Amann and K.
Knorr Cetina discuss the visual act of Interpreting DNA autoradiographs.5 Their
essay presents research conducted in a molecular genetics lab during 1986. During
this research the authors studied the types of “talk” that took place between
participants during the interpretation of autoradiographs. They show' that the image’s
meaning, its transition from data to evidence, is produced through this interactive
process and that the process itself is self-perpetuating. They explain:
Evidence is the aesthetically enhanced, carefully composed rendering
o f flexible visual objects that, through the meandering interrogatory
processes o f image analyzing talk, have been “embedded” and
entrenched in procedural reconstructions, local experiences and in the
landscape o f the data display.6
5K. Amann and K. Knorr Cetina, “The Fixation o f (Visual) Evidence,” in Representation in
Scientific Practice, ed. Michael Lynch and Steve Woolgar (Cambridge, Mass.: MIT Press, 1990), 85-
122 .
6Amann and Knorr Cetina, “The Fixation,” 115.
107
Sheila Jasanoff also examines a specifically visual aspect of DNA evidence.7 She
analyzes the use of DNA evidence in the O J . Simpson trial to address the
construction of authority over visual evidence. She notes the authority to “see”
within legal contexts, to have one’s views accepted as “expert,” is often determined
through the specific parameters of a case. Thus, in her analysis of the O J. Simpson
trial she emphasizes the role of Judge Ito as the arbiter o f what is and what is not
“seen” by the jury. She argues that in controlling what visual evidence was shown in
the trial, Judge Ito substituted his own vision for that of everyone so that Ms own
vision became that of the jurors.
Despite their particular emphases, the essays of Amann and Knorr Cetina and
Jasanoff demonstrate the mutability of visual evidence. Truth or fact is determined as
much through the framing o f what is being seen as through the image being
examined. TMs chapter complements the work of Amann and Knorr Cetina and
Jasanoff by focusing on the visual nature of DNA analysis. However, my work is
unique in its specific treatment of the visual. Whereas Amann and Knorr Cetina
examine the act of Interpreting autoradiographs In the research laboratory and
Jasanoff examines the authority associated with controlling the deployment of
autoradiographs as visual evidence in criminal trials, I stress the construction of
dominance over DNA analysis through the control of procedures for constructing
autoradiographs.
7 Sheila Jasanoff, “The Eye of Everyman: Witnessing DNA in the Simpson Trial,” Social
Studies o f Science 28, no. 5-6 (1998): 713-740.
108
Visualizing DMA: From Sample to Profile
Criminal identification by DNA involves the comparison of sample DNA (from
the crime scene) to DNA drawn from the suspect. Analysis begins with the collection
o f DNA from cells, usually from found samples o f blood, semen, bone marrow and
hair. Once extracted the DNA is cut into smaller pieces using restriction enzymes.
The separate pieces are of different lengths and represent specific groupings o f the
building blocks of DNA: adenine, cytosine, thymine, and guanine. From here, the
pieces are subjected to an electrical charge, through the process of gel electrophoresis.
This charge separates the pieces according to size and the speed with which they
move through the gel. After being transferred from the gel to a sheet of nylon
through a process called blotting, a radioactive probe is added. Because the building
blocks o f DNA only pair with their specific complement (adenine with thymine and
cytosine with guanine) the probe binds to its matching fragment and makes the
structure of the DNA known. The segments o f DNA with their bound probes are
rendered visible as the sheet o f nylon is placed against a sheet of photographic paper
and the probes register as dark bands. The final product is called an autoradiograph,
« ©
alternatively an autoradiogram or autorad (Figure 4.1). The parallel lanes of an
autoradiograph represent the sample DNA, the suspect’s DNA and control markers.9
8 There are a tremendous number of descriptions of this process available. Four particularly
clear sources include DNA Technology in Forensic Science (Washington, D.C.: National Research
Council, 1992); The Evaluation o f Forensic DNA Evidence; Thomas Curran, Forensic DNA Analysis:
Technology and Application (Parliamentary Research Branch, Government of Canada, 1997) BP-
443E; Eric Lander, "DNA Fingerprinting: Science, Law, and the Ultimate Identifier," in The Code of
109
aw mm&7
*
! m
m m
m
m
m
« k
m z *
* 1*.
mm
mm • •
•
•<*A
%
mm -# -
mm
mm s •
•
m* • •
at* #•
««!» ■m #
#:■
mm # il
Figure 4.1: The DNA autoradiograph. From the National Research Council’s DNA
Technology in Forensic Science.
Codes: Scientific and Social Issues in the Human Genome Project, ed. Daniel 1. KevSes and Leroy
Hood (Cambridge: Harvard University Press, 1992).
110
It is important to note the distinction between DNA samples and DNA profiles.
The blood, hair, semen or other source from which the DNA is derived is referred to
as the sample. The autoradiograph produced from the sample represents the DNA
profile. DNA profiles are distinguished from DNA samples by the type of
information they contain. DNA profiles contain information useful in differentiating
between individuals and establishing identity. They utilize non-coding fragments of
DNA, meaning the fragments used do not cany information regarding an individual’s
phenotypic, behavioural or pathological traits. In contrast to this DNA samples
contain entire genetic sequences of the individual from which they are drawn. DNA
samples contain an individual’s complete genetic code. Concerns regarding the
collection o f DNA generally refer to the collection of samples because of the wealth
o f personal information they contain, and its irrelevance for criminal identification.
This distinction will be taken up more fully in Chapter Five; here it is sufficient to
note that the autoradiograph is a DNA profile.
Once made visible, the DNA profiles from suspect and sample are compared.
Using a computer, the analyst(s) measure the specific displacement o f bands in the
suspect and sample profiles, comparing the measurements against one another. This
interpretation produces one of two results. If the displacement of bands in the sample
DNA and the suspect’s DNA are dissimilar the suspect can be removed from
suspicion, or exonerated if already convicted. If the two DNA profiles match, a
9 It is important to note that the specific fragments of DNA represented in the autoradiograph
are only a minutia of the entirety of the human genome. The sources referred to in the previous note
discuss the specific fragments of DNA that are used in analysis.
Ill
second procedure takes place. A probability is calculated to show the likelihood that
the matching DNA could have come from someone other than the suspect (but with
similar genetic makeup). Because only a fraction of the world’s population have had
their DNA analyzed, to calculate probability it is necessary to work with population
samples and statistics. Sample populations are broken down according to race-based
categories such as Caucasian, African-American, Asian and Hispanic. From here, a
numerical value is produced, such as “one in one million,” estimating the likelihood
that the sample DNA could have come from someone of the same sample population
other than the suspect. Together, the autoradiograph and the numerical probability
constitute the evidence used in DNA identification.10
In Chapter Three I argued that the fingerprint functions as a highly effective
immutable mobile. The mechanical objectivity of the camera captures the “natural
inference o f presence” o f the fingerprint and isolates it in the static image.11 The
m eaning, value and authority of the fingerprint remain unaffected as its photographic
representation moves seamlessly and immutably between contexts and locations. The
complexity o f DNA analysis immediately distinguishes it from fingerprinting. As it
operates from the molecular level, DN A analysis is dependent on inscriptions in a
way fingerprinting is not. DNA is only known through its material representation, the
autoradiograph. However, although DNA analysis is distinguished from
10A particularly good discussion of the calculation of probability can be found in The
Evaluation o f Forensic DNA Evidence.
11 The phrase is taken from George Wilton Wilton’s text on fingerprints. See Fingerprints:
History, Law and Romance (London: William Hodge and Company, 1938).
112
fingerprinting in this regard, it is similar in that the efficacy of the technique as visual
evidence depends on the ability of its inscriptions to function as immutable mobiles.
To function as an effective method of criminal identification the tremendous
complexity of DNA must be made into a coherent, unified evidentiary statement and
the information contained in the image must be made immutable.
Criticisms and concerns raised about DNA analysis pertain to the two-part process
of analysis. In the first instance, the technique is challenged based on the procedures
with which it was performed. The faulty collection of evidence, its mishandling or
misuse, and personal or laboratory error serve as sites o f contention. In the second
instance criticisms are leveled at the notion of population samples. Because the world
does not exist within strict race-based groups, critics argue, then sample populations
based on such groups will necessarily be flawed. In addition to these procedural
issues, DNA analysis continually raises concerns regarding the collection o f genetic
information. The storage and use of DNA information in databanks poses threats of
i 'y
discrimination against individuals based on their genetic makeup.
During the “DNA wars” o f the late 1980s and early 1990s proponents and
opponents of DNA analysis debated these issues and others related to the efficacy and
accuracy of the technique as it applies to personal identification. The debate took
place through newspapers, popular and specialized journals, government reports and
in the courtroom. The DNA Identification Act of 1993, enacted in 1994, attempted to
12For in-depth discussions of DNA analysis covering a diverse array of topics, perspectives
and approaches see: Billings, DNA on Trial; Mark A. Rothstein, ed. Genetic Secrets: Protecting
113
alleviate these concerns and to black box the technique.. As with earlier methods of
identification it was thought that standardized procedures would guarantee the
immutability of the evidence. Toward this end, the DNA Identification Act called for
the establishment of a DNA advisory board under the direction of the FBI to develop
and monitor standards for DNA analysis in law enforcement. This chapter turns to an
investigation of the “DNA wars,” from the introduction of DNA analysis into law
enforcement through the drafting of the DNA Identification Act in 1993. I locate the
lack of coherent procedures for analysis as the primary area of concern surrounding
the technique.
DNA From the Laboratory to the Courtroom
The first use of the technique known as “DNA fingerprinting” was by a British
geneticist, Alec Jeffreys, in March of 1985.13 Jeffreys’ method was patented by
Cellmark Diagnostics and was used primarily for paternity testing in relation to
immigration in the United Kingdom.14 In the United States, during the mid-to-late
1980s, the technique was performed by a series of private companies including
Cellmark Diagnostics of Maryland (the U.S. subsidiary of the British company which
Privacy and Confidentiality in the Genetic Era (New Haven: Yale University Press, 1997); “Contested
Identities,” a special issue of Social Studies o f Science, 28 no. 5-6 (1998).
13Jeffreys was the first to coin the term “DNA fingerprinting,” which has since been replaced
in favor of “DNA analysis,” “DNA typing,” or “DNA profiling.” DNA identification is based on the
calculation of probability and therefore does not determine absolute identification as fingerprinting
does. Because of this It was generally agreed that the technique’s affiliation to the standard practice of
fingerprinting was erroneous.
14Michael Lynch and Sheila Jasanoff, “Contested Identities: Science, Law and Forensic
Practice,” Social Studies o f Science 28 no. 5-6 (1998): 675-686; 676.
114
held the patent), Lifecodes Corporation of New York, Genelex of Washington State,
and Forensic Science Associates of Richland California. In 198S the Federal Bureau
of Investigation, became the first public laboratory to use the technique.
The transition of DNA analysis from the research laboratory to the courtroom was
surprisingly quick. Commenting on this move, the technique’s originator, Alec
Jeffreys, expressed Ms surprise in a Parade Magazine interview, stating: “it has
astonished me how rapidly the scientific community has used the technology, how
rapidly it has been viewed positively by many young lawyers.... I thought the
practical uses were years in the future.”15 The first conviction in the United States
based on DNA evidence was the 1987 case, State v. Andrews.16 The accused, Tommy
Lee Andrews, was convicted o f rape after DNA drawn from Ms blood matched that
found in a vaginal swab taken from the victim.
During the first few years of its use in cases of criminal identification, DNA
analysis held a reputation similar to fingerprinting. The technique was deemed
infallible and was likened to a “magic bullet” in criminal investigation by both
specialized and popular audiences. The use of the term “DNA fingerprinting”
throughout this early period attests to the technique’s successful reputation but also to
a general misunderstanding of its basis. As described at the start of this chapter,
DNA analysis identifies a person through the use of probability statistics. The
technique does not provide a unique identification as does fingerprinting. The
15Earl libel!, “Whodunit? Quick Check the Genes!,” Parade Magazine, March 31 1991: 12.
16State v. Andrews, 533, So.2d 841 (Fla. App. 5 Dist. 1988).
115
erroneous term “DNA fingerprinting” was gradually replaced with “DNA profiling,”
“DNA typing,” or “DNA analysis.”17 Throughout this Chapter I use the term “DNA
analysis” as it emphasizes the concrete processes of production and interpretation.
During the mid-to-late 1980s the reputation of DNA analysis as a method of
identification as equally infallible as fingerprinting was such that its incorporation
into criminal investigations regularly prevented cases from reaching trial.18
Combined with a general lack of familiarity with the technique, this reputation
prevented DNA analysis from being questioned. But when it did enter into trials,
DNA evidence confounded defense attorneys. In a New York Times article, Peter
Neufeid, a prominent defense lawyer and early critic of DNA evidence, commented
on the technique’s effect on the adversarial legal system. He notes:
In the first two dozen cases where DNA evidence was introduced, the
opposing attorney did not even challenge the evidence ... they felt
scientifically illiterate and unable to even perceive of questions. No
adverse experts were even retained by the counsel. Everyone just sort
o f lay down and died.19
17 It is interesting to note that contemporary private testing companies often use the term
“DNA fingerprinting” in advertisements for their services. This is an obvious yet inaccurate appeal to
the authority of fingerprinting.
18Peter J. Neufeid, “DNA - A Defense Perspective,” in Proceedings fo r the International

Symposium on Human Identification (Promega Corporation, 1989), 167-171.
19Gina Koiata, “Some Scientists Doubt the Value of'Genetic Fingerprint' Evidence,” New
York Times, January 29 1990, Al.
116
Whereas the Andrews case set a precedent as the first conviction involving DNA
analysis, the 1989 trial of Joseph Castro was the first to seriously question the
technique.20 Castro was accused of murdering a neighbor and her daughter in
February of 1987. Lifecodes Corporation was called in to perform DNA analysis for
the case. During a pre-trial hearing, lawyers Barry Scheck and Peter Neufeld and
population geneticist Eric Lander were among a team of lawyers and scientists that
questioned Lifecodes’ performance of the technique.21 As a result, the presiding
judge allowed some, but not all, of the DNA evidence to be used. As the New York
Times reported:
the judge held that DNA tests could be used to show that blood found
on the suspect’s wristwatch was not his. But the judge said other DNA
tests that purported to show it was the victim’s blood were
[procedurally] flawed and would not be admissible as evidence.
The next case involving DNA analysis which would prove to have lasting
significance was the 1990 murder trial, United States v. Yee.23 In this case,
defendants Steven Wayne Yee, Mark S. Verdi and John Ray Bonds were accused of
murdering David Hartlaub. The case would prove highly influential because of the
way in which the DNA evidence was handled. The judge in the case, James E. Carr,
20 People v. Castro, 545 N.Y.S. 2d 985 (Sup. Ct. 1989).
21 Lynch, “Contested Identities,” 678.
22 Robert D. McFadden, “Reliability of DNA Testing Challenged by Judge's Ruling,” New

York Times, August 15 1989, B1.
23 United States v. Yee, ND, Ohio 129 FRD 692 (1990).
117
gathered thirteen experts on DNA for a six-week hearing. The experts covered a
variety o f dements of DNA testing from procedural details to the use of population
samples in the analysis.24 Carr’s ruling after the six-week hearing was highly
debated. The thirteen experts returned with views that were, in many cases,
diametrically opposed. Despite this, Carr ruled that no unanimity or consensus was
needed for the evidence to be used in court.
Carr’s ruling in the Ohio case was provocative, as it seemed to contradict the only
existing precedent for scientific evidence in the courtroom. This vague precedent
came from the 1923 trial of James Frye.25 The resultant Frye principle stated that to
be accepted in court, any new scientific technique “must be sufficiently established to
have gained general acceptance in the particular field in which it belongs.” Carr’s
ruling in the Ohio trial directly refuted the Frye principle since he did not call for
consensus among Ms thirteen experts.
I stress that the immediate acceptance of DNA analysis as evidence in court can be
attributed largely due to the construction of the autoradiograph as a coherent and
unified evidentiary statement. Recall that DNA analysis originally developed within
24 Katherine Bishop, “A Victory for Genetic Fingerprinting,” New York Times, November 16
1990, B6.
25 Frye v. United States 293 F. 1013 (D.C. Cir. 1923).
26 The Frye principle is not the only method to determine the validity of a scientific technique,
although it was the most widely used during early uses of DNA analysis. The Federal Rules of
Evidence also offer a “helpfulness” model. The 1993 case, Daubert v. Merreil Dow Pharmaceuticals
Inc., offered a model parallel to that of the “helpfulness” model. For a discussion of these see: DNA
Technology in Forensic Science; Jasanoff, "The Eye of Everyman;” Lynch “Contested Identities.”
118
research laboratories and for use in paternity testing. Within these practices, DNA
analysis avoids many of the problems that plague its use in. criminal identification. In
non-forensic uses samples are abundant and contamination-free and tests do not rely
on the use of population samples. Paternity tests generally involve determining a
match or exclusion based on readily available samples of all parties involved. The
practice o f identification by DNA developed within this relatively idealized
environment.
When DNA analysis made the transition from the laboratory to the courtroom it
entered the latter as a coherent visual statement. Before the method was being used in
criminal trials its language was being created and revised within the controlled setting
of corporate laboratories. By the time it reached criminal trials, the product o f this
new method o f identification, the autoradiograph, was constructed as a newer, more
scientific, more technological manifestation o f the fingerprint. The autoradiograph
entered the practice of criminal identification as a genetic “fingerprint:” it was a
coherent, unified evidentiary statement that was understood to guarantee
identification. Once its cycle o f usage began it had a head-start on oppositional views
which first had to decode the language and then reformulate it to express the opposing
view. Thus, as Peter Neufeld explained, defense attorneys In the early court cases
“just sort of lay down and died.”27
When DNA evidence was eventually questioned, as in the Castro and Yee trials,
its status as a coherent, unified evidentiary statement was disrupted. The
27Neufeld*s fiili remarks are cited on page 115.
119
autoradiograph was deconstructed, revealing the fragmented and specialized
processes behind its construction. The image was shown to be the product of
complex and diverse scientific practices and its interpretation was bound to the
equally complex subject of population genetics. Once its unity was disrupted,
lawyers and scientists were able to question the autoradiograph based on the elements
of its construction. This is apparent in the pre-trial hearings in Castro and Yee where
experts challenged the procedures used in performing DNA analysis and the use of
population samples in the calculation of probability.
Developing Standards: The FBI in the DNA Identification Act
The immediate and unquestioned acceptance of DNA analysis in criminal
identification and its eventual contestation in the Castro and Yee trials points to the
contentious nature of the technique. A primary reason for its problematic
introduction to criminal identification was that, unlike Bertilkmage and fingerprinting
which were performed within the specific confines of criminal justice institutions,
DNA analysis was performed by numerous and different companies and laboratories,
each with its specific procedures of analysis. With independent labs such as Ceilmark
and Lifecodes producing the tests there was no way to assure consistency in the
method. In a 1988 New York Times article, Congressman Don Edwards (D, CA)
warned of “certain cautionary reservations about private firms galloping in and
testifying all over the place, because the procedures, the protocols, the standards have
120
not been established yet,”28 Eric Lander, a central early critic of the technique (and
also a critic of Lifecodes’ analysis during the Castro trial), noted the very serious
consequences of procedural inequalities, suggesting: “clinical laboratories have to
meet higher requirements to be allowed to diagnose a case of strep throat than a
forensic laboratory has to meet to put a defendant on death row.”29
Several professional groups and committees formed in response to the contentious
nature of DNA analysis. The rapidity o f the introduction of DNA evidence to the
legal system forced the issue of standardization into the foreground. The American
Society o f Crime Laboratory Directors (ASCLD), founded in 1974, established a
Laboratory Accreditation Board (ASCLD-LAB) in 1985.30 This board was initiated
by the FBI to bring together local and FBI forensic laboratories and provided a
voluntary accreditation program to crime laboratories, including those performing
DNA analysis. Whereas the American Society o f Crime Laboratory Directors offered
laboratory-level accreditation, the American Board o f Criminalistics, formed in 1989,

•5 |
provided certification to individual practitioners. Towards the end of the 1980s, and
again under the direction o f the FBI, a group of thirty scientists from forensic
laboratories in the United States and Canada formed the Technical Working Group on
28 Harold M. Jr. Schmeck, “DNA Findings Disputed by Scientists,” New York Times,
February 7 1988, 1.
29 Schmeck, “DNA Findings,” 1.
j0 DNA Technology in Forensic Science, 102. General information on the ASCLD, including
its history, can be found at the Society’s website, http://www.ascld.org.
31 Keith Inman and Norah Rodin, An Introduction to Forensic DNA Analysis (New York:
CRC Press, 1997), 139. See also the website of the American Board of Criminalistics at
http://criminalistics.eom/abc/A.php.
121
DNA Analysis and Methods (TWGDAM).32 This group did not provide accreditation
or certification; instead, it was an arena through which forensic scientists met to
discuss, monitor and examine various aspects of the technique including standards for
analysis. Finally, in direct response to the growing concern about DNA as evidence,
the National Academy of Sciences established the Committee on DNA Technology in
Forensic Science in 1990. Despite differences in their individual approaches, each of
these groups sought to alleviate the contentious nature o f DNA analysis through the
creation and regulation of standards. Two of these initiatives, both developed by the
FBI, the American Society of Crime Laboratory Directors’ Laboratory Accreditation
Board and the Technical Working Group on DNA Analysis and Methods, remain
prominent forces in shaping and monitoring DNA analysis.
The search for standards was also carried out through detailed governmental
reports. The first o f these was the 1990 Office o f Technology Assessment’s Genetic
Witness: Forensic Uses o f DNA Tests.33 The report was a response to the quick
movement o f DNA into the courtroom and to the contentious ruling in the Castro
trial. It provided a general review o f DNA technology, including discussions o f how
DNA is tested, how it is used in court, and the potential use and misuse of genetic
information by external parties. Early in the report a portion of highlighted text calls
attention to standardization: “Setting standards for forensic applications of DNA
32 Inman, An Introduction, 98.
33 Genetic Witness: Forensic Uses o f DNA Tests (Washington, D.C.: United States Congress,
Office of Technology Assessment, 1990).
122
testing is the most controversial and unsettled issue. Standards are necessary if high-
quality DNA forensic analysis is to be ensured, and the situation demands immediate
attention.”34 The 1990 report does not offer any specific conclusions on DNA
analysis. Instead it suggests, as in the passage above, that with the accelerated use of
DNA as evidence, setting standards was the pressing issue.
The second major report came in 1992 with the National Research Council’s DNA
Technology in Forensic Science?5 Unlike its 1990 predecessor, this report set out to
establish specific goals and recommendations. The Advisor to the Committee on
DNA Technology in Forensic Science, Victor A. McKusick, noted that the report
issued out o f the increasing concern about the use of DNA evidence in court.
Concluding his preface to the report, McKusick writes:
DNA typing for personal Identification is a powerful tool for criminal
investigation and justice. At the same time, the technical aspects of
DNA typing are vulnerable to error, and the interpretation of results
requires appreciation o f the skills o f population genetics. These
considerations and concerns arising out o f the felon DNA databanks
and the privacy o f DNA information made it imperative to develop
guidelines and safeguards for the most effective and socially beneficial
use o f this powerful tool.36
34 Genetic Witness, 10.
35 DNA Technology in Forensic Science, 185.
36 DNA Technology in Forensic Science, viii.
123
Again, as in the 1990 report, the National Research Council’s analysis attempted to
take into consideration the wide array of issues at work in DNA analysis. The report
includes chapters on technical considerations, the calculation of probability statistics,
the development of standards, DNA databases, and the social implications of these
practices. Each chapter also includes a brief summary of the committee’s
recommendations. But once again the development of standards for DNA analysis
took precedence.
The committee suggested that due to the powerful nature of the technique
(referring to its scientific potential and public benefit), “some degree of
standardization of laboratory procedures is necessary to assure the courts of high-
quality results.”37 The committee then detailed ways in which such standards could
best be achieved. Among the recommendations was a call for quality assurance and
quality control programs within individual laboratories. It was also suggested that
laboratories and individual practitioners should be required to seek professional
accreditation. To ensure that these recommendations were met appropriately, the
report also called for the development o f a professional organization to monitor these
activities. And finally, the report recommended that the federal government enact a
committee to oversee these processes and to ensure that the highest standards would
be met in each casef8
37 DNA Technology in Forensic Science, 98.
38 DNA Technology in Forensic Science, 97-110.
124
The search for standards in DNA analysis through the reports, committees and
groups detailed above is similar to law enforcement's search a century earlier for
standards in Bertillonage and fingerprinting. The result sought after in each case was
the same: standardized procedures for the creation of evidence would transform its
visual representation into an immutable mobile. The image’s certification as having
been produced according to accepted standards would guarantee its authority and
render the image’s unique and fragmented context of production invisible. The
standardized autoradiograph could move seamlessly and immutably between contexts
and locations. Without standards, autoradiographs could be challenged based on their
specific contexts of production and interpretation. The move to standardization in
Bertillonage, fingerprinting and DNA analysis took place through the control o f
inscriptions.
The development o f a uniform set o f procedures for the production o f DNA
evidence was a central concern behind the drafting o f the DNA Identification Act in
1993.39 Congressman Don Edwards sponsored the DNA Identification Act to provide
legislation surrounding the use o f DNA technology in criminal justice practices.
Recall that Edwards earlier warned of the dangers in having several independent
companies performing DNA tests. Edwards’s concern was that these private
companies would each perform the technique in an individual way and that this
would result in a lack o f consistency. As sponsor o f the 1993 Act, Edwards promoted
39 DNA Identification Act o f 1993, (Washington D.C.: United States House of

Representatives, 1993). The Act appears as an amendment to Title I o f the Omnibus Crime Control
and Safe Streets Act o f 1968.
125
the establishment of a regulated set of standards to be used in future cases involving
DNA analysis.
The DNA Identification Act covers five areas: funding to improve availability of
DNA analysis for criminal investigation; quality assurance and proficiency testing;
the development of an index for the exchange of genetic information; external testing
for the FBI laboratory; and the authorization o f funds. The Act includes a statement
of summary and purpose that is particularly clear:
The purpose of the DNA Identification Act o f 1993 is to promote the
use o f DNA identification technology for law enforcement purposes,
in accordance with appropriate quality control and privacy standards.
The Act authorizes federal funding for State and local governments to
establish or improve DNA forensic capabilities; requires grant
recipients and the FBI laboratory to adhere to minimum quality
control, proficiency testing and privacy standards; requires the FBI to
establish a forensic DNA advisory policy board to provide the FBI
with advice from a range o f DNA experts, including scientists from
outside the forensic field; and authorizes the FBI to establish a national
index o f DNA profiles of convicted offenders, pursuant to minimum
quality control, proficiency testing and privacy standards. The bill
does not set standards for the admission o f DNA evidence in court;
126
that question is left to the law of evidence and must be adjudicated on
a case-by-case basis.40
From this statement it is clear that the DNA Identification Act was drafted as a
conscious appeal to the concerns and recommendations set out in the 1992 National
Research Council report. 'The Act also acknowledges recommendations suggested by
the Office of Technology Assessment in its 1990 report. Moreover, in its deferment
to the courts, the document even refers back to the Frye principle. The 1993 Act
reads as a point-by-point response to the various criticisms leveled at the technique.
The Act did not go uncontested. A chief concern of critics was the control of
standards. Under subsection three o f the Act it is stated: “Not later than 180 days
after the date o f the enactment of this Act, the Director o f the Federal Bureau of
Investigation shall appoint an advisory board on DNA quality assurance methods.”41
Following this:
The Director o f the Federal Bureau of Investigation, after taking into
consideration such recommended standards, shall issue (and revise
from time to time) standards for quality assurance, including standards
for testing the proficiency o f forensic laboratories, and forensic
analysts, in conducting analyses o f DNA 42
40 DNA Identification Act, 4.
127
Under these provisions the development, implementation and control of standards for
performing DNA analysis in law enforcement practices would remain under the
direction of the FBI.
Responding to the FBI’s authoritative role, Barry Scheck and Peter Neufeld (from
the Castro trial and later part of the O J. Simpson defense team) argued that the FBI
depends on a certain success rate and therefore allowing this agency to establish DNA
standards would be problematic.43 In their view, the FBI would be able to define
privacy and proficiency criteria and could manipulate these criteria to serve their own
needs.44 John Hicks, then Assistant Director o f the FBI, responded by calling Scheck
and Neufeld’s suggestion ludicrous. “Where you have the technological competence
and mission,” Hicks defended, “that’s where you want to place your confidence.”45
Don Edwards, the sponsor o f the Act, took a similar stand. While Edwards
acknowledged that there were still problems with identification by DNA analysis he
ultimately insisted: “We’ve got to get it started, and the FBI is all set up to move
ahead quickly.”46
^ Scheck and Neufeld have been among the most vocal critics of DNA analysis throughout its
development. In 1992 the two formed the Innocence Project in the Benjamin N. Cardozo School of
Law. The project uses postconviction DNA evidence to prove the innocence of individuals wrongly
imprisoned. It is a non-profit “clinic” meaning that students work on cases under the supervision o f
lawyers. As o f November 27,2002 the program has exonerated one hundred and sixteen individuals.
Information is available from the program’s website, http://www.innocenceproject.org.
44 Naftali Bendavid, “Bill Would Expand Use of DNA Testing,” Legal Times, June 7 1993.
45 Bendavid, “Bill Would Expand.”
46 Bendavid, “Bill Would Expand.”
128
Despite continued debate, the DNA Identification Act of 1993 was adopted into
the Violent Crime Control and Law Enforcement Act on September 1 3 , 1994.47 The
specific DNA Identification Act appears under Title XXI (State and Local Law
Enforcement), Subtitle C (DNA Identification). Following on the heels of the
enactment, Eric Lander and Brace Budowle proclaimed that “the DNA fingerprinting
wars are over.”48 This triumphant remark appeared in a 1994 issue of Nature. The
combination o f previously opposing forces as Lander was an early critic of the
technique and Budowle was Director o f the FBI’s DNA typing laboratory, gave
particular and peculiar weight to their statement.49 In this article, Lander and
Budowle suggested that the National Research Council’s recommendation for a
national committee on DNA analysis had been met with the enactment o f the 1993
Act. For the authors this meant that the debate surrounding DNA analysis had been
solved. Their concluding remarks place central importance on standardization:
Most of all, the public needs to understand that the DNA fingerprinting
controversy has been resolved. There is no scientific reason to doubt
the accuracy o f forensic DNA typing results, provided that the testing
laboratory and the specific tests are on par with currently practiced
47 Violent Crime Control and Law Enforcement Act o f 1994 (Washington, D.C.: United
States Senate and House of Representatives, 1994).
48 Eric S. Lander and Bruce Budowle, “DNA Fingerprinting Dispute Laid to Rest,” Nature
371, no. October 27 (1994): 735-738: 735.
49 Saul Halfon provides an interesting and concise account of Erie Lander’s unique position
within the DNA wars. See: “Collecting, Testing and Convincing: Forensic DNA Experts in the
Courts,” Social Studies o f Science 28, no. 5-6 (1998). Michael Lynch and Sheila Jasanoff note that the
article by Lander and Budowle was also criticized as being an attempt to pre-empt attacks on DNA
evidence in the upcoming O.J. Simpson trial. See: “Contested Identities.”
129
standards in the field. The scientific debates served a salutary purpose:
standards were professionalized and research stimulated.50
The “DNA wars” of the late 1980s and early 1990s were essential in developing,
questioning and testing DNA analysis for criminal identification. Lander and
Budowles’ remark that the activity of the “DNA wars” served its “salutary purpose”
belittles the importance of critical debate and discussion in practices as far reaching
as DNA analysis. The various government reports, scientific and popular articles and
legal battles o f this period were essential in providing much needed critical
investigation of a technique that posed serious questions regarding individual rights
and freedoms and one that problematized the intersections o f science, law
enforcement and criminal justice. The development o f committees and pre-trial
hearings to debate the accuracy, efficacy and social ramifications o f genetic
identification emphasizes the problematic nature o f the technique. Attempts to
alleviate the controversy surrounding DNA analysis within this context took place
almost exclusively through the search for standards o f analysis.
The formalization of standards for DNA analysis under the DNA Identification
Act did much to allay the concerns, highlighted by Congressman Edwards, regarding
inconsistent analyses. However, standardized procedures did not wholly resolve the
controversy surrounding the technique. In response to Lander and Budowle5s claim
in the journal Nature that “the DNA fingerprinting wars are over,” the following two
50 Lander and Budowle, “DNA Fingerprinting,” 738.
130
volumes of the journal included articles suggesting the wars were very much still on.
In the December 1 issue of the journal, Richard C. Lewontin, of Harvard University,
accused the authors of “completely distortfmg] the current situation” in discussions of
DNA analysis.51 For Lewontin serious questions remained concerning laboratory
reliability, the calculation of probabilities and the ability of jurors to make logical
decisions based on probabilities. The second of these concerns, the calculation of
probabilities, was the subject of another response to Lander and Budowle appearing
In the same issue, authored by Daniel L. Haiti, a colleague of Lewontin5s at Harvard.
The debate did not end there. Five more responses appeared in the January 12 edition
o f the journal. The concerns associated with the calculation o f probabilities and with
the use of population samples were extensive enough to warrant a second National
Research Council report in 1996.52 In addition, Sheila Jasanoffhas suggested that,
contrary to solving issues associated with Inconsistent analyses, standardization
allowed DNA evidence to be challenged in a new way. By providing a step by step
account o f how DNA evidence should be performed, she argues, established
standards allowed defense attorneys to map the actual performance of the technique
against its ideal model.53
51 “Correspondence: Forensic DNA Typing Dispute,” Nature 371, December 1 1994: 398-
399.
52 The Evaluation o f Forensic DNA Evidence. Also see: Donald A. Berry, “Statistical Issues
in DNA Identification,” in Billings, DNA on Trial, 91-108.
5j Jasanoff, “The Eye o f Everyman.”
131
Although standardization did not wholly resolve the conflict surrounding DNA
analysis, I stress that it had two significant and far-reaching effects. First, I argue that
the enactment of the DNA Identification Act in1994 and the standardization of
procedures under the control of the FBI had the effect of streamlining DNA analysis
in law enforcement, transforming the fragmented inscriptions of its production into an
immutable and mobile autoradiograph. In a 1992 essay on the potential uses and
applications of DNA analysis Paul Billings warns:
when a scientific discovery is sped into application, inadequate time
may be allowed for the scientific community to scrutinize the
techniques, for the peer review process to eliminate spurious
Information, and for controversies to be resolved.”54
The comments of John Hicks and Don Edwards cited earlier in this section illustrate
the extent to which DNA analysis was sped into application. I argue that
standardization sped DNA analysis into application and brought a premature end to
much of the critical Investigation of the technique.
The second significant effect o f standardization was that is served as a necessary
precondition for the development of “dominance” over DNA analysis in law
enforcement by the FBI. As Latour notes, dominance over inscriptions “Is not a
given but a slow construction.”55 I stress that while standardization was an essential
component in black-boxing DNA analysis in law enforcement, It was not in itself
54Paul R. Billings, “Gene Technology: Views of its Criminal Justice Applications,” in

Billings, DNA on Trial, 1-4:2.
55Latour, “Drawing,” 56.
132
sufficient for this task. As part of a process of slow construction, standardization
worked in concert with existing FBI initiatives toward this goal. To understand the
construction of the FBI’s dominance over inscriptions as well as the ramifications of
that position it is necessary to address these initiatives and their intersections with the
DNA Identification Act. Specifically, in addition to developing FBI controlled
standards for DNA analysis, the DNA Identification Act effectively delineated a
workable boundary for the FBI within the larger field o f DNA analysis and provided
them with funding necessary to function In what is a highly competitive market.
Disseminating Standards and Developing Dominance over Inscriptions
The DNA Identification Act defines its purpose as being concerned with the
promotion of “DNA technology for law enforcement purposes.” I stress that the
explicit reference to “law enforcement” was necessary in establishing a workable
boundary within the larger, competitive and profit-driven field o f DNA analysis. In
Ms examination o f DNA-typing companies Arthur Daemmrich notes that
biotechnology companies derive substantial profits from paternity testing and are
attracted to criminal cases because of the visibility and credibility they provide.56
Cellmark Diagnostics’ (now OrcMd Cellmark) 2002 fees range from $600.00 to
56Arthur Daemmrich, “The Evidence Does Not Speak for Itself: Expert Witnesses and the
Organization ofDNA-Typing Companies,” Social Studies o f Science 28, no. 5-6 (1998): 741-772. For
a discussion from the point of view of biotechnology companies see: P.G. Debenham, “DNA
Fingerprinting; a Biotechnology Business,” in DNA Fingerprinting: Approaches and Applications, ed.
Terry Burke, et a!. (Basel, Switzerland: Birkauser Verlag, 1991), 342-348. Debenham was an
employee for Cellmark Diagnostics at the time of writing.
133
$1500.00 (USD) per sample for DNA analysis and $2000.00 (USD) plus expenses per
day for expert witness testimony.37 With the substantial remuneration of DNA
analysis, individual and corporate reputations and careers can be shaped through
participation in civil and criminal cases. Private laboratories, therefore, have a vested
interest in both the production and the interpretation of their product. Toward this
end, Daemmrich notes that these corporations control all aspects of the creation and
sale of their product He describes the reasoning for this practice:
All DNA-typing companies face formidable challenges in presenting a
credible product to courts. Only by tightly coupling the company
organization and testing procedures to the expert witness do they gain
the competitive edge needed to enter the courtrooms around the
country and, in each, to persuade a different cast o f judges, lawyers,
and jury.58
DNA-typing companies create a monopoly over their specific product. This allows
them to monitor and guarantee every aspect of its production and interpretation. This
is essential in the for-profit business o f DNA analysis because the product, the
autoradiograph, represents not only the results of the analysis but the credibility of the
corporation performing the test and its employees.
57 Cellmark’s 2002 fee schedule can be found on their website at http://www.ce!!mark-

labs.com.
58 Daemmrich, “The Evidence,” 749.
134
During the mid-1980s DNA analysis was the domain of private biotechnology
companies such as Cellmark. These companies competed for dominance over the
technique: the company that produced the test inscriptions and immutable mobiles, in
this case the best autoradiographs, procured higher profits and greater visibility
through participation in high profile cases. Therefore, when the FBI opened its
laboratory in 1988 it was entering an already competitive, profit-driven market. It
had to compete against well-funded biotechnology corporations that were already
establishing themselves within civil and criminal cases, and that had developed highly
specialized and controlled procedures for producing their product, the autoradiograph.
The DNA Identification Act was essential in delimiting a specific field of DMA
analysis within which the FBI could effectively compete and, ultimately, control.
Importantly, although the DMA Identification Act established FBI regulated
standards for DNA analysis in law enforcement it did not prevent private
biotechnology companies from performing the technique. These companies are able
to perform analysis in forensic cases provided they adhere to the FBI’s standards.
Preventing private companies from performing forensic work would have appeared as
an overt attempt to control the technique and would have caused wide-spread concern
among scientific, academic and professional communities. However, I stress that the
exclusion of biotechnology companies would also have eliminated a valuable
resource for the FBI. DNA analysis is a time consuming and expensive process. The
technique is still in its infancy and is constantly undergoing development. The
collection o f DNA samples within law enforcement proceeds at a rate considerably
135
faster than that of its analysis producing large backlogs of samples. Law enforcement
agencies often rely on biotechnology corporations, which have the most up-to-date
equipment and numerous staff, to analyze these backlogs of DMA samples. Orchid
Cellmark lists law enforcement agencies in South Dakota, Oregon, Wyoming, New
York and Nevada as former “clients” and notes that they are “currently under contract
to perform the STR [short tandem repeat] analysis of thousands of backlogged sexual
assault kits from the New York City Police Department, the Illinois State Police, and
the Phoenix. Police Departments.”59 Similarly, Genelex advertises that they have
“completed DNA profiles under contract on more than 25,000 individuals that are
currently in law enforcement databanks, including the FBI’s National DNA Index
System.”60
The utilization of these biotechnology companies assists law enforcement agencies
in clearing backlogs of DNA samples and in amassing genetic identification data. It
also serves to further mobilize the FBI’s standards o f analysis by disseminating them
throughout the private DNA-typing companies. At the very least this dissemination
of FBI standards strengthens the Bureau’s position of dominance within DNA
analysis in law enforcement. However, it also threatens to influence and interact with
other applications, such as paternity testing and medical and genetic research. This
59From Celimark’s website at http://www.ceIlmark-Iabs.com/companypfofiIe.htoiI.

Originally accessed in November 2002, the page has since been removed. For information on
Cellmark laboratories see their website at http://www.cellmark-Iabs.com, or that of their parent
company, Orchid Biosciences at http://www.orchid.com.
60From the forensic page of Genelex, see

http://www.genelex.coni/forensics/forensicshome.fataiL Accessed November 2002.
136
issue and others associated with the databanking of DNA profiles will be addressed at
length in Chapter Five.
The dissemination of FBI standards for performing D N A analysis was equally
important within law enforcement agencies. As was the case with fingerprinting at
the turn of the nineteenth and twentieth centuries, the potential of DNA analysis as a
method of criminal identification was quickly seized and Implemented by individual
law enforcement agencies throughout the country.61 I argue that, by legalizing and
funding the establishment o f a federal level DNA database, called the National DNA
Index, the DNA Identification Act disseminated the FBI’s standards of analysis
throughout local and state law enforcement agencies, strengthening their control over
the technique.
The National DNA Index has its roots in the establishment of the Combined DNA
Index System (CODIS). This was a program initiated by the FBI in 1990 to collect,
store and exchange genetic identification data between law enforcement agencies.
The potential databanking o f genetic information was a primary concern of critics
from the inception o f DNA identification. Concerns were raised that this information
could be accessed and shared between law enforcement, health, insurance and other
agencies and used to discriminate against individuals. Critics also expressed concern
that a large genetic index developed for use in criminal identification would
eventually expand to other uses, posing threats to personal privacy. I have omitted
61 On the development of DNA databanks in law enforcement see: Jean E. McEwen and
Phillip R. Reilly, “A Review o f State Legislation on DNA Forensic Data Banking,”American Journal
o f Human Genetics 54 (1994): 941-958; Jean E, McEwen, “Forensic DNA Data Banking by State
Crime Laboratories,” American Journal o f Human Genetics 56 (1995): 1487-1492.
137
the discussion here to focus on the attempts to universalize procedures and because
the issue is addressed in Chapter Five. Nonetheless to fully understand the slow
construction of dominance over DNA analysis and the ramifications of that position it
is necessary to touch briefly on the development of the National DNA Index.
In order for a program such as the Combined DNA Index System to be effective
its data, DNA profiles, have to be standardized. The notion of different organizations
using different methods of analysis and producing different results is antithetical to
the proper function of communication programs such as CODIS. The FBI’s
initiatives to standardize procedures through both the American Society of Crime
Laboratory Directors and the Technical Working Group on DNA Analysis and
Methods were essential pretexts for CODIS. The enactment of the DNA
Identification Act further mobilized CODIS by universalizing its data and
establishing the National DNA Index. Importantly the DNA Identification Act also
provided the FBI with $40,000,000 over five years to implement the National Index
and CODIS. A significant portion of the money was to be used for establishing DNA
analysis facilities and databases at the local and state level. This substantial financial
allotment gave the FBI the leverage necessary to strengthen their position within the
field o f DNA analysis. In contrast to the high fees of private laboratories, the FBI
will perform DNA analysis free o f charge.62 Crucially, in order to access the FBI’s
services, law enforcement agencies must accept FBI standards for DNA analysis.
62 Private companies can perform analyses faster that the FBI, which remains plagued by a
backlog o f DNA evidence to analyze. This has caused some agencies to opt for paying for analysis.
138
Local and state agencies that agree to this requirement receive CODIS software,
training for the program and continued support from the FBI at no cost. The benefits
include access to the national database and the backing and credibility of the FBI
laboratory in cases of identification.
The implementation of a program such as CODIS and its National Index serves to
disseminate FBI procedures for performing DNA analysis and thus to further
strengthen their position of dominance over the technique. This process was
expedited through the reallocation of funds outlined In the DNA Identification Act.
In the original 1994 Act the $40,000,000 allotted to the FBI was to be distributed
from 1996 to 2000 in increments of 1, 3, 5 , 13.5, and 17.5 million dollars
respectively. An amendment to the Act, entitled the “DNA Identification Grants
Improvement Act o f 1995,” redistributed the funds into the Increments o f 1 , 15,14,6,
and 4 million dollars.63 The amendment explains that the FBI and the American
Society o f Crime Laboratory Directors “recommended that this modification be made
because of the significant start-up costs to States in creating DNA testing programs
and databases.” I argue that the redistribution of funds also functioned to expedite
control over DNA analysis. By amassing the support of state and local laboratories
through the implementation of CODIS, the FBI hastened its control over inscriptions.
The redistribution o f funds In 1995 enabled state and local laboratories to implement
63“DNA Identification Grants Improvement Act o f 1995,” (Washington, D.C.: United States
House of Representatives, 1995).
139
CODIS quickly, further strengthening the FBI’s dominant position and limiting
opposition.
Participation in CODIS by law enforcement agencies also gives the FBI access to
the various DNA databases of participating agencies, giving rise to a centralized
database of genetic information. The aggregates of genetic data are subject to the
dangers associated with law enforcement practices based on latent identification
described in Chapter Three and suggest distinct possibilities for misuse. Databases of
DNA could be accessed according to changing notions of criminality, a practice that
threatens to discriminate against individuals based on variables including race, gender
and sexual orientation. Genetic data could be used to track and identify criminal
behaviour patterns, or to research a biological basis for crime. The potential misuse
o f this information is further problematized by the interaction o f forensic and non-
forensic DNA analysis practices facilitated through CODIS. DNA databases are
constructed by law enforcement agencies, the military, private biotechnology
companies, health and medical institutions and university laboratories.64 Information
derived from non-forensic Investigation such as that o f the Human Genome Project
could be used in conjunction with law enforcement databases to geneticize crime.
Discussing the central role of the FBI in DNA analysis Simon Cole writes, “as it
did with fingerprinting, the FBI is positioning Itself less as the keeper of a national
database than as coordinator and facilitator for the communication of criminal
64Jean E. McEwen, “DNA Data Banks,” in Genetic Secrets: Protecting Privacy and
Confidentiality in the Genetic Era, ed. Mark A. Rothstein (New Haven: Yale University Press, 1997),
231-251:231.
140
identification information between states.” fo I argue that it is precisely the role of
coordinator and facilitator that affords the FBI control over DNA analysis and
heightened power within law enforcement. In conjunction with the National Index,
CODIS disseminates the FBI’s procedures and protocols throughout laboratories
across the country7. This not only strengthens their position of dominance within
DNA analysis but also gives them access to large and diverse repositories o f genetic
information. In a self-perpetuating process, the production, reproduction and
dissemination of autoradiographs validates the FBI’s authority and extends the
boundaries of its reach.
Conclusion
When DNA analysis entered the courtroom for use in criminal identification it did
so as a highly regarded scientific practice. Its specialized nature prevented it from
being questioned by lawyers, judges and juries who were forced to rely on the
interpretations o f “experts.” The evidence, the autoradiograph, functioned as a
genetic fingerprint. It was taken to testify to the absolute identity of the criminal. As
it was developed within the controlled environment of the research laboratory, the
autoradiograph entered the courtroom as an immutable mobile. The unique variables
o f its construction were rendered invisible in the image enabling it to move
seamlessly and immutably between laboratory and courtroom. Beginning with the
65 Simon Cole, Suspect Identities: A History o f Criminal Identification and Fingerprinting

(Cambridge: Harvard University Press, 2001), 294.
141
Castro trial defense attorneys and scientists started to pose questions regarding the
production and interpretation of DNA evidence. The autoradiograph was
deconstructed and challenged based on two of its constituent features: its specific
method of production and the use of population samples in the calculation of
probability. What entered the courtroom as an immutable mobile was made mutable.
The DNA wars which were waged through professional and public spaces and
contexts including journals,, government reports, newspaper articles, books, and the
courtroom served as necessary critical forums for the investigation of a tremendously
powerful and far-reaching method of personal identification. These wars addressed
the two central problems facing DNA analysis: its production and the use of
population samples. However, they also pointed to the larger social ramifications of
DNA analysis such as its potential impact on personal privacy and individual rights.
In its attempt to address the contentious issues of the DNA wars, the DNA
Identification Act established universal procedures for the technique to be used in
cases of criminal identification. This was seen as a positive step in alleviating the
procedural inequalities of DNA analysis, as the technique was performed by
numerous private and public laboratories.
In this chapter I have argued that the FBI gained a position of dominance over
DNA analysis in law enforcement practices and that the DNA Identification Act
concretized this position. The essential precondition for the Bureau’s dominance was
the development and regulation of standards for the production o f autoradiographs.
However, standardization alone was not enough to cede control over the technique to
142
the FBI. The DNA Identification Act delimited the authority of the FBI over DNA
analysis to Its specific use In law enforcement practices. This enabled the FBI to
differentiate its practices from those of highly funded biotechnology corporations that
were already established in the competitive market of DNA analysis. Further, the
DNA Identification Act legalized the construction of a National DNA Index, and
provided the FBI with $40,000,000 funding to expand their Combined DNA Index
System. This had the effect of disseminating the FBFs procedures for performing
DNA analysis throughout local and state law enforcement laboratories. The
standardization of DNA analysis, the specific delimiting of the FBFs authority over
the technique in law enforcement, and the allotment of funds for the dissemination of
CODIS were the essential components of the FBFs construction of dominance over
the technique.
The potential consequences of the FBFs dominant position within DNA analysis
and programs such as the Combined DNA Index System and the National DNA Index
are far from being mere theoretical concerns. In a 1992 essay, Jeroo S. Kotval, a
member of the New York State legislative commission on Science and Technology
described that state’s confrontation with the FB I.66 During the “DNA wars” New
York State legislature sought to alleviate concerns by requiring that all tests used in
state cases be accredited by the State’s Department of Health, a notion that would
have only strengthened the credibility of the evidence. The FBI opposed this
66Jeroo S. Kotval, “Public Policy for Forensic DNA Analysis: The Model of New York
State,” in Billings, DNA on Trial.
143
proposition, specifically rejecting the notion that their own tests would have to
receive further accreditation by New York State’s Department of Health. Kotval
notes that in their opposition, the FBI presented the State a clear choice:
If the state were to require the FBI test to be validated and licensed by
state authorities before introduction into New York State cases, the
FBI would refuse both to do the test for New York State and to allow
it access to the nationwide FBI databank.67
Kotval’s essay highlights the rhetorical nature of the “choice” facing states as well
as the ramifications of their decision. Success stories of CODIS linking crimes and
identifying serial offenders throughout participating states, combined with the
severity of cases for which DNA analysis is used (sexual and violent offenses) would
assure significant public pressure for states to participate in the program. When
CODIS began as a pilot project in 1990 there were fourteen participating local and
state laboratories. Currently, over one hundred and thirty laboratories throughout
forty-four states, the FBI, the Army and Puerto Rico participate in the program.68
These laboratories and those of private companies such as Orchid Cellmark and
Geneiex that perform analyses for law enforcement agencies are continually
collecting, analyzing and storing DNA information according to FBI protocols.69
67Kotval,“PublicPolicy,” 111.
68 Federal Bureau o f Investigation, CODIS Statistical Clickable Map, available from

http:www.fci.gov/hq/lab/codis/clIckmap.htm, accessed February 2003.
69 Both Orchid Cellmark and Geneiex advertise their accreditation with the FBI’s ASCLD-
LAB and their ability to perform analyses consistent with FBI and Combined DNA Index System
standards.
144
Rather than just alleviating the problem of procedural inequalities, the continued
production of autoradiographs within these laboratories solidifies the FBI’s position
of authority, extends their power and reach and severely limits potential opposition.
145
CHAPTER FIVE
THE BODY IN THE DIGITAL ARCHIVE: REDEFINING CRIME AND
POTENTIAL CRIMINALITY
In Chapter Three 1 argued that the specific collaboration of photography and
fingerprint Identification at the turn of the nineteenth and twentieth centuries
produced a new mode of law enforcement defined by a new role of the camera. The
ability to Identify criminals and track crimes independent of the physical presence of
the body and free o f traditional time constraints announced by fingerprinting
depended on the camera’s ability to capture crime scene data. This transformation in
law enforcement practices resulted in the construction o f a new archive of
identification information. In addition to housing records on known, captured and
incarcerated criminals this new archive housed latent identification data. Within this
new practice the camera collected aggregates of identification data on known and
unknown bodies according to its potential use.
Although fingerprint identification and photography announced the potential for
law enforcement to collect, store and use latent identification data, this practice is
only now being realized with the advent of digital technologies. The ease of
collecting, storing and sharing images and information made possible through
scanners, computers and digital cameras at the close of the twentieth century has
greatly extended the reach, capabilities and power o f law enforcement. Images of
146
faces, fingerprints and DNA are now routinely collected and compiled into vast
information databases, largely free of physical constraints and readily expandable.
The continued development of this digital arcMve gives rise to a space similar to that
described by Allan Sekula in Ms essay, “The Body and the Archive.” 1 Within this
space the scientific, social, technological and visual interact in the production of both
criminal and normal bodies.
This chapter investigates the contemporary digital arcMve, distinguishing it from
its nineteenth and twentieth century ancestors. My concern is not the larger, more
abstract inclusive archive described by Sekula but the specific territorialized arcMves
of law enforcement agencies in the United States.2 I argue that the limitations posed
by the physicality o f the nineteenth-century arcMve are largely absent in the new
digital space. Subsequently, contemporary law enforcement agencies increasingly
employ digital technologies in the collection o f vast sums o f personal identification
data. The construction and use of the resultant databases are dictated and driven by
technological capability and prowess and pose serious threats of discrimination.
This chapter argues that practices of mass data collection and archiving in
contemporary law enforcement and criminal identification practices constitute a
redefinition of criminality and the body. Combined with mutable conceptions of
crime, criminality and deviance, digital technologies enable law enforcement
agencies to bring increasing numbers of bodies into the archive. Once archived, the
1 Allan Sekula, “The Body and the Archive,” October 39 (1986): 3-64.
147
faces, fingerprints and DNA of these bodies are identified and re-identified according
to the parameters and needs of the specific investigation. The body in the digital
arcMve exists as a site of potential criminality.
Physicality, Inclusivity and Utility
The archives of law enforcement agencies of the nineteenth and twentieth
centuries functioned as spaces for record keeping and for tracking, arresting and
incarcerating recidivists. True to Sekula’s description, it was the filing cabinet and
not the photograph that defined this archival mode. For Sekula the arcMve provided
“a relation of general equivalence between images.”3 This equivalence was
guaranteed through “the metrical accuracy of the camera:” the camera translated the
disparate events of the live world into a system of comparable, static, two-
dimensional representations.4 Similarly, I argue that the digital arcMve provides for
“a relation of general equivalence between images;” however, rather than being
grounded in the metrical accuracy of the camera the digital archive is grounded in
electronic, binary code. The disparate and diverse representations and inscriptions in
contemporary law enforcement and criminal identification practices, the faces,
fingerprints and autoradiographs, are reduced to binary code and are made
exchangeable, searchable and comparable with one another.
148
The emergence of fingerprint identification at the turn of the nineteenth and
twentieth centuries greatly extended the reach of law enforcement. Fingerprint
records contained identification data in a compact and concise format that was highly
amenable to classification and exchange. Similarly, the latency of the fingerprint and
the mobility of the camera meant that police could collect identification data
independent of the physical presence of the body and independent of its knowledge.
By the opening decades of the twentieth century and equipped with cameras, law
enforcement agencies throughout the Western world began to build archives of
identification data.
Despite this effort, the large fingerprint archives that developed throughout the
twentieth century remained limited by their physicality. Although considerably less
cumbersome to produce and to search than Bertillon’s identification cards, fingerprint
records still occupied distinct physical space and, more importantly, required
significant effort to produce, file and search. The practices of photographing
fingerprints necessarily reflected these physical limitations of law enforcement.
Although in theory police could photograph a seemingly infinite number of
fingerprints, in practice this would render the arcMve unusable. As the arcMve grew,
searches became increasingly difficult and laborious and positive matches
increasingly improbable.
The move towards digital technology in law enforcement initiated in the late
1960s and early 1970s offered a solution to the physical limitations of the arcMve.
Reduced to a binary code, visual and textual identification records occupied
149
significantly smaller space than their paper predecessors. Similarly, algorithmic
searches performed by computer processors achieved results and speeds far greater
than anything possible manually. The result was significant Whereas the collection
and archiving of identification data in the nineteenth and twentieth centuries had to
account for the physical limitations associated with the classification, storage and
retrieval o f paper records, the digital archive exists largely free of these previous
constraints.
I do not want to suggest that the digital archive is absent of all labour or without its
own particular constraints. Information databases in law enforcement still require the
constant participation of law enforcement personnel. And the rate at which computer
technologies have developed has made the upgrade o f equipment and programs a
constant concern o f law enforcement agencies. However, I stress that the physical
limitations restricting the collection o f identification data during the nineteenth and
most of the twentieth century are disappearing with the continued development and
use o f digital technologies. This transformation in the capabilities for data collection
highlights a primary feature of the digital archive, and one that clearly distinguishes it
from its predecessors: inclusivity. Michael Smith, a member of the National Institute
of Justice’s National Commission on the Future of DNA Evidence, explains the
notion o f inclusivity in relation to information databases. He comments, “the more
inclusive a searchable identification database is, the more useful it is for the excluding
150
or including of individuals based on evidence found at a crime scene,” He adds “the
more inclusive, the more useful.”5
This is the paradox of the digital archive: whereas the utility of the paper archives
of the nineteenth and twentieth centuries waned in relation to their growth, the utility
of the digital arcMve increases with each addition to the database. A completely
inclusive digital arcMve, containing identification data on every member of society,
would ensure that no crime committed would go unsolved and no criminal
unidentified. The construction of such an archive raises immediate questions about
individual rights and freedoms and privacy. These concerns notwithstanding, I stress
that the construction of the digital archive threatens to redefine criminality and the
body. Constant data collection and ubiquitous police surveillance, in the form o f an
inclusive identification database, positions the body as something that is potentially
criminal. Criminality exists in every record o f the digital archive and in every body
those records represent It is something that is latent and, therefore, potentially
identifiable.
Whereas fingerprint identification and photography announced the potential for
law enforcement to collect and use masses of identification data, the adoption of
digital technologies a century later has made it a reality. The filing cabinet has been
replaced by the computer, making the M l extent o f the archive available to law
5 National Institute o f Justice, National Commission on the Future of DNA Evidence,

Proceedings, Legal Issues Working Group Report and Discussion (September 27,1999). Accessed
May 21,2003, available from http:///www.ojp.usdoj.gov/nij/dnamtgtrans7/welcome.html. Also cited
in Rebecca Sasser Peterson, “DNA Databases: When Fear Goes Too Far,” American Criminal Law
Review 37 (2000): 1219-1238.
151
enforcement. To exemplify this I turn to an examination of two digital archives
fundamental to contemporary and future law enforcement, the National Crime
Information. Center 2000 and the Combined D NA Index System. I use these
databases to argue that contemporary law enforcement practices reflect a trend toward
the accumulation and archiving of identification data.
NCIC 2000 and CODIS
The National Crime Information Centex was initiated in 1967 to facilitate
communication between law enforcement agencies in the United States. The program
originally housed data on wanted persons and on stolen vehicles, articles, guns and
license plates. A missing persons database was added in 1975 and an unidentified
persons database was added in 1982.6 Data is entered into the National Crime
Information Center by local law enforcement agencies. These agencies use
standardized data fields and entry codes to promote the uniformity of the information
entered. Data fields include various forms of identifying information: license plate
numbers, serial numbers or manufacturing numbers in the case of objects and
variables such as height, weight, age and place of birth in the case of people.7 A
unique identification code given to each participating agency is also entered to
indicate the source of the data and to protect against unauthorized access. Entered
6 William D. Haglund, “The National Crime Information Center (NCIC) Missing and
Unidentified Persons System Revisited,” Journal o f Forensic Sciences 38, no. 2 (! 993): 365-378.
7 The Federation o f American Scientists offers a detailed description o f the various data fields
as well as a general overview o f the National Crime Information Center. Accessed May 21,2003,
available from, http://www.fas.org/Irp/agency/doj/fbi/is/ncic.htm.
152
data is managed through a designated state agency, called a Control Terminal
Agency, or CTA, and forwarded into the centralized database of the FBI, Law
enforcement agencies from local, state and federal levels share and access
information over a secure telecommunications network.
Use of the National Crime Information Center is restricted to local, state, federal
and other law enforcement agencies, such as the Royal Canadian Mounted Police.
Agencies participating In the program must adhere to security safeguards
administered and monitored by the FBI.8 Searches proceed according to the fields
used for data entry. Searches cross-reference all databases so that a search for an
individual’s name will cross-reference the missing and wanted persons files as well as
other relevant databases.9 Potential matches are ranked through a scoring system
based on the number o f data fields matching the search parameters.10 For example,
an officer at a local agency In Florida can search the system using the name, height,
weight and age o f a suspect If the suspect was pulled over in relation to a traffic
violation, data on the vehicle such as its make, model, serial number and license plate
could also be entered. The system returns records matching the search parameters
enabling the officer in Florida to see if the person apprehended is wanted in relation
to other crimes throughout the country or if the vehicle has been reported lost ox
* See source In note 7.
9Paul L. Woodward, “Criminal Justice "Hot" Files,” (Washington, D.C.: U.S. Department of
Justice, Bureau o f Justice Statistics, 1986), 19.
10 William D. Haglund, “The National Crime Information Center (NCIC) Missing and
Unidentified Persons System Revisited,” Journal o f Forensic Sciences 38, no. 2 (1993): 365-378.
153
stolen. The officer can then use the agency identification code to contact the various
agencies responsible for entering the original data and can access the entirety of
records on the individual and vehicle in question.
During the 1990s the FBI received $183,000,000 in funding to upgrade and
expand the system. The new program, entitled National Crime Information Center
2000, houses seventeen cross-referenceable databases. The separate databases are:
stolen articles, foreign fugitives, stolen guns, criminal history queries, stolen license
plates, deported felons, missing persons, criminal justice agency identifier, stolen
securities, stolen boats, gang and terrorist members, unidentified persons, U.S. Secret
Service Protective File, stolen vehicles, persons subject to protection orders, wanted
persons, and the Canadian Police Information Center.1*
The upgrade o f the National Crime Information Center coincided with a larger
project o f the FBI to reevaluate and restructure their Identification Division. In 1989,
the FBI, in cooperation with the National Crime Information Center Advisory Policy
Board, conducted a study on the Division.12 As noted in Chapter Three, the
Identification Division was founded in 1924 to serve as a national site for the storage
and exchange of identification information. A central feature driving the
establishment of the Division was the need to develop a national fingerprint archive,
11Federal Bureau of Investigation, “National Crime Information Center (NCIC) 2000,” (Press
Release, July 15,1999). Accessed May 21,2003, available from
http://www.fbi.gov/pressrel/pressrel99/ncic2000.htm.
12 Bruce J. Brotman and Rhonda K. Pavel, “Identification: A Move Toward the Future,” Law
Enforcement Bulletin, July (1991).
154
thereby uniting the disparate fingerprint archives throughout the country. Similarly,
the 1989 study of the Identification Division was in response to the need to create
better storage and exchange of criminal justice information. The adoption of digital
technologies by law enforcement agencies across the United States gave rise to a
disparate array of digital archives, represented in the numerous Automated
Fingerprint Identification System databases discussed in Chapter Three.
The result of the 1989 study was the complete reconfiguration and relocation of
the Identification Division. Renamed the Criminal Justice information Services
Division in 1992, the new division constructed a new 500,000 square foot complex on
nine hundred and eighty-six acres of land in Clarksburg, West Virginia.13 The
Criminal Justice Information Services Division remains the largest unit of the FBI.
In addition to housing the seventeen databases o f the National Crime Information
Center 2000, the Division also houses the National Incident-Based Report System and
the Integrated Automated Fingerprint Identification System, a program uniting the
various AFIS databases in law enforcement agencies throughout the country.14
The central feature o f the National Crime Information Center upgrade was the
adoption of digital imaging technologies. The new program enables officers to
collect, store and share digital images and information from a variety of locations. To
13 Federal Bureau o f Investigation, Criminal Justice Information Services Division. Accessed

May 21,2003, available from http://www.fbi.gov/bq/cjisd/about.htm.
14 The Integrated Automated Fingerprint Identification System (IAFIS) became operable in

July, 1999. See 'Technology Update: Unsolved Case Fingerprint Matching," Law Enforcement
Bulletin, (December 2000): 12-13; “Inauguration o f the Integrated Automated Fingerprint
Identification System (IAFIS),5' Federal Bureau o f Investigation (Press Release, August 1 0 , 1999).
Accessed November 2002, available from http://www.fbi.gov/pressrel/pressrel99/iafis.htm.
155
take ftill advantage of the system, participating agencies are encouraged to purchase a
computer, laser printer, document scanner, single finger print scanner and a digital
camera. Agencies are responsible for hardware costs but the FBI provides software
and support free of charge. Using the digital camera, scanner or single fingerprint
scanner officers can enter digital images of faces, fingerprints, signatures and up to
ten other identifying images per record (the program suggests scars, tattoos and other
bodily marks). These images can be entered directly from the field and can be
relayed to other officers through Mobile Data Terminals, or M.D.T.’s, hardmounted
to police cruiser dashboards and through laptops.15 In addition, some departments
are exploring hand-held wireless devices such as Palm Pilots to link officers on foot
and bicycle to each other and to the information databases.16
Searches of the National Crime Information Center 2000 proceed in the same
manner as its predecessor. However, with the advent of digital imaging technologies,
officers are no longer restricted to textual descriptors but can enter and access image
files from a variety o f remote locations. The officer in Florida submitting the
information described above can now submit and receive mugshots, fingerprints and
15Keith J. Cutri, “Laptop Computers: New Technology for Law Enforcement,” Law
Enforcement Bulletin, February / March (1996); Catherine Greenmail, “A Well-Equipped Patrol
Officer: Gun, Flashlight, Computer,” (New York Times, 1999, accessed January 10 2003), available
from http://www.nytimes.com/library/tech/99/01/circuits/articles/21howw.html; Catherine Greenman,
“Computers on Patrol Have Got Your Number,” (New York Times, 1999, accessed January 10 2003);
available from http://www.nytimes.eom/library/tech/99/01 /circuits/artides/2! howw-side.html;
Christine Moran, “Computers Making Job of Police Easier,” (The Intelligencer, accessed January 10
2003); available from http://www.pMllyburbs.eom/intelligenceiTecord/artidel.asp7F_miiiFl427054.
16“Bike Cops Get Info on Palm - Bellevue Police Test 30 New Palm Pilots,” (Eastside
Journal Online, 2000, accessed January 10 2003), available from http://www.easteidejoumal.com.
156
other identifying images for comparison. Using the single fingerprint scanner, the
officer can scan the print of the individual in question for a search against the records
of the Integrated Automated Fingerprint Identification System.
The National Crime Information Center processed 2,000,000 transactions over the
entirety of Its Inaugural year. By the middle of the 1980s the program processed 400,
000 transactions per day.17 Statistics for the year 2000 show that the updated system
processes 2,500,000 transactions per day.18 On March 15, 2002 the National Crime
Information Center 2000 set a new record with 3,295,587 transactions In a single
day.19 Although only officially launched in 1999 the National Crime Information
Center 2000 now links 80,000 criminal justice agencies throughout the United States,
the U.S. Virgin Islands, Puerto Rico and Canada.20 As an example o f the depth and
size o f this developing archive, the centralized fingerprint database currently houses
over 35,000,000 fingerprint records with over 630,000,000 individual fingerprint
’'Woodward, “Criminal Justice,” 19.
18Stephanie L. Hitt, “NCIC 2000,” Law Enforcement Bulletin 69, no. 7 (2000): 12 - 15.
19Federal Bureau of Investigation, “An NCIC Milestone,” (Press Release, March 23,2002).
Accessed March 5 2003, available from http://www.fei.gov/pressrel/pressrel02/acic032302.htm.
20 Federal Bureau of Investigation, “Inauguration of the Integrated Automated Fingerprint

Identification System (IAFIS),” (Press Release, August 10, 1999). Accessed May 21,2003, available
from http://www.fbi.gov/pressrel/pressrel99/iafis.htm.
157
’j |
images. Fingerprint searches that took four to six months in the 1980s take one to
two hours in 2001.22
Contemporaneous with the update of the National Crime Information Center and
the transformation of the Identification Division the FBI began a pilot program called
the Combined DNA Index System (CODIS). Like the National Crime Information
Center 2000, the purpose of the DNA Index System is to facilitate the exchange of
identification data between law enforcement agencies. The program unites Local
DNA Index Systems (LDIS) and State DNA Index Systems (SDIS) with the National
DNA Index System (NDIS). As outlined in Chapter Four, participating laboratories
must meet guidelines for analysis established and monitored by the FBI.
Local agencies collect and forward genetic data through specified state agencies
and into the centralized National Index of the FBI laboratory. The specific
displacement of bands in the DNA profile is measured with the use of a computer and
the profile is stored as a numeric relation. Searches proceed through matching
comparable numeric codes. As with the National Crime Information Center 2000 a
match to the National DNA Index provides additional reference information including
21 Raymond Dussauit, “Cause and Effect,” (Government Technology, April 2000, accessed
January 9 2003), available from
http://www.govtech.net/publications/crimetech/AprOO/CTEfmgerprints.phtml.
22Michael D. Kirkpatrick, “Solving Cold Cases with Digital Fingerprints,” Sheriff'S3, no. 4
(2001): 14-17.
158
“ [a] laboratory Identifier, sample identifier, the associated DNA profile, and details
of the search parameters.”23
Compared to the National Crime Information Center 2000, the Combined DNA
Index System is relatively small in size and limited in its capabilities. This is largely
due to the type o f data being stored and the infancy of DNA analysis. DNA contains
a wealth o f personal information far beyond that of other modes o f Identification, and
therefore presents increased dangers associated with the misuse of data. Because of
this DNA databases are not currently interfaced with other systems, as the Integrated
Automated Fingerprint Identification System is with the National Crime Information
Center 2000. Further, the production o f DNA profiles, represented in the
autoradiograph, is a time and labor intensive process. The initial technique for
performing DNA analysis, called Restriction Fragment Length Polymorphism
(RFLP) analysis, took up to six weeks to produce a DNA profile. By the close of the
1990s a newer, faster method of analysis called Short Tandem Repeat (STR) analysis
had been developed, producing results within forty-eight hours.24 Although possible
in such a short time-frame, in actual practice most analyses take between four and
23Randall S. Murcfr and Bruce Budowle, “Are Developments in Forensic Applications of

DNA Technology Consistent with Privacy Protections”?, in Genetic Secrets: Protecting Privacy and
Confidentiality in the Genetic Era, ed. Mark A. Rothstein (New Haven: Yale University Press, 1997),
212-230; 223.
24Tod Newcombe, “Slow Spiral: State DNA Lag Databases,” {Government Technology, April
2000, accessed January 9 2003), available from
http://www.govtech.net/publications/crimetech/AprOO/CTENDA.phtoiI.
159
five weeks,25 In a January 15,2003 Press Release, Orchid Cellmark announced Its
new “DNA Express Service” which promised DNA analysis results to law
enforcement agencies In five days.26
Despits these technological innovations, the production of DNA profiles remains
considerably slow in comparison to fingerprinting. Combined with the ease with
which DNA samples can be collected, law enforcement agencies remain plagued by
backlogs of unanalyzed samples. Year 2000 estimates suggest there are roughly 1.5
million backlogged samples throughout agencies in the United States and that
$75,000,000 In fending would be needed to clear the backlog.27
In 1990, during the program’s trial phase, there were fourteen laboratories
participating in the Combined DNA Index System. In 1994, the DNA Identification
Act provided the FBI with $40,000,000 to expand its program.2® As discussed in
Chapter Four the majority o f this money was used to develop analysis facilities and
communications capabilities at the state and local level. Currently there are over one
hundred and thirty participating laboratories throughout the United States and Puerto
Rico. The FBI’s National Index houses over 1,200,000 DNA profiles from convicted
offenders and over 40,000 unidentified or latent profiles from crime scenes.
25Orchid Biosciences, “Orchid Cellmark Launches Rapid Forensic DNA Analysis Service,”
(Press Release, January 15, 2003). Accessed May 20,2003, available from
http://www.orchid.com/news/view_pr.asp?ID=280.
26 Orchid Biosciences, “Orchid.”
27 Newcombe, “Slow Spiral.”
28 “DNA Identification Act o f 1993,” (Washington D.C.: United States House of

Representatives, 1993), 15. The Act was enacted into the 1994 Violent Crime and Law Enforcement
Act.
160
The adoption of digital technologies by law enforcement and their cooperation
with initiatives such as the National Crime Information Center 2000 and the
Combined DNA Index System are justified through rhetoric that describes these
systems as a necessary response to increasingly mobile criminals. Mobile Data
Terminals, laptops and wireless devices such as Palm Pilots connect officers on foot,
bicycle and in patrol cars to each other and to the digital archive.29 Law enforcement
journals and government reports and press releases are filled with success stories in
which an offender apprehended in one state is linked to crimes in another through
these technologies. Such success stories are instrumental in securing fending for law
enforcement to continue to develop and implement these technologies and to clear the
backlogged DNA samples.
I do not question that the capability to exchange identification data electronically
is useful for law enforcement agencies in apprehending serial offenders. However, I
stress that digital technologies have also greatly extended the reach and capabilities of
law enforcement agencies to collect and archive masses of identification data. “Live
scan” systems such as that illustrated in Figure 5.1 enable officers to capture and
archive electronic images of an individual’s fingerprints, palm prints and face. When
the suspect is not in custody and therefore the “live scan” system cannot be used,
digital cameras enable officers to collect vast amounts of identification data.
Fingerprints can be entered directly into the digital archive from the field, or they can
29 See notes 11 and 12.
161
be “lifted” and brought back to the police station to be scanned and archived using
additional equipment (Figure 5.2). With funding received from the National Institute
of Justice, Nanogen of San Diego has developed a portable, hand-held DNA
scanner.30 This device fits in the palm of the hand and allows the user to produce a
DNA profile directly from the crime scene (Figure 5.3).
Technological innovations such as those illustrated above contribute to an
expanding digital arcMve. The increased capabilities of law enforcement in this
effort have met surprisingly little opposition. Legislation governing the permissible
collection and storage of identification data and the enforcement of that legislation is
either absent or insufficient in relation to the heightened capabilities of police.
The National Crime Information Center 2000 has received minimal critical
attention. This is primarily due to the type of information it contains. Mugshots,
fingerprints, and serial numbers of stolen vehicles, boats and guns are accepted as
relatively banal identification data. Which is to say that there is little concern that this
information could be used for anything other than identification. Further, I would
argue that a common belief regarding the collection and storage of data in the
National Crime Information Center 2000 is that it is restricted to known criminals and
therefore does not warrant the concern of the everyday citizen.
j0 Gunjan Sinha, “DNA Detectives,” Popular Science 255, no. 2 (1999): 48-52. More
information on the specific equipment can be found at Nanogen’s web-address, available at
http://www.nanogen.com, Accessed May 26,2003.
162
Figure 5.1: NEC’s LS21™ Live Scan Workstation. Law enforcement agencies can
capture, archive, search and exchange fingerprints, palm prints and mugshots from
this workstation. Photo courtesy of Sylvia Welch, Manager, Marketing and
Communications, NEC Solutions (America) Inc.
163
Figure 5.2: A latent fingerprint workstation by NEC. Allows law enforcement
personnel to scan latent prints for examination and archive and exchange prints within
an AFIS program. Photo courtesy of Sylvia Welch, Manager, Marketing and
Communications, NEC Solutions (America) Inc.
Figure 5.3: Nanogen’s NanoChip® portable DNA scanner. As advertised on
Nanogen’s website, available at http://www.nanogen.com.
164
Of the limited work addressing the National Crime Information Center, two 1993
studies showed the system to have significant flaws.31 The studies tested the accuracy
of the system and found it to be compromised in two ways: data was entered
incorrectly and entire fields of data were often omitted. The reports concluded the
system was subject to misidentifications which could in tom lead to false arrests.
Indeed, a 1985 New York Times article reported that the National Crime Information
Center was producing an average of 12,000 false database matches per day.32
Not only has data been entered incorrectly and produced false identifications but it
has also been subject to misuse. In 1993 the United States General Accounting
Office issued a report highlighting the misuse of information contained, in the
National Crime Information Center.33 The report was a response to a 1991 incident in
which twenty individuals with access to the National Crime Information Center sold
criminal history information obtained from the system to third parties. The report
also found several instances where records had been altered or completely deleted.
The upgrade of the National Crime Information Center during the 1990s attempted
to take account of the problems and potential dangers associated with the system that
were identified through these studies and reports. The upgrade Included extensive
35 Gary L. Bell, “Testing the National Crime Information Center Missing I Unidentified
Persons Computer Comparison Routine,” Journal o f Forensic Sciences 38, no. 1 (1993): 13-22;
William D. Haglund, “The National Crime Information Center (NCIC) Missing and Unidentified
Persons System Revisited,”Journal o f Forensic Sciences 38, no. 2 (1993): 365-378.
32David Burnham, “FBI Says 12,000 Faulty Reports on Suspects are Issued Each Day,” New
York Times, August 25 1985.
33Laurie E. Ekstrand, '‘National Crime Information Center: Legislation Needed to Deter

Misuse of Criminal Justice Information,” (Washington, D.C.: United States General Accounting
Office, 1993), 32.
165
data-entry checks, including point-of-entry checks for errors and data validation
procedures to assure the uniformity of information and that all mandatory data fields
had appropriate entries.34 To protect against unauthorized access to the information,
agencies now communicate over local Intranets or through Virtual Private Networks,
or VPNs. Continued tests of the National Crime Information Center, such as those of
1993, are needed to address procedural issues associated with the program. More
importantly, critical investigation and inquiry from external parties is needed to
address the larger social and cultural issues associated with the archiving of personal
identification information.
Importantly, regulation governing the collection of mugshots, fingerprints,
signatures and other identification data (excluding DNA which will be addressed
shortly) varies according to the nature o f the investigation. A hypothetical case
presented during my work with the Metropolitan Toronto Police serves to illustrate
the point A crime occurs at a restaurant and officers are called to the scene. All
property within the venue is subject to investigation. Officers lift and or photograph
fingerprints from every drinking glass in the establishment. This results in hundreds
of fingerprints, which are then archived into an Automated Fingerprint Identification
System database and linked to the crime through a reference number. The individual
committing the traffic violation in Florida, years after the fictional example, can be
linked to a past crime through their patronage of a Toronto restaurant.
Federal Bureau of Investigation, Criminal Justice Information Services Division, National

Crime Information Center. Accessed May 21, 2003, available from
http://www.fbi.gov/hq/cjisd/ncic.hta.
166
The hypothetical example seems improbable or unlikely; however, it clearly
illustrates the paradoxical nature of the digital archive. Digital technologies enable
law enforcement agencies to collect and store tremendous sums of identification data.
However, this information is often collected independent o f consideration of its
actual, practical use. This, I argue, is indicative of a fundamental disconnect in
contemporary law enforcement. Identification data is increasingly collected based
solely on its potential use. The dangers of this practice are two-fold. First, practices
of mass data collection transform the subject of law enforcement from the individual,
physical body to a larger, abstract social body creating a pervasive, ever-present mode
o f police surveillance. Second, identification data collected and stored in the digital
archive can be continually accessed and reinterpreted according to the changing needs
o f law enforcement and to mutable definitions o f crime and criminality.
In contrast to the National Crime Information Center 2000, the Combined DNA
Index System has received a significant amount of critical attention. More
specifically, the construction of the National DNA Index has caused concern from a
variety o f legal, scientific and social perspectives. The concern radiates from the fact
that DNA contains a wealth of personal information far beyond what is necessary for
identification purposes. Work within sociology, cultural studies and the social study
o f science points to the dangers of discrimination based on genetic information.05 For
35Three central texts (each is a collection o f essays) addressing the social and cultural issues
associated with DNA identification are: Carl F. Cranor, ed. Are Genes Us? The Social Consequences
o f the New Genetics (New Brunswick: Rutgers University Press, 1994); Ruth Hubbard and Elijah
Wald, Exploding the Gene Myth: How Genetic Information is Produced and Manipulated by
Scientists, Physicians, Employers, Insurance Companies, Educators, and Law Enforcers (Boston:
167
example, medical insurance companies and health care providers could set a different
fee schedule for persons exhibiting the BR.CA1 gene, which is linked to breast cancer.
Similarly, employers could access and use genetic information to discriminate against
individuals during the hiring process. In addition to these social and cultural
concerns, work from medicine, law and public policy has addressed the legal and
’y/r
policy issues associated with privacy and confidentiality. Of chief importance in
this debate is the protection of privacy, confidentiality and Fourth Amendment rights
(protection against unreasonable search and seizure), as well as the right of

'I 'l
individuals to maintain control over their own genetic information.
The DNA Identification Act addressed these concerns by placing tight restrictions
on the collection, storage and use of genetic information within law enforcement
practices. Proponents of the National DNA Index and the Combined DNA Index
System rightly note that, under the DNA Identification Act, only persons convicted of
certain sexual and violent crimes will have their DNA stored in the National Index.
Proponents also note that the database contains DNA profiles and not samples. As
Beacon Press, 1993); Daniel I. Kevles and Leroy E. Hood, The Code o f Codes: Scientific and Social
Issues in the Human Genome Project (Cambridge, Mass.: Harvard University Press, 1992).
•’s George I. Annas, “Privacy Rules for DNA Databanks: Protecting Coded ’Future Diaries’,”
Journal o f the American Medical Association 270, no. 19 (1993): 2346-2350; George J. Annas,
Leonard H. Glantz, and Patricia A. Roche, “Drafting the Genetic .Privacy Act: Science, Policy, and
Practical Considerations,” Journal o f Law, Medicine and Ethics 23 (1995): 360-366; Andrea De
Gorgey, “The Advent of DNA Databanks: Implications for Information Privacy,” American Journal o f
Law and Medicine 16, no. 3 (1990): 381-398.
37The Fourth Amendment reads: “The right of the people to be secure in their persons,
houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no
Warrants shall issue, but upon probably cause, supported by Oath or affirmation, and particularly
describing the place to be searched, and the persons and things to be seized.”
168
noted in Chapter Four, profiles represent specific non-coding fragments of DNA and
do not offer biological information about an individual. In this sense, it is argued, the
National DNA Index employs appropriate safeguards to protect against the misuse of
genetic identification data.
However, I stress that while the DNA Identification Act does safeguard practices
surrounding the inclusion o f DNA profiles into the National DNA Index it does not
regulate the collection and storage of DNA at the local and state level. Subsequently,
local and state agencies across the country routinely collect and store genetic data
from an increasingly diverse group o f offenses. Recall that DNA’s use in law
enforcement was developed to aid in the investigation of specific, serial sexual and
violent crimes. Initial laws governing the collection o f DNA restricted it to those
convicted o f such crimes. To exemplify, California enacted the first DNA
databanking law, restricting the collection o f DNA to adult registered sex offenders.
However, by 1994 its law was among the broadest in the nation. The California law
was expanded to allow the collection o f DNA from various non-sex related crimes
including “certain types o f assault and battery on custodial officers, transportation
personnel, or jurors; contributing to the delinquency of a minor, and inducing
disobedience to a court order.”38
California’s expanding law is indicative of a trend throughout the United States.
In a 1994 study o f state legislation governing DNA collection and databanking, Jean
j8 Jean E. McEwen and Phillip R. Reilly, “A Review o f State Legislation on DNA Forensic
Data Banking,” American Journal o f Human Genetics 54 (1994): 941-958; 945.
169
McEwen and Philip R. Reilly found tremendous discrepancy between state laws and a
continued trend in broadening laws to be more inclusive.j9 The authors found that
several states allowed for the collection of genetic material from an incredibly diverse
set of offences. These include “lewd and lascivious conduct,” “indecent exposure,”
“promoting or compelling prostitution,” “engaging in or abetting bigamy,” “loitering
near public toilets,” and “possession o f child pornography.” At least one state
allowed for samples to be taken from persons arrested but not convicted.40 A
subsequent study in 2000 by Rebecca Sasser Peterson supports McEwen and Reilly’s
findings 41 Peterson notes that Alabama, New Mexico, Virginia and Wyoming
require DNA samples from anyone convicted of a felony; Louisiana requires samples
from all individuals arrested on sex offenses, whether convicted or not; and Arizona,
Kansas and Oregon require samples from juveniles arrested in relation to sex
crimes.42 This last case is particularly alarming as the data is collected and stored for
future use when the individual reaches the age of majority.
To reiterate, my concern is not with the exchange o f identification data between
law enforcement agencies, nor is it with the misuse of that information by external,
third parties, although these remain very significant issues. Instead, my concern rests
with the practice of data collection. The continued collection o f data in the digital
39 McEwen and Reilly, “A Review.”
40 McEwen and Reilly, “A Review.”
41 Rebecca Sasser Peterson, “DNA Databases: When Fear Goes Too Far,” American Criminal
Law Review 37 (2000): 1219-1238. See also: D.H. Kaye, “Bioethical Objections to DNA Databases
for Law Enforcement: Questions and Answers,” Selon Hall Law Review 31, no. 4 (2001): 936-948.
170
archive, based on its potential use in future law enforcement practices constructs the
body as a site of potential criminality, the specific notion of crime being determined
by those controlling the production and use of the archive. In Chapters Three and
Four I described fingerprints and DNA as immutable mobiles. Importantly, as
inscriptions, fingerprints and autoradiographs attest only to the presence of a specific
body at a specific time and location. Once archived these inscriptions remain
immutable references to innumerable bodies, throughout innumerable contexts, times
and locations. The individual body’s identity as normal or deviant changes with the
use o f the archive.
DNA analysis was introduced to law enforcement for the express purpose of
assisting in the investigation o f serial sexual and violent crimes. It was useful in these
cases because traditional forms of evidence such as the eye-witness were absent and
biological materials were often all that remained. Because DNA contains a wealth of
personal information beyond that necessary for the identification process, initial laws
placed tight restrictions on its collection and use. However, these laws quickly
expanded to include a diverse number of offenses. As a result, law enforcement
agencies have amassed 1,500,000 backlogged samples of DNA. As with practices of
data collection in the National Crime Information Center 2000, this attests to a
fundamental disconnect in contemporary law enforcement between data collection
and its use. Under expanding laws, DNA is collected at a rate far beyond that which
42Sasser Peterson, “DNA Databases,”1225.
171
it can be analyzed and used. What then are the purpose and rationale behind this data
collection? And what is at stake in this process?
A clear effect of this practice of mass data collection is to bring increasing
numbers of bodies into the archive, making it more inclusive. More bodies in the
archive means more database hits, more crimes solved, more success stories and
continued funding. However, I stress that the accumulation of DNA samples and
other identification data in this effort betrays the rationale for using these
technologies in law enforcement and poses clear threats of discrimination.
Archived DNA profiles o f persons convicted of indecent exposure pose no clear
use for police. Serial flashers are not apprehended through a DNA database search.
Importantly, once archived, DNA can be re-accessed and re-interpreted. While it is
true that the National DNA Index stores only profiles and not samples, local and state
databases routinely store both the profile and the sample from which it was derived.43
Profiles entered into local, state and federal databases are assigned a reference
number based on the agency responsible for entering the data and the specific case.
These reference numbers can be used to access the sample from which the profile was
produced. The potential danger in this practice is that the information will be
accessed and used based on changing definitions of crime and criminality. Genetic
research on behaviour, sexuality, race, gender or any other variable can be used to
establish new criteria for searching and using the information in the digital archive.
43 One o f the most common reasons given for storing samples is that they may need to be used
during a criminal investigation, such as to produce farther autoradiographs or to re-test the DNA in the
face of evidentiary dispute.
172
To exemplify, studies appearing in a 1996 issue of the journal Nature Genetics
suggested a link between a specific gene (the D4 Dopamine receptor gene) and the
behavioral trait of novelty seeking.44 Law enforcement agencies can use this
information to search the archive for individuals having this specific gene. The
individual brought into the archive on a charge of indecent exposure could be re
arrested and incarcerated based on Ms now presumed genetic predisposition for risky
behavior. The search for genetic links for human traits and behavior such as that of
the Dopamine receptor and the search for the “homosexual” gene presupposes a
simplistic cause and effect link between gene and trait.45 Such an approach
erroneously treats the gene as what Evelyn Fox Keller describes as a “master
molecule” and ignores the complex relationships between genes in producing human
traits as well as the myriad environmental factors involved.46
Just as the laws governing DNA collection are expanding so too can the laws
governing its use. The DNA Identification Act permits the collection and use of
DNA for “law enforcement purposes.” As Barry Scheck warns, tMs term is
“disturbingly broad,” and could permit the use o f genetic information and exchange
between a variety of institutions including immigration, child support and welfare
44 R. P. Ebstein et al., “Dopamine D4 Receptor (D4R4) Exon III Polymorphism Associated

with the Human Personality Trait of Novelty Seeking,” Nature Genetics 12 (1996): 78-80; I. Benjamin
et a!., “Population and Familial Association Between the D4 Dopamine Receptor Gene and Measure of
Novelty Seeking,” Nature Genetics 12 (1996) 81-84.
45Two research articles that propose a genetic link for male homosexuality are: Stella Hu et
a!., “Linkage Between Sexual Orientation and Chromosome Xq28 in Males but not in Females,”
Nature Genetics 11, November (1995): 248-256; Dean H. Hamer et ai, “A Linkage Between DNA
Markers on the X Chromosome and Male Sexual Orientation,” Science 261, July 16 (1993): 321-327.
46Evelyn Fox Keller, “Master Molecules,” in Cranor, Are Genes Us?
173
agencies.47 In the same way that research on the D4 Dopamine receptor gene can
lead to a new notion of criminality and yield new parameters for searching the
archives of law enforcement agencies, so too can this information be extended
through the social programs and apparatuses described by Scheck. Children carrying
the gene could be identified as pre-criminal, placing them under constant surveillance
and subjected them to continued discrimination. Similarly, non-United States citizens
carrying the gene and trying to enter the country could be denied access, and those
already residing in the United States could be deported.
Conclusion
To further exemplify the problematic nature o f the digital archive and to conclude
I want to discuss briefly the current effort to collect and use identification data under
the rubric o f a “war on terror.” Within this war United States Attorney General John
Ashcroft’s National Security Entry-Exit Registration System digitally captures the
fingerprints and faces o f visitors to the United States who originate from countries
deemed a national threat. Law enforcement personnel at border crossings, in airports
and other ports of entry use “live scan” systems such as that depicted in Figure 5.1 to
collect this identification data. The Registration System has already produced “hits”
(fingerprint matches) to fingerprints in the databases o f the National Crime
47 Barry Scheck, “DNA Data Banking: A Cautionary Tale,” American Journal of Human
Genetics 54, no. 6 (1994): 931-933; 932.
174
Information Center 2O0O,48 Individuals identified through these hits are acted upon
according to a specific, contemporary conception of criminality. This conception
propagates a connection between race, gender, nationality and the propensity to
commit criminal acts, in this case defined as terrorist acts, and It carries severe and
maximum penalties. An individual bom in one of the countries deemed a national
threat that is brought into the archive in relation to a petty crime in 1990 can be re
identified in 2003 as a potential threat to national security. This re-identification is
made based solely on the individual’s place of birth. Similarly, a visitor to the United
States brought into the archive in 2003 under the Registration System can be re-
identified as criminal in 2020 according to mutable definitions of what constitutes
crime and criminality. In each case, the individual’s presence in the digital archive as
an image o f his face, fingerprint or DNA remains immutable but his identity, rights
and freedoms as an innocent citizen, petty thief, terrorist, national threat or otherwise
changes with the nature o f the investigation.
The National Security Entry-Exit Registration System illustrates the problematic
nature o f the digital archive. Practices of mass data collection such as that of the
Registration System and their cooperation with programs such as the National Crime
Information Center 2000 and the Combined DNA Index System are made possible
through digital technologies. Discourses driven by notions of fear and national threat
mobilize these technologies and legalize practices o f data collection that are overtly
48 Statistics show the System produces an average o f 67 hits a week and, in a preliminary test
between January and May, 2002, led to the apprehension o f 1400 people. See: National Security
Entry-Exit Registration System (United States Department of Justice, 2002, accessed January 13 2003),
available from http://www.usdoj.gov/ag/speeches/2002/natlsecentryexittrackingsys.htm.
175
discriminatory and that are degrading and humiliating to those subjected to collection,
The significant and far-reaching social ramifications of these practices are ignored in
favour of technological capability and the continued expansion of the archive.
It is expected that, by the year 2005, Ashcroft’s Registration System will record
each of the country’s 35,000,000 annual visitors.49 The mandatory collection of
images of faces and fingerprints in this effort is defended as a necessary response to
the events of September 11,2001 and to a terrorist threat promoted as invisible and
ever-present I question the purpose and rationale for collecting and storing
identification data on such a vast scale. Long after the “war on terror” has come to
completion, the Registration System will continue to collect data for its archive.
What is the purpose of this process of collection? Does the collection o f data fulfill a
clear need o f law enforcement or is it motivated by paranoia and fear? What are the
social, political and cultural consequences of practices o f data collection that are
clearly based in racial and religious discrimination?
The enhanced capabilities of law enforcement and the unchecked, exponential
expansion of the digital archive demand a reinvestigation of the rationale and ethical
issues associated with the collection, storage and use of identification data. Digital
technologies do not merely expand the breadth of the archive and increase the speed
with which it can be searched. Rather, they announce a restructuring of law
enforcement and criminal identification practices. In the nineteenth-century archive
49 Eric Schmitt, “Ashcroft Proposes Rules for Foreign Visitors,” (New York Times, 2002,
accessed June 10 2002), available from http://www.nytimes.com/2002/06/06/politics/06VlSA.html.
176
the image captured the criminal, representing Ms identity within its frame. Within the
new digital space the image captures the body independent of identity. The body in
the digital arcMve exists as data, it is an object of ever-present surveillance and is
always a site of potential criminality.
177
CONCLUSION
The development of photomechanical reproduction in the nineteenth century
coincided with the growth of social institutions that sought ways of recording and
documenting the body. The camera was quickly employed by these institutions,
resulting in the development o f vast photographic archives. John Tagg, Allan Sekula,
David Green, Suren Lalvani and other photographic historians have examined the
specific use of the camera in the documentation of the criminal body.1 I argue that, in
its strict treatment o f the photograph as representation, this work fails to address the
concrete practices involved in the production and use o f photographs within law
enforcement and criminal identification.
To complement existing work in the history of photography and to address the
concrete practices associated with the production and use of photographic
representation, the dissertation drew from work in the field of science and technology
studies. I argue that the notion of inscription and the concept of the immutable
mobile, as outlined by Bruno Latour and Steve Woolgar, are particularly effective in
addressing the use of photographic representation in law enforcement and criminal
Identification practices. Positioning the photograph as an inscription, I argue that the
production, archiving and exchange of photographic representations in law
enforcement and criminal identification function in the construction o f identity.
1 See the sources listed in footnote 3 o f Chapter One.
178
I stress that the history of criminal identification practices from the nineteenth
century to the present can be read through the history of the production and use of
inscriptions of the body. From the development of modem policing in the nineteenth
century, law enforcement agencies have been dependent on inscriptions to reduce the
complexities and idiosyncrasies of live, mobile bodies into two-dimensional
representations that are mobile, immutable, presentable, readable and combinable
with one another.2
From its adoption into law enforcement practices in the mid-nineteenth century,
photographic representation has served as a primary means o f translating the body
into a two-dimensional document. The role of the photograph in law enforcement
and criminal identification was transformed from strict representation to inscription as
it was mobilized through the classificatory system o f Alphonse Bertiilon. However,
Bertillonage was time and labour-intensive and the recording o f data on Bertiilon
identification cards was inconsistent. Fingerprinting supplanted Bertillonage in large
part because it yielded better inscriptions o f the body. This was essential in law
enforcement practices that were increasingly reliant on the communication and
exchange of information. Fingerprints provided the necessary identification data in a
compact and concise visual format and, as tactile traces of the body, they were
2 These are the five features of Latour’s conception of the immutable mobile. See Bruno
Latour, "Drawing Things Together," in Representation in Scientific Practice, ed. Michael Lynch and
Steve Woolgar (Cambridge, Mass.: MIT Press, 1990), 19-68.
179
accepted as 44the natural inference of presence” of the criminal.3 Similarly, as the
inscription of an individual’s unique, genetic code, the autoradiograph functions as a
powerful method of identification. The specific displacement of bands in the visual
representation of DNA is promoted by law enforcement agencies as an image of
identification to rival, and eventually replace, its dermal-based predecessor.
Despite its specific manifestation as an image of a face, fingerprint or DNA, visual
representation has remained a central tool in law enforcement and criminal
identification practices. I argue that the importance of visual representation can be
traced to the development of a new conception of the criminal body in the nineteenth
century. Work in physiognomy and phrenology of the eighteenth and nineteenth
centuries defined the outward surface of the body, specifically the head and face, as
indicative of an individual’s character. Nineteenth-century work in the then emerging
field of criminology expanded on the tenets of this work and defined the body as a
text that offered visual signs of its identity as normal or deviant Championed in the
work of Cesare Lombroso, criminality was defined as something that could be read
directly from the physical body and therefore captured in its visual representation.
Whereas the photograph was initially employed by law enforcement agencies as a
tool of representation, I argue that the specific collaboration of photography and the
theory of fingerprint identification at the close of the nineteenth century announced a
new practice in law enforcement, defined by a new role of the camera. The mobility,
3 George Wilton Wilton, Fingerprints: History, Law and Romance (London: William Hodge
and Company, 1938).
180
ubiquity and objectivity of the camera, combined wife the latency of the fingerprint,
enabled law enforcement agencies to collect and archive latent identification data.
TMs, I argue, transformed the subject of law enforcement from fee individual,
criminal body to a larger, aggregate social body within which any specific body was
potentially criminal.
In my analysis of fingerprint identification I traced the concrete production of
fingerprint evidence to stress the centrality of photographic representation in that
process. In contrast to this, my analysis of DNA identification examined the
development of authority over DNA analysis in law enforcement to stress the
importance o f the control of inscriptions. I argue that the FBI gained significant
power through the gradual development o f control over the inscription of DNA, the
autoradiograph. The DNA Identification Act o f 1993 was central in this effort. The
Act gave the FBI control over the standards for DNA analysis, delimited their
authority over the technique to its use within law enforcement practices, and provided
the funding necessary for the Bureau to develop a National DNA Index.
The theoretical possibilities associated with the collection and use o f latent
identification data announced at the turn of the nineteenth and twentieth centuries are
a reality in 2003. The defining feature of Sekula’s nineteenth-century archive, the
filing cabinet, has been replaced by the computer. Digital technologies have greatly
extended the capabilities of police to collect aggregates of identification data
independent o f the physical presence and knowledge of the body. Images of faces,
fingerprints and DNA are quickly and easily collected by law enforcement personnel
18!
and are archived in the databases of the National Crime Information Center 2000 and
the Combined DNA Index System.
The transformation of the FBI’s Identification Division to the Criminal Justice
Information Services Division and its relocation to a 500,000 square foot complex
attests to the primary importance placed on identification data in contemporary law
enforcement and criminal identification practices. The amalgamation, in January
2003, of over twenty government agencies including the United States Coast Guard,
Secret Service, Immigration and Naturalization Service, and the Department of
Energy into the unified Department of Homeland Security points to a similar drive to
centralize the collection and dissemination of criminal justice information.4 These
initiatives are indicative o f the emphasis in modem and contemporary law
enforcement practices to collect, archive and exchange identification data.
As I have stressed throughout this dissertation, while the exchange of
identification data between law enforcement personnel and agencies is undoubtedly
an effective tool in apprehending offenders, the practices of data collection are cause
for concern. In conjunction with increased technological capabilities for producing
inscriptions of the body, changing conceptions of crime and criminality contribute to
the collection of identification data toward a complete and inclusive digital archive.
These practices o f mass data collection reflect a transformation in the role of law
enforcement in society and threaten a redefinition of the body. The body in the
4 The Department of Homeland Security was brought into operation on January 2 4 , 2003. A
complete history of the Department, including its schematic organization is available from
http://www.whitehouse.gov/homeland/index.htmL Accessed May 26,2003.
182
digital archive is always a site of potential criminality. Its identity as normal or
deviant, and its attendant rights and limitations are determined through the production
and interpretation of images in the digital archive.
183
WORKS CITED
Frye v. United States, 293 F. 1013 (D.C. Cir. 1923).
People v. Castro, 545 N.Y.S. 2d 985 (Sup. Ct. 1989).
State v. Andrews, 533, So.2d 841 (Fla. App. 5 Dist. 1988).
United States v. Yee, ND, OMo 129 FRD 692 (1990).
"Finger-Prints via Radio." Scientific American 127, December (1922): 386.
Genetic Witness: Forensic Uses o f DNA Tests. Washington, D.C.: United States
Congress, Office of Technology Assessment, 1990.
DNA Technology in Forensic Science. Washington, D.C.: National Research Council,

1992.
DNA Identification Act o f 1993. Washington D.C.: United States House of

Representatives, 1993, 103-45.
"Correspondence: Forensic DNA Typing Dispute." Nature 372 (December 1 1994):

398-399.
Violent Crime Control and Law Enforcement Act o f 1994. Washington, D.C.: United
States Senate and House o f Representatives, 1994, P.L. 103-322.
"Correspondence: More on DNA Typing Dispute." Nature 373 (January 12 1995):

98-99.
DNA Identification Grants Improvement Act o f 1995. Washington, D.C.: United

States House of Representatives, 1995, 104-393.
The Evaluation o f Forensic DNA Evidence. Washington, D.C.: National Research

Council, 1996.
“A Capsule Chronological History of Fingerprint Identification.” Metropolitan

Toronto Police Identification Bureau. August 1996.
184
“Fingerprints as a Means of Identification.” Metropolitan Toronto Police

Identification Bureau. August 1996.
“History of Fingerprints.” Metropolitan Toronto Police Identification Bureau. August

1996.
“History o f Identification by Fingerprints.” Metropolitan Toronto Police

Identification Bureau. August 1996.
Scenes o f Crime Manual. Ontario Police College. August 1996.
“Structure of the Skin” Metropolitan Toronto Police Identification Bureau. August

1996.
“National Crime Information Center (NCIC) 2000.” Federal Bureau of Investigation

(Press Release, July 15, 1999). Accessed May 21, 2003, available from
“Inauguration of the Integrated Automated Fingerprint Identification System

(IAFIS).” Federal Bureau of Investigation (Press Release, August 10,1999).
Accessed November 2002, available from,
http://www.fbi.gov/pressrel/pressrel99/iafis.litm.
“Bike Cops Get Info on Palm - Bellevue Police Test 30 New Palm Pilots.” Eastside
Journal Online, 2000. Accessed January 10 2003. Available from
http://www.eastsidejoumal.com.
"Technology Update: Unsolved Case Fingerprint Matching." Law Enforcement

Bulletin. December (2000): 12-13.
National Security Entry-Exit Registration System. United States Department of

Justice, 2002. Accessed January 13 2003. Available from
http://www.usdoj.gov/ag/speeches/2002/natisecentryexittrackingsys.htm.
“An NCIC Milestone.” Federal Bureau of Investigation (Press Release March 23,
2002). Accessed March 5 2003. Available from
“Orchid Cellmark Launches Rapid Forensic DNA Analysis Service.” Orchid

Biosciences (Press Release 2003). Accessed May 20 2003. Available from
http://www.orchid.com/news/view_pr.asp?ID=280.
Allison, Harrison C. Personal Identification. Boston: Holbrook Press, 1973.
185
Amann, XL, and K. Knorr Cetina. "The Fixation of (Visual) Evidence." In

Representation in Scientific Practice, ed. Michael Lynch and Steve Woolgar,
85-122. Cambridge, Mass.: MIT Press, 1990.
Annas, George I. "Privacy Rules for DNA Databanks: Protecting Coded 'Future
Diaries’." Journal o f the American Medical Association 270, no. 19 (1993):
2346-2350.
Annas, George I., Leonard H. Glantz, and Patricia A. Roche. "Drafting the Genetic
Privacy Act: Science, Policy, and Practical Considerations." Journal o f Law,
Medicine and Ethics 23 (1995): 360-366.
Barak-Glantz, Israel L., and C. Ronald Huff, ed. The Mad, the Bad, and the Different:
Essays in Honor o f Simon Dinitz. Lexington: Lexington Books, 1981.
Beavan, Colin. Fingerprints: The Origins o f Crime Detection and the Murder Case
that Launched Forensic Science. New York: Hyperion, 2001.
Beime, Piers, ed. The Origins and Growth o f Criminology: Essays on Intellectual
History, 1760-1945. Edited by Gerald Mars and David Nelken, International
Library o f Criminology, Criminal Justice and Penology. Aldershot:
Dartmouth, 1994.
Bell, Gary L. "Testing the National Crime Information Center Missing / Unidentified
Persons Computer Comparison Routine." Journal o f Forensic Sciences 38,
no.l (1993): 13-22.
Bendavid, Naftali. "Bill Would Expand Use of DNA Testing." Legal Times, June 7
1993.
Benjamin, J., Lin Li, C. Patterson, B.D. Greenberg, D.L. Murphy, and D.H. Hamer.
"Population and Familial Association Between the D4 Dopamine Receptor
Gene and Measure of Novelty Seeking." Nature Genetics 12 (1996): 81-84.
Berry, Donald A. "Statistical Issues in DNA Identification." In DNA on Trial:

Genetic Identification and Criminal Justice, ed. Paul R. Billings, 91-108.
Plainview: Cold Spring Harbor Laboratory, 1992.
Bertiilon, Alphonse. La Photographie Judiciaire avec un Appendice sur la

Classification et L ’identification Anthropometriques. Paris: Gauthier-Villars,
1890.
________ . Identification Anthropometrique: Instructions Signaletiques -Album.

Melon: Imprimerie Administrative, 1893.
186
. Alphonse Bertiilon’s Instructions for taking descriptions for the

identification o f criminals, and others by the means o f anthropometric
indications. New York: AMS Press, 1975.
Billings, Pan! R., ed. DNA on Trial: Genetic Identification and Criminal Justice.
Plainview: Cold Springs Harbor Laboratory, 1992.
________. ’’Gene Technology: Views of its Criminal Justice Applications." In DNA

on Trial: Genetic Identification and Criminal Justice, ed. Paul R. Billings,
1-4. Plainview: Cold Springs Harbor Laboratory, 1992.
Bishop, Katherine. "A Victory for Genetic Fingerprinting." New York Times,
November 16 1990, B6.
Bloor, David. Knowledge and Social Imagery. London: Routledge, 1976.
Brotman, Brace J., and Rhonda K. Pavel. "Identification: A Move Toward the
Future." Law Enforcement Bulletin, July (1991).
Burke, Terry, Gaudenz Dolf, Alec J. Jeffreys, and Roger Wolff, ed. DNA
Fingerprinting: Approaches and Applications. Basel, Switzerland: Birkauser
Verlag, 1991.
Burnham, David. "FBI Says 12,000 Faulty Reports on Suspects are Issued Each
Day." New York Times, August 25 1985.
Byrnes, Thomas. Professional Criminals o f America. New York: Cassell and

Company, 1886.
Carey, Thomas, Assistant Special Agent in Charge, Washington Field Office, Federal
Bureau of Investigation. “Communication with the Law Enforcement
Community.” Congressional Statement, November 13, 2001. Accessed
September 2002, available from
http:/Avww.fbi.gov/congress/congressO 1/carey 111301.htm.
Chapel, Charles Edward. Fingerprinting: A Manual o f Identification. New York:

Coward McCann, 1941.
Chen-ill, Frederick R. The Finger Print System at Scotland Yard. London: Her
Majesty's Stationary Office, 1954.
Cole, Simon. Suspect Identities: A History o f Criminal Identification and

Fingerprinting. Cambridge: Harvard University Press, 2001.
187
________. "Manufacturing Identity: A History of Criminal Identification Techniques

From Photography Through Fingerprinting.” Ph.D. diss, Cornell University,
1998.
. "Witnessing Identification: Latent Fingerprinting Evidence and Expert

Knowledge." Social Studies o f Science 28, no. 5-6 (1998): 687-712.
Collins, H, M. Changing Order: Replication and Induction in Scientific Practice.

London; Beverly Hills: Sage Publications, 1985.
Cooper, H.S. "The Evolution of Canadian Police." In The Police Function in Canada,
ed. William T. McGrath and Michael P. Mitchell, 37-52. Toronto: Methuen,
1981.
Coopersmith, Jonathan. "Facsimile's False Starts." IEEE Spectrum, February (1993):

46-49.
Costigan, Daniel M. Electronic Delivery o f Documents and Graphics. New York:

Van Nostrand Reinhold Company, 1978.
Cranor, Carl F., ed. Are Genes Us? The Social Consequences o f the New Genetics.
New Brunswick: Rutgers University Press, 1994.
Curran, Thomas. Forensic DNA Analysis: Technology and Application.

Parliamentary Research Branch, Government o f Canada, 1997, BP-443E.
Cutri, Keith J. "Laptop Computers: New Technology for Law Enforcement." Law
Enforcement Bulletin, February / March (1996).
Daemmrich, Arthur. "The Evidence Does Not Speak for Itself: Expert Witnesses and
the Organization of DNA-Typing Companies." Social Studies o f Science 28,
no. 5-6 (1998): 741-772.
Daston, Lorraine, and Peter Galison. "The Image o f Objectivity.” Representations,

no. 40 (1992): 81 - 128.
Debenham, P.G. "DNA Fingerprinting; a Biotechnology Business." In DNA

Fingerprinting: Approaches and Applications, ed. Terry Burke, Gaudenz
Dolf, Alec J. Jeffreys, and Roger Wolff, 342-348. Basel, Switzerland:
Birkauser Verlag, 1991.
188
Diiworth, Donald €., ed. The Blue and the Brass, American Policing: 1390-1910.
Police History Series. Gaithersburg: International Association of Chiefs of
Police, 1976.
________. Identification Wanted: Development o f the American Criminal

Identification System 1893 - 1943. Police History Series. Gaithersburg:
International Association of Chiefs of Police, 1977.
Duncan. Chief Inspector J.H. An Introduction to Fingerprints. London: Butterworth

and Co., 1942.
Dussauit, Raymond. “Cause and Effect.” Government Technology, April 2000.

Accessed January 9 2003. Available from http://wwrv.govtech.net/
publicatioEs/crimetech/AprOO/CTEfingerprints.phtmL
Ebstein, R.P., O. Novlck, R. Umansky, B. Priel, Y. Osfaer, D. Blaine, E.R. Bennett, L.

Nemanov, M. Katz, and R.H. Belmaker. "Dopamine D4 Receptor (D4R4)
Exon III Polymorphism Associated with the Human Personality Trait of
Novelty Seeking." Nature Genetics 12 (1996): 78-80.
Ekstrand, Laurie E. National Crime Information Center: Legislation Needed to Deter

Misuse o f Criminal Justice Information. Washington, D.C.: United States
General Accounting Office, 1993, GAO/T-GGD-93-41.
Ellis, Havelock. The Criminal. Vol. 7. 22 vols. Contemporary Science Series, ed.
Havelock Ellis. London: Walter Scott, 1890.
Faulds, Henry. "On the Skin-Furrows o f the Hand." Nature 22, Oct. 28 (1880): 605.
Foucault, Michel. Discipline and Punish: the Birth o f the Prison. Trans. Alan
Sheridan New York: Vintage Books, 1979.
Galison, Peter. "Judgment Against Objectivity." In Picturing Science Producing Art,

ed. Caroline A. Jones and Peter Galison, 327 - 359. New York: Routledge,
1998.
Galton, Francis. "Personal Identification and Description: 1." Nature 38, June 21
(1888): 173-77.
________. "Personal Identification and Description: 2." Nature 38, June 28 (1888):
201- 02 .
________ . Finger Prints. London: Macmillan, 1892.
189
________. Inquiries Into Human Faculty and its Development. London: J.M. Dent
and Sons, 1907.
________ . Memories o f My Life. London: Methuen, 1908.
Gilman, Sander L. Difference and Pathology: Stereotypes o f Sexuality, Race, and

Madness. Ithaca: Cornell University Press, 1985.
Gilman, Sander L., Hugh Welch Diamond, and John Conolly. The Face o f Madness:
Hugh W Diamond and the Origin o f Psychiatric Photography. New York:
Brunner/Mazel, 1976.
Gorgey, Andrea De. "The Advent of DNA Databanks: Implications for Information
Privacy.” American Journal o f Law and Medicine 16, no. 3 (1990): 381-398.
Green, David. "Classified Subjects." Ten.8 , no. 14 (1984): 30-37.
________ . "On Foucault: Disciplinary Power and Photography." Camerawork 32

(1985): 6-9.
________ . "Veins of Resemblance: Photography and Eugenics." The Oxford Art

Journal 7, no. 2 (1985): 3-16.
Greenmail, Catherine. “Computers on Patrol Have Got Your Number.” New York
Times, 1999. Accessed January 10 2003. Available from
http://www.nytimes.eom/IIbrary/tech/99/01 /circuits/articles/
21 howw-side.html.
. “A Well-Equipped Patrol Officer: Gun, Flashlight, Computer.”

New York Times, 1999. Accessed January 10 2003. Available from
http://www.nytimes.eom/library/tech/99/01/circuits/articles/21howw.html.
Gunning, Tom. "Tracing the Individual Body: Photography, Detectives, and Early
Cinema.” In Cinema and the Invention o f Modern Life, ed. Leo Chamey and
Vanessa R. Schwartz, 15-45. Berkeley: University of California, 1995.
Haglund, William D. "The National Crime Information Center (NCIC) Missing and
Unidentified Persons System Revisited." Journal o f Forensic Sciences 38,
no. 2 (1993): 365-378.
Halfon, Saul. "Collecting, Testing and Convincing: Forensic DNA Experts in the
Courts." Social Studies o f Science 28, no. 5-6 (1998).
190
Hamer, Dean H., Stella Hu, Victoria L. Magnuson, Nan Hu, and Angela M. L.
Pattatucci. "A Linkage Between DNA Markers on the X Chromosome and
Male Sexual Orientation.'' Science 261, July 16 (1993): 321-327.
Hamilton, Peter, and Roger Hargreaves. The Beautiful and the Damned: The Creation
o f Identity in Nineteenth Century Photography. London: Lund Humphries,
2001.
Henry, Sir E.R. Classification and Uses o f Finger Prints. 7th [1934] ed. London: His
Majesty's Stationary Office, 1900.
Herschel, W. J. "Skin Furrows of the Hand." Nature 23, Nov 25 (1880): 76.
Hitt, Stephanie L. "NCIC 2000." Law Enforcement Bulletin 69, no. 7 (2000): 12 - 15.
Holt, James. Finger Prints Simplified: A Handbook o f the Science o f Finger Print
Identification. Chicago: Frederick I. Drake and Co., 1920.
Horn, David G. "This Norm Which is Not One: Reading the Female Body in
Lombroso's Anthropology." In Deviant Bodies: Critical Perspectives on
Difference in Science and Popular Culture, ed. Jennifer Terry and Jacqueline
Urla, 109-128. Bloomington: Indiana University Press, 1995.
Hu, Stella, Angela M. L. Pattatucci, Chavis Patterson, Lin Li, David W. Fulker,
Stacey S. Cherny, Leonid Kraglyak, and Dean H. Hamer. "Linkage Between
Sexual Orientation and Chromosome Xq28 in Males but not in Females."
Nature Genetics 11, November (1995): 248-256.
Hubbard, Ruth, and Elijah Wald. Exploding the Gene Myth: How Genetic
Information is Produced and Manipulated by Scientists, Physicians,
Employers, Insurance Companies, Educators, and Law Enforcers. Boston:
Beacon Press, 1993.
lies, Chrissie, and Russell Roberts, ed. In Visible Light: Photography and
Classification in Art, Science and The Everyday. Oxford: Museum of Modem
Art, 1997.
Inman, Keith, and Norah Rudln. An Introduction to Forensic DNA Analysis. New
York: CRC Press, 1997.
Jasanoff, Sheila. "The Eye o f Everyman: Witnessing DNA In the Simpson Trial."
Social Studies o f Science 28, no. 5-6 (1998): 713-740.
191
Jones, Caroline A., and Peter Galison, ed. Picturing Science Producing Art. New
York: Routledge, 1998.
Kaye, D.H. "Bioethical Objections to DNA Databases for Law Enforcement:

Questions and Answers." Seton Hall Law Review 31, no. 4 (2001): 936-948.
Keller, Evelyn Fox. "Master Molecules." In Are Genes Us? The Social
Consequences o f the New Genetics, ed. Car! F. Cranor, 89-98. New Jersey:
Rutgers University Press, 1994.
Kelly, William and Nora. Policing in Canada. Toronto: Macmillan, 1976.
Kevles, Daniel J., and Leroy E. Hood. The Code o f Codes: Scientific and Social
Issues in the Human Genome Project. Cambridge, Mass.: Harvard University
Press, 1992.
Kirkpatrick, Michael D. "Solving Cold Cases with Digital Fingerprints." Sheriff 53,
no. 4 (2001): 14-17.
Kolata, Gina. "Some Scientists Doubt the Value of 'Genetic Fingerprint' Evidence."
New York Times, January 29 1990, A l.
Kotval, Jeroo S. "Public Policy for Forensic DNA Analysis: The Model o f New York
State." In DNA on Trial: Genetic Identification and Criminal Justice, ed. Paul
R. Billings, 109-118. Plainview: Cold Spring Harbor Laboratory Press, 1992.
Kuhn, Thomas S. The Structure o f Scientific Revolutions. [2d, eni ed. Chicago]:
University o f Chicago Press, 1970.
Kufane, Frederick. The Finger Print Instructor. New York: Muirn and Company,
1916.
Lalvani, Suren. Photography, Vision, and the Production o f Modern Bodies.

INTERRUPTIONS: Border Testimony(ies) and Critical Discourse/s, ed.
Henry A. Giroux. Albany: State University of New York, 1996.
Lander, Eric. "DNA Fingerprinting: Science, Law, and the Ultimate Identifier." In
The Code o f Codes: Scientific and Social Issues in the Human Genome
Project, ed. Daniel J. Kevles and Leroy Hood. Cambridge: Harvard University
Press, 1992.
Lander, Eric S., and Bruce Budowle. "DNA Fingerprinting Dispute Laid to Rest."
Nature 371, October 27 (1994): 735-738.
192
Latour, Bruno. Science in Action: How to Follow Scientists and Engineers Through
Society. Cambridge, Mass.: Harvard University Press, 1987.
. "Drawing Things Together." In Representation in Scientific Practice, ed.

Michael Lynch and Steve Wooigar, 19-68, Cambridge, Mass.: MIT Press,
1990.
Latour, Bruno, and Steve Wooigar. Laboratory Life: The Construction o f Scientific
Facts. 2nd ed. Princeton: Princeton University Press, 1979.
Lavater, Johan Caspar. Essays on Physiognomy, Designed to Promote the Knowledge

and the Love o f Mankind. London: T. Bensley, 1810.
Lescarboura, Austin C. "Sending Photographs Over Wires." Scientific American 123,

no. 19 (1920): 474-84.
Lombroso, Cesare. Crime: Its Causes and Remedies. Translated by Henry P. Horton,
M.A. Patterson Smith Reprint Series in Criminology, Law Enforcement, and
Social Problems. Montclair: Patterson Smith, 1968 (c. 1911).
Lombroso, Cesare, and William Ferrero. The Female Offender. The Criminology
Series, ed. M.A. W. Douglas Morrison. New York: D. Appleton and
Company, 1895.
Lombroso-Ferrero, Gina. Criminal Man: According to the Classification o f Cesare

Lombroso Patterson Smith Reprint Series in Criminology, Law Enforcement,
and Social Problems. Montclair: Patterson Smith, 1972 (c. 1911).
Lynch, Michael, and Sheila Jasanoff. "Contested Identities: Science, Law and
Forensic Practice." Social Studies o f Science 28, no. 5-6 (1998): 675-686.
Maynard, Patrick. The Engine o f Visualization: Thinking Through Photography.

Ithaca, N.Y.: Cornell University Press, 1997.
McEwen, Jean E. "Forensic DNA Data Banking by State Crime Laboratories."

American Journal o f Human Genetics 56 (1995): 1487-1492.
________ . "DNA Data Banks." In Genetic Secrets: Protecting Privacy and

Confidentiality in the Genetic Era, ed. Mark A. Rothstein, 231-251. New
Haven: Yale University Press, 1997.
McEwen, Jean E., and Phillip R. Reilly. "A Review of State Legislation on DNA
Forensic Data Banking." American Journal o f Human Genetics 54 (1994):
941-958.
193
McFadden, Robert D. "Reliability of DNA Testing Challenged by Judge's Ruling."

New York Times, August 15 1989, Bl.
McGrath, William T., and Michael P. Mitchell, ed. The Police Function in Canada.
Toronto: Methuen, 1981.
Moran, Christine. “Computers Making Job of Police Easier.” The Intelligencer,

Accessed January 10 2003.. Available from http://www.phillyburbs.com/
inteliigencerrecord/article 1.asp?F_nutn=l427054.
Murefa, Randall S., and Brace Budowle. "Are Developments in Forensic Applications
of DNA Technology Consistent with Privacy Protections?" In Genetic
Secrets: Protecting Privacy and Confidentiality in the Genetic Era, ed. Mark
A. Rothstein, 212-230. New Haven: Yale University Press, 1997.
Neufeld, Peter J. "DNA - A Defense Perspective." A paper delivered at the

Proceedings for the International Symposium on Human Identification, 1989.
Newcombe, Tod. “Slow Spiral: State DNA Lag Databases.” Government Technology,
April 2000. Accessed January 9 2003. Available from
http://www.govtech.net/pubiications/crimetech/AprOO/CTENDA.phtmi.
Peterson, Rebecca Sasser. "DNA Databases: When Fear Goes Too Far." American
Criminal Law Review 37 (2000): 1219-1238.
Phillips, Sandra, Mark Haworth-Booth, and Carol Squires. Police Pictures: The
Photograph as Evidence. San Francisco: Chronicle Books, 1997.
Phillips, Sandra S. "Identifying the Criminal." In Police Pictures: The Photograph as

Evidence, ed. Sandra S. Phillips, Mark Haworth-Booth, and Carol Squires,
11- 33. San Francisco: Chronicle Books, 1997.
Porter, Theodore M. The Rise o f Statistical Thinking: 1820 -1900. Princeton:

Princeton University Press, 1986.
________. Trust in Numbers: The Pursuit o f Objectivity in Science and Public Life.
Princeton: Princeton University Press, 1995.
Rabinow, Paul. "Gallon's Regret: O f Types and Individuals." In DNA on Trial:

Genetic Identification and Criminal Justice, ed. Paul R. Billings, 5-18.
Plainview: Cold Spring Harbor Laboratory, 1992.
194
Rafter, Nicole Hahn. "Criminal Anthropology in the United States." Criminology 30,
no. 4 (1992): 525-545.
Ray, leffer C. Criminology: An Interdisciplinary Approach. Englewood Cliffs:

Prentice-Hall, 1990.
Reed, Bill. "Automated Fingerprint Identification: From Will West to Minnesota

Nine-Fingers and Beyond." Journal ofPolice Science and Administration 9,
no. 3 (1981): 317-326.
Rejlander, O. G. "Hints Concerning the Photographing of Criminals." The British

Journal Photographic Almanac (1872): 116-117.
Rhodes, Henry T. F. Alphonse Bertillon, Father o f Scientific Detection. London:

Harrap, 1956.
Rothstein, Mark A., ed. Genetic Secrets: Protecting Privacy and Confidentiality in
the Genetic Era. New Haven: Yale University Press, 1997.
Scheck, Barry. "DNA Data Banking: A Cautionary Tale." American Journal o f

Human Genetics 54, no. 6 (1994): 931-933.
Schmeck, Harold M. Ir. "DNA Findings Disputed by Scientists." New York Times,
February 7 1988,1.
Schmitt, Eric. “Ashcroft Proposes Rules for Foreign Visitors.” New York Times,
2002. Accessed June 10 2002.. Available from http://www.n3dimes.com/
20O2/O6/Q6/politics/O6VI5A.fatmi.
Sekula, Allan. "The Body and the Archive." October 39 (1986): 3-64.
Sinha, Gunjan. "DNA Detectives." Popular Science 255, no. 2 (1999): 48-52.
Tagg, John. The Burden o f Representation: Essays on Photographies and Histories.

Communications and Culture, ed. Stuart Hall and Paul Walton. Basingstoke:
Macmillan, 1988.
Talbot, C.K., C.H.S. Jayewardene, and T.J. Juliani, ed. Canada’s Constables: The
Historical Development o f Policing in Canada. Ottawa: Crimcare, 1985.
Terry, Jennifer, and Jacqueline Urla. Deviant Bodies: Critical Perspectives on

Difference in Science and Popular Culture. Race, Gender, and Science, ed.
Anne Fausto-Sterling. Bloomington: Indiana University Press, 1995.
195
Tuttle, Harris B. "History of Photography in Law Enforcement.’’ Finger Print and

Identification Magazine 43, no. 4 (1961): 3-28.
Ubell, Earl. "Whodunit? Quick Check the Genes!" Parade Magazine, March 31
1991.
Wentworth, Bert, and Harris Hawthorne Wilder. Personal Identification: Methods fo r

the Identification o f Individuals, Living or Dead. Chicago: T.G. Cooke, 1932.
Wilson, Thomas F. "Automated Fingerprint Identification Systems." Law

Enforcement Technology, August / September (1986): 17-46.
Wilton, George Wilton. Fingerprints: History, Law and Romance, London: William
Hodge and Company, 1938.
Woodbury, W. B. "Photographing Criminals.” The British Journal Photographic

Almanac (1875): 128.
Woodward, Paul L. Criminal Justice "Hot” Files. Washington, D.C.: U.S.

Department of Justice, Bureau of Justice Statistics, 1986, NCJ-101850.

Police Photo

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Police Photo

Transféré par

Droits d'auteur :

Formats disponibles

Police Photography : Visual Representation, Scientific Evidence

and the Construction of the Criminal

Submitted in Partial Fulfillment

Requirements for the Degree

Professor Lisa Cartwright

Program in Visual and Cultural Studies

Department of Art and Art History

Arts and Sciences

Rochester, New York

ProQuest Information and Learning Company

of study in the Art History Department of Leeds University, England. Jonathan

the University of Rochester in 1997. Jonathan pursued his research in forensic

of Arts degree from the University of Rochester in 2001.

This dissertation, “Police Photography: Visual Representation, Scientific Evidence

and the Construction of the Criminal,” examines the use of photographic

representation and photographic technologies in modem and contemporary American

and Canadian law enforcement and criminal identification practices. I combine

science and technology studies to position the photograph as an inscription, as a

material representation at work in the constmction of scientific evidence. I argue that

the production, archiving and exchange of photographic representations in law

enforcement and criminal identification practices function in the constmction of

The dissertation addresses three identification practices fundamental to modem

and contemporary law enforcement: fingerprinting, DNA analysis, and information

databases. I stress that each is dependent on photographic representation. I argue that

the specific collaboration of photography and the theory of fingerprint identification

representation of DNA, the autoradiograph.

The dissertation concludes with an examination of two digital archives used in

century digital archive is constructed as a site of potential criminality.

Chapter One Introduction 1

Chapter Two Picturing the Criminal: Photography and 19

Criminality in the Nineteenth Century

Chapter Three Photographing Fingerprints: Data, 60

Evidence and Latent Identification

Chapter Four DNA Analysis: Competition, 102

Standardization and the Control of

Chapter Five The Body in the Digital Archive: 145

Redefining Crime and Potential

Works Cited 183

Figure Title Page

Photographic representations of the criminal: a 1

mugshot, fingerprint and autoradiograph.

2.1 The Inspector’s Model. 21

2.2 Two examples of the visibly deviant body used as 34

illustrations in Lombroso’s work. “Italian Criminal. A

Case of Alcoholism,” (L) and “Anton Otto Krauser,

2.3 Tools used by Lombroso for measuring the “signs” of 37

criminality. The algometer (L) for measuring “general

sensibility and sensibility to pain,” and the campimeter

(R) for measuring field of vision.

2.4 “Physiognomy of Russian Female Offenders.” 40

2.5 Galton composite photographs showing the “criminal 44

2.6 Galton composite photographs showing the “Jewish 44

2.7 A page of images depicting male criminals from 50

Thomas Byrnes’ Professional Criminals o f America.

2.8 A page of images depicting female criminals from 51

Thomas Byrnes’ Professional Criminals o f America.

2.9 Measuring the width of the skull for anthropometric 54

2.10 Bertillon’s “synoptic table of facial expressions.” 55

2.11 A Bertillon identification card showing the recording of 56

anthropometric measurements, physical description and

full face and profile photographs.

3.1 The ridge characteristics of the fingerprint. 91

3.2 Fictional crime scene: the first photograph positions the 92

dinner plate within the context of the crime scene.