ANTHROPOLOGY TODAY VOL 25 NO 2, APRiL 2009 9
Imprisonment, having for its only object the detention of indi-
viduals, might become rare, when they were held as it were by
an invisible chain. Jeremy Bentham
Biometrics, from the Ancient Greek bios and metron,
denotes the recognizing of humans on the basis of intrinsic
physical or behavioural traits. Biometric systems differ
from more traditional forms of security by identifying the
body alongside knowledge or possession tokens, such as
passwords or documents. The common recognition charac-
teristics are the face, fingerprint or iris. However, ongoing
research is developing recognition of vascular pattern,
hand geometry, DNA and even body odour. Behavioural
biometrics encompasses handwriting, voice and keystroke
patterns, together with growing fields such as gait recogni-
tion. Since the late 1990s there has
been a revolution in biometrics for
primary security access and, in some
cases, as replacement technologies
for older forms of identification.
Most people experience bio-
metrics in an airport or restricted
building. Here, it is sufficient to
note that the process involves taking
an image of characteristics and
‘enrolling’ it in a database, com-
paring it to stored data and running a
performance system to test accuracy and filter noise, such
as motion or light. Biometric security thus involves an
information system with multiple levels of access, degrees
of openness and potential for inaccuracies. Unsurprisingly,
the technical literature is dominated by encryption, efforts
to model efficacy and establish ‘best practice’ – and with
good reason. In recent years, there have been a number of
well-publicised debacles. In 2002, for example, a security
industry conference in Las Vegas heard that researcher
Tsutomu Matsumoto gained access to an unfamiliar bio-
metric system in under 60 minutes using fake fingerprints
moulded from Gummi Bear sweets.
1
Other examples
include recent erroneous imprisonment cases in the US
due to ‘false positives’.
Civil liberties organizations have been quick to point to
the serious implications for privacy and the dangers of reli-
ance on single biometric technologies, but this has done
little to diminish investment in areas ranging from ‘smart’
borders and e-banking to benefit and welfare payments.
Indeed, according to recent estimates, by 2010 the market
for security solutions for European airports will reach 10
billon euros. And if we follow Naomi Klein, the future may
well be cultivated in more distant ‘laboratories’, such as
the Chinese port city and export zone of Shenzhen, where
experimental bio-IDs store data on
one’s reproductive history and even
landlords’ phone numbers.
For some commentators, the cur-
rent revolution in biometrics marks
a new era of surveillance in which
the body is both the target and the
instrument of control. Speaking of
‘smart’ borders, David Lyon (2007)
recently argued that citizenship
becomes less significant in a world
in which the body is the passport
to spaces and privileges. The insights of Michel Foucault,
particularly his work on panopticism, are central to sur-
veillance studies, and there is a growing literature on the
relationship between modern governmentality and what
Nikolas Rose has termed the ‘securitization of identity’
(1999: 240).
Biometric security has received little anthropological
attention despite historical associations with the discipline.
This article represents elements of a larger project on tech-
nologies of mobility control, and here I wish to stake out a
The birth of biometric security
Fig. 2. Dutch national IDs
require static ‘mug-shots’ so
official identification matches
face recognition scanners. In
early 2007, Robert Coleman,
at the time a security guard
of ‘Het Torentje’ (the round
tower which serves as the
office for the Dutch prime
minister in The Hague),
managed to procure a Dutch
national ID dressed as The
Joker from Batman (claiming
his attire was necessary
for religious reasons), used
it to fly to London and
was photographed outside
10 Downing Street. Thus
illustrating, at the very least,
the human error potential in
modern security systems.
Fig. 1. Three-dimensional
Face Recognition Shown at
a Biometrics Conference,
London, 2004.
Mark Maguire
Mark Maguire is a Lecturer
in the Department of
Anthropology, National
University of Ireland,
Maynooth. He was the 2008
Fulbright Advanced Research
Scholar in the Western
Institute for Irish Studies,
Stanford University, where he
is also a Visiting Associate
Professor in the Department
of Anthropology. His email is
Mark.H.Maguire@nuim.ie
iAN WALDie/GeTTY imAGes
De VOLkskRANT
10 ANTHROPOLOGY TODAY VOL 25 NO 2, APRiL 2009
position in relation to this topic. However, my perspective
will be informed by historical insights rather than ethno-
graphic ones. Essentially, I am asking the question: what
may be learned from a history of the present revolution in
biometric security?
Use of biometrics can be identified in a variety of his-
torical moments, from Assyrian payment receipts to inky
footprints on ancient Chinese divorce records. However,
biometric security, in the modern sense, was born in the
19th century, through the innovations of administrators,
anthropologists and French detectives and in contexts as
varied as colonial India, metropolitan London and fin-
de-siècle Paris. Unsurprisingly, much has already been
written about the emergence of police photography and
fingerprinting as chapters within the history of crimi-
nology and forensics. Other biometric solutions such as
anthropometry, once regarded as scientifically superior,
now appear as historical curiosities. But the history of bio-
metrics reveals more than simply periods of innovation,
testing, and varied levels of legal and public acceptance.
Fingerprinting and face recognition, which are at the heart
of the contemporary revolution in biometrics (and my focus
here), offered 19th-century innovators more than the pros-
pect of criminal identification: early biometrics promised a
utopia of bio-governmentality in which individual identity
verification was at the heart of population control.
In the brief historical survey below I will argue that fig-
ures such as the colonial administrator William Herschel,
the anthropologist and eugenicist Francis Galton and the
detective Alphonse Bertillon did not simply ‘discover’
technologies of human identification, classification and
data archiving; rather, each foresaw and spoke of the pos-
sibilities for systems of biometric security. Today, with
exponential growth in imaging and processing technolo-
gies and the increasing drive towards the securitization of
identity, the utopia of bio-governmentality evoked in the
late 19th century is finding new champions.
Fingerprints, photographs and filing cabinets
In 1880 the stirrings of a hundred-year-long controversy
were to be found in the pages of the journal Nature. Henry
Faulds, a Scottish medical missionary in Tokyo, published a
short letter outlining his observations on fingerprints, from
those of primates to human impressions on ancient pottery.
Faulds noted that these ‘nature-copies’ could be compared
visually, potentially leading to the ‘scientific identification
of criminals’ (1880: 605). Soon after William Herschel, a
former Assistant Joint Magistrate in India, wrote a follow-
up article in Nature pointing out that he had been working
with fingerprints for more than 20 years in a broader effort
to verify the identities of colonial subjects. The ‘discovery’
controversy continues to this day; however, a considerable
body of scholarship now argues that fingerprints may have
been used to authenticate identity in a variety of cultures
for many thousands of years (see Beavan 2002). Looking
beyond the development of fingerprinting for criminal
identification, the contribution of William Herschel is
most remarkable as an early attempt to roll out biometric
security in a variety of civil areas.
Herschel traced the genesis of modern fingerprinting
to his epiphanic encounter with ‘native’ signatures. In
1858 he was contracting road-building materials in India’s
Hooghly River region on behalf of the British administra-
tion (initially for the East India Company). He demanded
that a local man, Rajyadhar Kōnāi, sign a contract supple-
mented by his palm print with the intention of ‘frightening’
him. Local use of palm prints and fingertips, tip sahi, in
written agreements was not uncommon, but Herschel
was struck by the power of these nature copies to be both
reproducible and real. Colonial texts spoke of the obscu-
rity that arose because individual identity was unfixed in
the colonial gaze, and Herschel saw biometrics as the prac-
tical solution. Later, as Magistrate for Hooghly, he used
fingerprinting to combat fraud in pension claims, where
it was rumoured that one could hire an elderly ‘relative’;
Fig. 3. ‘Biometrics at War’.
US soldier scans and Iraqi
youth’s iris using a biometrics
digital system camera during
an enlistment program
session to join the Sahwa
group in Mahmudiyah, some
30 kms (18 miles) south of
Baghdad, on May 5, 2008.
mAURiCiO LimA/AFP/GeTTY
ANTHROPOLOGY TODAY VOL 25 NO 2, APRiL 2009 11
apparently prisoners were also available to serve one’s
sentence.
In Mimesis and alterity, Michael Taussig refers to the
work of Herschel as the ‘organized control of mimesis’,
a modernist sorcery through which the copy takes on the
attributes of the real, becoming even more real. The argu-
ment that Taussig makes is that ‘fingerprinting as a modern
state practice owes everything to modern colonialism’
(1993: 222). However, when one looks at the development
of fingerprinting one quickly comes to the conclusion that
it is not just a state practice or a specifically colonial one.
Chandak Sengoopta’s excellent history of fingerprinting
in India, Imprint of the Raj, speaks of colonialism as a
process simultaneously unfolding at different scales of
empire, involving state and non-state actors. One may
push this further by arguing that what is at stake here is the
application of biometrics for an emergent governmentality
in which, to borrow from Sengoopta, ‘one needed to know
individuals’ (2003: 47).
Despite his utopian efforts, when William Herschel
retired fingerprinting was still an ad hoc security solu-
tion lacking in scientific credibility. In 1880, prompted
by Faulds’s letter in Nature, he wrote of his experiences
and passed his notes to Darwin’s cousin, the Victorian
polymath, anthropologist and father of eugenics Francis
Galton.
2
Galton was at that time crossing the archipelago
of Victorian institutions – schools, the army, hospitals,
asylums and prisons – searching within the multitude of
individuals for a hidden order: heritable characteristics of
‘race’. In the 1870s he experimented with composite pho-
tography at the request of the Director General of Prisons.
As the British prison population swelled so too did an
unsystematic photographic archive, which failed to fulfil
its promise of literally capturing the habitual criminal.
Galton’s interest was not in the individual per se but rather
in identity as a marker of generic types. Working initially
with Herbert Spencer, Galton began to treat a facial image
as a landscape on which locational or ‘register marks’
could be made, thus rendering the human face as a series
of points that could then be mechanically selected (see
Galton 1879). However, his experiments failed to conclu-
sively locate generic types: the ghostly composite faces
were not apparitions of criminality or insanity but merely
physiological averages. Nonetheless, he understood the
implications of moving in the opposite statistical direction,
towards personal identification. The technical literature
on biometric face recognition still salutes his pioneering
work on mechanical selection for composites, in particular
systems using the Principal Component Analysis tech-
niques developed by Galton’s protégé Karl Pearson (see
Quintiliano and Rosa 2006, Galton 1889).
Galton’s interest in fingerprints, sparked by Herschel,
was similarly motivated by a desire to see beyond indi-
vidual differences towards heredity and ‘race’. However,
one of his greatest scientific contributions was indexing
and classification for individual identification. Throughout
1. Disturbingly, in 2007,
German Interior Minister
Wolfgang Schäuble’s
fingerprints were copied
from a drinking glass at
a university reception by
members of the Chaos
Computer Club and an
image published in a
magazine with instructions
on how it could be used to
deceive security systems.
2. The ‘discovery’
controversy also involved
Galton. Prior to the letter
in Nature, Henry Faulds
wrote to Charles Darwin
hoping for his assistance.
Darwin forwarded the letter
to Francis Galton, who
did not reply to Faulds,
but forwarded the letter to
the Royal Anthropological
Society where it received no
attention and was returned
to Galton in 1894. By
the early 1880s scientific
journals had moved on to
other matters, and Faulds
decided to write to police
chiefs around the world to
campaign for fingerprinting.
He received few replies,
as many police forces had
become enraptured by
anthropometry.
3. Biometrics such as
fingerprints are normally
used with respect to
populations composed of
enrolled individuals, e.g.
a criminal database. The
uses of biometrics comprise
authentication, which
involves matching within
acceptable error levels the
biometrics presented with
specific stored information,
and identification, where
the data is compared to all
data within the sample.
Since the Indian Evidence
Act of 1899, the world’s
first legal endorsement of
fingerprinting as a criterion
of identification, an image
of fingerprinting as the most
Fig. 5. ‘Mensuration de
la coudée’ From Alphonse
Bertillon’s exhibition at the
1893 World’s Columbian
Exposition in Chicago.
Permission required from the
National Gallery of Canada,
Ottawa.
xxx
Fig. 3. ‘The Jewish Type’
Example of composite
portraiture from Life of
Francis Galton by Karl
Pearson Vol 2 : image 0341.
xxx
12 ANTHROPOLOGY TODAY VOL 25 NO 2, APRiL 2009
the 1880s, while operating his celebrated Anthropometric
Laboratory at the International Health Exhibition in
Kensington, he pushed fingerprinting along the path to
scientific credibility. Though no trace of temperament or
‘race’ could be found in the arches, loops or whorls, Galton
was keenly aware of the significance of his contribution.
Habitual criminals in the vast unmapped spaces of met-
ropolitan London, immigrants and (in his words) native
‘coolies’ in the furthest reaches of empire could all be secu-
ritized. Later, he imagined a role for biometric security in
a eugenic utopia, and a fragment of that vision remains.
Towards the end of his life, the scientific flâneur dictated
a novel that was partially destroyed titled, The Eugenics
College of Kantsaywhere. In the manuscript, Professor
I. Donoghue arrives at a colony segregated according to
genetic fitness. Enforced celibacy and labour camps await
those of inferior stock, though upward mobility is pos-
sible through College examinations. The protagonist is
genetically screened in a manner apparently resembling
that of the Army, Indian Civil Service or a careful insur-
ance office. The final process is to leave one’s fingerprints
for future identification. The Professor notes, ‘All immi-
grants are more or less suspected’ (1906: no pagination).
‘I had never been so keenly looked over before,’ remarks
Donoghue, before explaining:
What they are concerned with in one another are the natural,
and therefore the only heritable characteristics. […] His ‘brute
value’ would be a proper expression if employed in the original
sense of that word. […] In Kantsaywhere they think much more
of the race than of the individual.
In the 1880s, when he was experimenting with finger-
printing and directing his anthropometric lab, Galton was
compelled to investigate parallel efforts in Paris, where
Alphonse Bertillon was fast achieving fame as a detec-
tive who thought much more of the individual than of
the ‘race’. If Galton epitomized the polymath scientist,
Bertillon – not unlike William Herschel – worked on the
practical and administrative side of research. Fears over
crime conditioned the birth of biometric security in fin-de-
siècle Paris. In the 1860s the city was ‘modernized’, with
only 40 percent of its buildings remaining untouched by
Baron Haussmann. The goal was to open streets out and
allow circulation; however, for many it was the city’s cir-
culation that was the cause of its problems. Paris was now
open to the ‘dangerous’ class of récidivistes (recidivists)
– people who were mobile in identity as well as move-
ment – and into this context stepped Bertillon, an obses-
sive police clerk.
Efforts had been made to identify recidivists through sil-
houettes and, later, as in Britain, crude and unsystematic
photographs accompanied by unverifiable names. By 1882
Bertillon had agreement to anthropometric measurement
of arrested persons. However, this could not confirm iden-
tity conclusively, so Bertillon added standardized photo-
graphs and the portrait parlé or verbal portrait. The details
were then entered on cards and meticulously filed. In the
years to follow, Bertillon’s legacy would be composed
of such visual innovations, though his key contribution
was, as Allan Sekula notes, ‘not the camera but the filing
cabinet’ (1986: 16). When one looks back at Bertillon’s
methods once can see a curious statistical mechanism: that
of recording the body’s markers, normally common to all,
in sufficient detail as to render them specific to one. The
exercise was essentially a working out of three problems.
Firstly, Bertillon noted the problem of human vision. He
argued that to see ‘tall’ or ‘short’ people is to perceive them
in a meaningful way. His goal was to standardize the police
gaze. Secondly, he wished to standardize the recording of
physical and social traits. Seeing was also writing, a coded
gaze. Thirdly, he archived the file on each subject by ‘indi-
vidualising him in the midst of the multitude of human
beings’ (1893: 4).
By 1885 Bertillonage was receiving worldwide atten-
tion and its inventor even eclipsed Sherlock Holmes as the
foremost detective in Europe in Sir Arthur Conan Doyle’s
story, The Hound of the Baskervilles. That year, his mind
on fingerprinting, Francis Galton visited Bertillon to dis-
cuss his methods and be entered into the system. Galton
regarded the whole business as less of a triumph in science
and more of a triumph in filing – perhaps correctly so. But
Bertillon was still riding the wave of success. His pub-
lisher’s introduction to the English edition of Signaletics
is effusive:
[E]very human being should be partially signalised. […] The
process of signalment would take the place of passports at
every frontier, and signalments would appear on all life assur-
ance policies, permits and other papers. […] It would then be
possible to find any person at once whenever desired, whether
for its own good or that of society at large, in whatever place
he might be and however he might alter his appearance or his
name. Crime would thus be rooted out, elections purified,
immigration laws effectively enforced, innumerable misunder-
standings and much injustice prevented and all business rela-
tions greatly facilitated. (1896: viii)
reliable of all biometrics has
emerged – an unquestioned
legal technology through which
the criminal’s prints may even
be, one-to-one, scientifically
‘matched’ with those left at
a crime scene. This image
is at variance with reality,
practically and theoretically.
In terms of enrolment, prints
may be partial or otherwise
distorted. There is also the
issue of scientifically verifiable
standards for ‘matching’.
Up until 2001 the UK used
a 16-point standard of
similarity, and many other
jurisdictions favour ridgeology.
The key issue is statistical
individualization, or the
process used to determine a
match to the exclusion of all
others. Confidence levels can
be attributed to either a positive
or negative result only with
reference to the population.
Beavan, C. 2002. Fingerprints:
The origins of crime
detection and the murder
case that launched forensic
science. London: Hyperion.
Bertillon, A. 1893.
Identification
anthropométrique:
Instructions signalétiques.
Paris: Imprimerie
Administrative.
— 1896. Signaletic instruction
including the theory and
practice of anthropometrical
identification. Chicago,
New York and London: The
Werner Company.
Cole, S. 2001. Suspect
identities: A history of
fingerprinting and criminal
identification. Cambridge,
MA: Harvard University
Press.
Fig. 6. An Unwilling Sitter in
Pinkertons Criminal Mug Shot
& Information Book, 1895’.
Allen Pinkerton, like Bertillon
in France, was a pioneer of
standardised, description and
photographic practice.
xxx
ANTHROPOLOGY TODAY VOL 25 NO 2, APRiL 2009 13
Conan Doyle, A. 1998. The
hound of the Baskervilles:
Another adventure of
Sherlock Holmes. Oxford:
Oxford University Press.
Faulds, H. 1880. On the
skin-furrows of the hand.
Nature 28 October: 605.
Foucault, M. 2004. Security,
territory, population:
Lectures at the Collège
de France, 1977-1978.
Basingstoke and New
York: Palgrave Macmillan.
Galton, F. 1879. Composite
portraits made by
combining those of many
different persons into a
single figure. Journal
of the Anthropological
Institute 8: 132-148.
— 1889. Personal
identification and
description. Journal of the
Anthropological Institute
18: 177-191.
— 1906. ‘The Eugenic
College of Kantsaywhere’.
Unpublished MS,
University College
London, FG 138/6.
Herschel, W.J. 1880. Skin
furrows of the hand.
Nature 25 November: 76.
Lyon, D. 2007. Surveillance
studies: An overview.
Cambridge: Polity Press.
Maghiros, I. et al. 2005.
Biometrics at the frontiers:
Assessing the impact
on society. European
Parliament Committee on
Bertillonage, based primarily on anthropometry, though
later incorporating fingerprints, worked with the visual,
textual and archival. By turning the body into code, it
had the potential to allow security data to travel. As it
spread from Europe to the Americas and along the routes
of empire, where fingerprinting was simultaneously on
the rise, a utopia shimmered on the horizon. If you were
trading under a certain name in Bengal in, say, 1898 but
were an impostor and swindler, the Bertillon system prom-
ised your undoing. You had, even in the 19th century, a
modern shadow – a so-called data double.
However, Bertillonage was flawed. By 1898 Bengal
kept over 200,000 Bertillon cards, but it demanded exact
use, and its human error potential increased out of sight of
its originator. In India, Edward Henry, who later helped
develop the Henry System of fingerprinting, pushed for
the replacement of the Bertillon system with his own.
Again, the history of biometrics is noteworthy: both
anthropometry and fingerprinting were primarily applied
in civil areas, from employment records to migration. This
resonates with historical analysis of the same era in South
America and the USA. For example, Ruggiero’s (2004)
research in Argentina documents a turn-of-the-century
period of testing and jockeying for position, leading to the
eventual primacy of the Vucetich fingerprinting system.
Biometric identification of criminals was historically
linked to issues of immigration in Argentina and there, as
in Paris, London and colonial India, a utopia of bio-gov-
ernmentality opened. Luis Reyna Almandos, a successor
of Vucetich, later unsuccessfully argued for fingerprinting
to underpin an international ‘book of personality’, linked
by personal numbers tattooed on the body: a universal
system of individual identity verification.
The history of the present
Much of the contemporary literature on biometrics is con-
cerned with the efficacy of the technologies, the implica-
tions for privacy and the relative openness of databases.
An understanding of the birth of biometric security has an
important role to play in framing the research challenges.
It is a history that swings pendulum-like from knowledge
of populations through racial theories to knowledge of the
individual units of those same populations; it is a history
marked by scientific innovation as well as administrative
and governmental deployments; however, perhaps most
intriguingly, it speaks of a vision of security – a utopian
excess – that pushed against the limits of 19th-century
technology and acceptability.
Considering the genealogy of biometric security draws
attention to security as an appropriate field of anthropolog-
ical enquiry. If we follow Michel Foucault, security must
be considered as a productive apparatus: it allows ‘circula-
tions to take place […] controlling them, sifting the good
from the bad, ensuring that things are always in movement
[…] but in such a way that the inherent dangers of this cir-
culation are cancelled out’ (2004: 65). This insight remains
underdeveloped in his work, but links individuals and
their mobility to bio-political power and governmentality.
Biometric security, in the 19th century as today, oscillates
between knowledge of and power over populations and the
securitization of individual identity.
For Francis Galton anthropometry, photographs and
fingerprints held the promise of finding in a multitude of
individuals, types whose heritable characteristics would
illuminate evolutionary and racial thought; his statistical
genius instead found individual security. But while overt
racial theory no longer has a place in the literature on bio-
metrics, so-called ‘categorical suspicion’ does: an indi-
vidual fingerprint or face is significant because it is a unit
within a population – a population enrolled in a database, a
population not enrolled or denied access, potentially a sus-
pect population. In this sense, biometrics are indexical.
A genealogy of biometrics also demands that we con-
sider security as an assemblage of different elements oper-
ating in specific contexts. The worlds of Francis Galton’s
anthropometric laboratory and Alphonse Bertillon’s police
archive were very different; however, they both responded
to contexts in which the mobility of bodies and the mal-
leability of individual identity was rendered problematic.
Both London and Paris were imperial metropolises, mag-
nets for large-scale immigration and landscapes darkened
by real and imagined criminals, marked off as different
and potentially set against the polis. While Galton’s mind
wandered from photography to fingerprinting and anthro-
pometry, indexing and classifying, the utopia he imagined
was one ordered by heredity and supported by a system of
biometric security. For Bertillon anthropometry was at the
core of a utopia always just beyond his reach. Signaletics,
contrary to his expectations, ultimately failed to slip its
moorings in criminal identification and move into ever
newer areas of civil life. As Simon Cole has convincingly
argued, while Bertillon’s anthropometric system was con-
sidered to be far more ‘scientific’ than its competitors well
into the 20th century, its demise was more of a victory
for ‘industrial-style speed, efficiency, productivity, and
economy over what was seen as scientific accuracy and
precision’ (2001: 93).
In practical terms, an understanding of the birth of bio-
metric security also helps temper images of the newness
of the contemporary revolution. While face recognition
systems developed into their modern form through techno-
logical advances made in the 1960s, the standardized gaze
of the police mug-shot, its augmentation of other security
apparatuses and pioneering work on mechanical selection
have a much older history. Fingerprinting, on the other
hand, quickly found a place in the craft of policing, where
technical experts succeeded in rendering it an objective
and ostensibly sure-fire method of criminal identification
emptied of socio-cultural meaning.
3
Fig. 7. Anthropometry card
of Francis Galton, with
profile and full-face photos
and spaces for key body
measurements, taken by
Alphonse Bertillon, 1893.
University College London,
FG, 84.
xxx
