The Matrix in Context: Taking Stock of Police Gang Databases in London and Beyond
| DOI | 10.1177/1473225419883706 |
| Published date | 01 April 2020 |
| Date | 01 April 2020 |
| Author | James A. Densley,David C. Pyrooz |
| Subject Matter | Special Issue Articles |
883706YJJ0010.1177/1473225419883706Youth JusticeDensley and Pyrooz
research-article2019
Special Issue Article
Youth Justice
2020, Vol. 20(1-2) 11 –30
The Matrix in Context: Taking
© The Author(s) 2019
Article reuse guidelines:
Stock of Police Gang Databases
sagepub.com/journals-permissions
https://doi.org/10.1177/1473225419883706
DOI: 10.1177/1473225419883706
journals.sagepub.com/home/yjj
in London and Beyond
James A. Densley and David C. Pyrooz
Abstract
Technology has ushered in a new era of intelligence-led and ‘big data’ policing, and police gang databases
are part of this paradigmatic shift. In recent years, however, gang databases have come under intense public
scrutiny. For example, Amnesty International and others argue that London’s Gangs Matrix is discriminatory
and violates data-protection laws. This article draws on evidence and examples from a wide range of sources
– gang legislation, surveys of young people, police gang records and research on gangs – to put the Matrix
controversy into broader context, and to adjudicate between common validity and civil liberties critiques
of gang databases.
Keywords
civil liberties, gangs, police databases, risk assessment, violence intervention
The London Metropolitan Police Service (hereafter, the Met) maintains a database of
purported gang members, termed the ‘Gangs Matrix’ (hereafter, the Matrix). The Matrix
was launched in 2012 as part of the UK Government’s ‘war on gangs’ (Densley and
Mason, 2011; Fraser et al., 2018), although there is precedence for UK gang databases
before that, with the National Football Intelligence Unit files on football hooligans in the
1990s (Garland and Rowe, 1999). The Matrix is a risk management tool used to respond
to the problems associated with the estimated 200 gangs in London, such as the 110 gang-
related homicides that occurred between 2016 and 2018 (Mayor’s Office for Policing and
Crime (MOPAC), 2018: 10). At any given time, there are an estimated 3000–4000 indi-
viduals listed in the Matrix from a London population in excess of 8 million.
A recent Amnesty International UK (2018) investigation raised serious concerns about
the Matrix. About 80 per cent of the individuals found in the database were Black, com-
pared with 13 per cent of London’s population, and many had a ‘zero-harm’ score, reflect-
ing the lowest risk of committing violence. A subsequent review of the database for the
Corresponding author:
James A. Densley, School of Law Enforcement and Criminal Justice, Metropolitan State University, 9110 Brooklyn Blvd.,
Brooklyn Park, MN 55445, USA.
Email: james.densley@metrostate.edu
12
Youth Justice 20(1-2)
Mayor of London (MOPAC, 2018) found that too many Black people were in the Matrix
compared with their likelihood of offending or their chances of being a victim. The regu-
latory Information Commissioner’s Office (ICO) (2018) further concluded that the list of
gang suspects had breached data protection laws. The net result has been calls from civil
liberties groups to either reform the ‘racist’ Matrix or abolish it altogether (Scott, 2019;
Williams, 2018).
Of course, controversy over police gang databases is neither new nor exclusive to
London (Huff and Barrows, 2015). In recent years, cities across the United States –
where gang databases originated – have faced a range of similar criticisms, including
how people came to be included on gang lists, how they could be removed, the overrep-
resentation of racial and ethnic minorities, and the way in which information was stored
securely (Durán, 2013; Jacobs, 2009; Klein, 2009; Muniz, 2014; Pyrooz and Densley,
2018). Gang databases are among the most controversial issues to have emerged in the
broader discussion surrounding ‘big data’ and intelligence-led policing (Ferguson, 2017;
Ratcliffe, 2016).
The aim of this article, therefore, is to better situate the debate surrounding the efficacy
of the Matrix. As Bryan and Arditti (2018) observe, the evidence-base for and against the
Matrix is ‘methodologically weak’ and ‘limited to single, often uncorroborated sources’,
hence why it is necessary to take stock of what is known about gang databases in general
in order to chart directions forward. We acknowledge the limitations of exporting research
findings from the United States to the United Kingdom, but begin by discussing why
police document gangs and gang members in the first place. Next, we detail the common
critiques of gang databases in London and elsewhere. We then draw on evidence and
examples from a wide range of sources – gang legislation, surveys of young people, police
gang records, and research on gangs – to provide broader context on gang databases and
to adjudicate between common critiques of them. We find there is merit to civil liberties
concerns and claims that gang data are invalid and unreliable, but the picture is more
complex than law enforcement or advocacy groups make it out to be. We conclude that
eliminating gang databases entirely could have the unintended consequence of making it
more difficult to understand and respond to violence in communities, but there are steps
that could and should be taken to improve gang databases in the future.
Why Do Police Document Gangs and Gang Members?
Starting with the premise that a gang is either an artificial construct or one that is impos-
sible to define for any practical utility, many critics take aim at how and why the concept
forms the basis of police prioritisation in the first place (Hallsworth, 2013). Some argue
criminologists and practitioners alike focus too much on gangs and not enough on youth
crime and violence, some of which is gang-related (e.g. Abt, 2019; Kennedy, 2011;
Sullivan, 2005). This is the ‘red herring’ argument – gangs are the shiny object that dis-
tracts us from the bigger problem of youth violence.
But the fact is that gang members account for a disproportionately large share of crime
and violence (Pyrooz et al., 2016), and in London especially, gangs are the ‘driving force’
behind an estimated 60 per cent of all shootings resulting in injury and 30 per cent of all
Densley and Pyrooz
13
Figure 1. The frequency and prevalence of gang and non-gang homicides in London, January 2008 to April
2018 (N = 1242 homicides). Primary axis refers to the count of homicides; secondary axis refers to the
prevalence of homicides that are gang-related. Gang homicide N = 265; non-gang homicide N = 977.
Source: Ager (2018). There is no readily available London homicide dataset in the public domain – Ager (2018) construct-
ed the original dataset using a number of open sources, most notably https://www.murdermap.co.uk, but also Google
News, https://courtnewsuk.co.uk and https://qgis.org/en/site/.
homicides (MOPAC, 2018: 4). As can be observed in Figure 1, youth homicide in London
is very much a gang issue. Gang-related violence is also ‘significantly more likely to
result in serious injury’, with 57 per cent of gang-related stabbings featuring a serious or
fatal injury, compared with 34 per cent for non-gang-related stabbings (MOPAC, 2018:
10). Beyond street violence, ‘the harm of gangs extends further . . . [to] violence against
women and girls, acquisitive crime and drug supply’ (p. 4). The MOPAC is of course a
political organisation publishing data for public consumption, which may be limited by
partiality. Such criticisms notwithstanding, few would claim that gangs are the youth vio-
lence problem in London, but even fewer should claim that the violence associated with
gangs is merely a sensationalised distraction (Densley, 2013).
The numbers are even more striking in the United States, where there are around 2000
gang-related homicides every year, which amounts to about 13 per cent of all homicides
(National Gang Center, 2012). The gang-related homicide rate in the United States alone
(about 2 per 100,000 persons) exceeds the total homicide rate in nearly every European
Union country. National Gang Center (2012) data also show that gang members constitute
one-fifth of 1 per cent of the US population, revealing an exceptionally high level of dis-
proportionality in the violence attributed to gangs. There are, of course, individual- and
14
Youth Justice 20(1-2)
macro-level explanations for such violence (Decker, 2007; Papachristos, 2009). Still,
those explanations do not negate the fact that violence drives police strategies (Abt, 2019;
Kennedy, 2011). Disputing whether or not gangs are sensationalised does little to comfort
a mother who lost her child to violence.
Law enforcement employ a range of strategies in response to the criminal and violent
activity associated with gangs. Sometimes, these strategies place police at the core of the
intervention, other times they are a partner in a broader collaborative. But regardless of
whether focused deterrence (Braga et al., 2018), gang crackdowns (Ratcliffe et al., 2017),
comprehensive strategy (Gebo et al., 2015) or civil gang injunctions (Ridgeway et al.,
2019) are deployed to respond to gang violence, all strategies share a common problem:
they need data on gangs and gang members. They need this information to ‘audit’
(Papachristos, 2012) the gang landscape of cities, target offenders with deterrence mes-
saging (Deuchar, 2013), and flag gang members who might benefit from social services
or even leave gang life, if given the right nudge (Roman et al., 2017).
Where do the data come from? Who is responsible for gathering it? Hypothetically,
...
To continue reading
Request your trial