Social Networks and Gang Violence Reduction

Abstract

This review traces the origins, development, and use of social network analysis in gang research and gang violence reduction strategies. Although early gang scholars intuitively recognized the networked nature of gangs and gang violence, such insights were not always leveraged by gang violence reduction efforts that became increasingly enforcement-centric throughout the twentieth century. This review describes these historical shifts, the recent advent of social network analysis in research and gang interventions, and future directions that research and interventions can take to develop a more victim-focused approach to gang violence reduction. Expected final online publication date for the Annual Review of Law and Social Science Volume 13 is October 13, 2017. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Full text

LS13CH19-Papachristos ARI 14 July 2017 10:7
R
E
V
I
E
W
S
I
N
A
D
V
A
N
C
E
Social Networks and Gang
Violence Reduction
Michael Sierra-Ar´
evalo and Andrew V. Papachristos
Department of Sociology, Yale University, New Haven, Connecticut 06511;
email: andrew.papachristos@yale.edu
Annu. Rev. Law Soc. Sci. 2017. 13:19.1–19.21
The Annual Review of Law and Social Science is
online at lawsocsci.annualreviews.org
https://doi.org/10.1146/annurev-lawsocsci-
110615-085017
Copyright c
2017 by Annual Reviews.
All rights reserved
Keywords
gangs, social networks, violence prevention, street outreach, victimization
Abstract
This review traces the origins, development, and use of social network anal-
ysis in gang research and gang violence reduction strategies. Although early
gang scholars intuitively recognized the networked nature of gangs and gang
violence, such insights were not always leveraged by gang violence reduc-
tion efforts that became increasingly enforcement-centric throughout the
twentieth century. This review describes these historical shifts, the recent
advent of social network analysis in research and gang interventions, and
future directions that research and interventions can take to develop a more
victim-focused approach to gang violence reduction.
19.1
Review in Advance first posted online
on July 21, 2017. (Changes may still
occur before final publication.)
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
INTRODUCTION
The street gang occupies a unique, almost mythical position in American society. For some, street
gangs are public enemy number one, a scourge on cities and a representation of the decay of
contemporary youth culture (Chic. Crime Comm. 1995). For others, gangs represent a reality of
everyday life—at times a necessity—for young people trying to feel safe or empowered as they
walk through dangerous streets in disadvantaged and disempowered neighborhoods (Garot 2010).
Regardless of where one might fall on such a spectrum, few deny that certain gang behaviors—
especially violence—represent a serious social problem. In cities like Chicago (Papachristos &
Kirk 2015), Boston (Braga et al. 2014), and Los Angeles (Tita et al. 2010), gangs are involved
in somewhere between 30% and 60% of all homicides annually. Such figures say nothing of the
fear instilled in community residents by acts of gang violence that often occur on neighborhood
streets, involve firearms, and include a greater number of perpetrators and victims (Curry et al.
2002, Maxson et al. 1985).
How we conceive of gangs greatly influences society’s approaches to gangs and gang violence. If
gangs are seen as wayward youth or lost boys, we might consider a softer touch, including diversion
programs, social services, and educational or employment opportunities to get young people back
on track. If gangs are nothing more than “urban terrorists” (Penley 2011), however, then one
might feel less inclined to glove the heavy hand of the criminal justice system. Complicating
matters is the manner in which conceptions of gangs are influenced by the intersection of race,
gender, and geography (Ludeke 2006, Papachristos & Hughes 2015, Petersen & Howell 2013):
The solution to gang violence might well vary with the skin color or neighborhood of the gang
member in question.
The social sciences have a lot to say about street gangs and can provide important nuance to
sweeping generalizations of gangs. For example, most street gangs in the United States are not
highly organized, corporate, drug-dealing enterprises but rather small, informal groups that are
limited in size, scope, and capacity for crime (Howell 2012, Klein 1995). Yet rigorous research
on street gangs has not always led to equally nuanced policy recommendations or gang violence
reduction programs. At times—especially during the 1960s and at some points in the 1990s—gang
research was central to the design of gang violence reduction efforts. At other times, however,
especially in the 1980s, social science was largely absent from such discussions. Today, policy
makers and funding agencies have once again made a cognizant (at least external) push toward the
promotion of data-driven solutions to social and policy problems, creating, perhaps, an important
moment for scientists to engage in such pressing debates.
This article examines how one area of science—social network analysis—can play an important
role in informing efforts to reduce gang violence.In an era of almost ubiquitous digital intercon-
nection, many confound the ideas of social media and social networks; the former refers to digital
platforms that allow us to connect with our social relations, whereas the latter refers to the tan-
gle of kinship, friendship, and other relationships that comprises our social worlds. The field of
social network analysis has been around since at least the 1930s, but the massive proliferation
of handheld devices, the advent of Big Data, and exponential increases in statistical computing
power have spurred the proliferation of network analysis as a tool for asking new questions and
revisiting old ones. Indeed, some have argued that a new field of network science has emerged
that transcends traditional boundaries of the physical, social, and hard sciences as similar models,
theories, and methods have developed to explain brain organization (Bullmore & Sporns 2009),
the spread of disease (Eubank et al. 2004), and the social contagion of political influence (Kahler
2015).
19.2 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Although criminology has long theorized about the role of networks qua criminal or deviant
groups, the field is a latecomer to the implementation of network methods in the study of gangs
(Papachristos 2011, Sierra-Ar´
evalo & Papachristos 2015a). The history of gang research is ripe
with examples of network thinking; some of the earliest studies recognize that the gang as a group
is constituted by the connection between its members. Recent advances in data, methods, and
computing power have enabled these foundational insights to be reexamined and have given new
life to our understanding of gangs, gang structure, and gang behavior.
We believe network science can advance theoretical and empirical investigations of street
gangs, both academically and in practical attempts to reduce gang violence. Social network anal-
ysis also holds potential for new approaches to gang violence reduction. By measuring (rather
than assuming) the underlying structure of street gangs (McGloin 2007, Morselli 2009), gang
violence reduction efforts can provide a roadmap that can more strategically identify the groups
and individuals driving gang violence—a much-needed reprieve from overly broad policies like
stop-and-frisk that generate massive societal inequities (Pettit & Western 2004, Wakefield &
Wildeman 2016, Western & Wildeman 2009).
GANGS, NETWORKS, AND GANGS AS NETWORKS
What Is a Gang?
Any academic study of street gangs must spend the obligatory ink on the definitional debate, the
longstanding attempt to decipher what does and does not constitute a gang. To summarize, the
central points of contention in this debate are the extent to which any definition is (a) structural,
based on group organization or leadership, especially its structure or organization; (b) behavioral,
focusing on the involvement of the group in specific behaviors, usually related to crime and
delinquency; or (c) some combination therein (Bouchard & Spindler 2010, Spindler & Bouchard
2011).
Although both structure and behavior are important elements for conceptualizing what is and
is not a gang, we need not definitely adjudicate between these aspects of the gang for the purposes
of this review. Instead, we start from the premise that the gang is first and foremost a social
network. A gang, both theoretically and empirically, is defined by the social relationships of its
members with each other and with those outside the group. Though it is true that gangs can
exist as groups in part because there are those that are not in the gang, it is crucial to recognize
that gangs and gang members are embedded in a complex web of social relationships that cut
across social, geographic, and even virtual space (Fleisher 1998, Pattillo 1998, Sierra-Ar´
evalo &
Papachristos 2015a). As such, though the relationships between gang members that share a group
identity, neighborhood, and set of conflicts are important, so too are the relationships of the
gang’s members to other gangs, friends, parents, schoolmates, and the police. It is precisely this
constellation of relationships that social network analysis is suited to measure and analyze.
What Is Social Network Analysis?
Social network analysis (hereafter, simply SNA) refers to a set of theoretical and methodological
tools focused on observing and modeling the relationships among actors (Carrington et al. 2005,
Wasserman & Faust 1994). SNA is based on graph theoretical notions of networks that consist of
two (or more) types of data: vertices, often visualized as points or nodes that represent the actors
under investigation, and edges, depicted as lines or arrows that represent the relationships among
www.annualreviews.org •Social Networks and Gangs 19.3
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
the actors.1In combination, these vertices and edges display the sets of relationships between
actors, and the resulting visualization of these connections is referred to as a network graph, or
sociogram.
More practically, SNA tries to understand how social relationships affect what we feel, think,
and do. From its beginnings in field-driven anthropology and group-focused social psychology and
sociology, SNA has developed into a much broader field of network science that currently stretches
from theoretical physics and computer science to public health interventions and popular culture.
Network methods and models have been used to describe a whole range of phenomena, including
the diffusion of ideas and technologies (Angst et al. 2010, Cho et al. 2012, Ferrier et al. 2016),
the spread of disease (Christakis & Fowler 2007, Eubank et al. 2004), political and organizational
behavior (Boushey 2010, Elliott et al. 2014), and friendship and romantic relationships (Ball &
Newman 2013, Bearman et al. 2004, McFarland et al. 2014).
Though scholars of crime, law, and deviance arrived a bit late to this network turn in the
academy (Papachristos 2011), a growing body of work recognizes the salience of relational con-
cepts and methods in the study of street gangs. With the increase of computing power and the
development of more advanced network methods, SNA is being used to ask new questions about
gangs and gang behavior, as well as to revisit and reorient long-standing questions about gang
structure, membership, cohesion, and violence. In fact, this recent work echoes the long-standing
insight of storied gang scholarship that identifies the importance of intra- and intergang ties not
only to gang life but to the role of gangs in neighborhood and urban life more broadly.
THE EARLY YEARS: THE GANG AND STREET CORNER SOCIETY
Fredric Thrasher and the Gang
In what is widely considered to be the first scientific study of gangs, Fredric M. Thrasher [2013
(1927), p. 3] summarized his raison d’ ˆ
etre for studying gangs: “The gang, in short, is life, often rough
and untamed, yet rich in elemental social processes significant to the student of society and human
nature.” For Thrasher, the gang was a worthy subject of scientific investigation not only because
it was a unique social phenomenon that demanded its own classification and explanation but also
because understanding of the gang stood to shed light on social processes that affected urban life as
a whole. In his sprawling study of 1,313 gangs in 1920s Chicago, Thrasher plants the flag for gang
research within sociology and establishes the foundational notion of gangs as part of the social
fabric of cities. The gangs studied by Thrasher are not alien invaders of urban neighborhoods
but rather the by-products of the same social forces that generate other social groups within the
industrial city; though gangs are unique entities within the city’s broader ecology, they are but
one piece of an interconnected urban tapestry.
Although Thrasher does not use formal network methods, models, or terminology, he spends
many pages discussing the connected nature of the gang and gang life. For example, he de-
scribes how gangs can form in the context of a sports team or club, solidified through common
interests or their shared expulsion/exclusion from neighborhood institutions. Thrasher also rec-
ognizes that gangs and gang members are indelibly tied to the social and geographical interstices
of the city—the gaps and fissures between dominant social institutions and physical spaces. These
places become part of gang identity—Forty-Seventh Street is synonymous with the Bat-Eyes, just
1See Wasserman & Faust (1994) for a discussion of the mathematical underpinnings of SNA, as well as Scott & Carrington
(2011) for an updated account of SNA’s historical and methodological development.
19.4 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
as Sixty-Third Street is Dirty Dozen territory—and are crucial to delineating the contours of
intergang relationships. As Thrasher describes, alliances and feuds are maintained because of and
across space, such as in the alliance between gangs in Bucktown to more effectively fight rivals
across the river in Pojay Town. Even without the aid of network methods, Thrasher’s view of
gangs, gang conflict, and life in the gang reveals an implicit relational understanding of the con-
nected nature of the gang and its activities, as well as the gang’s relationship to local businesses,
schools, politics, and the broader ecology of the city.
William Foote Whyte and Street Corner Society
Whereas Thrasher went for breadth in understanding the relationships between the gang and an
entire city, William Foote Whyte’s classic, Street Corner Society, went for depth by documenting
the relationship between street corner groups and Boston’s Northend community. Like Thrasher,
Whyte presents many important findings that are relevant to today’s gang scholar; one of Whyte’s
more significant contributions is what is perhaps the first known network analysis of a street gang—
the Nortons. Beginning with the deceptively simple declaration, “The Nortons were Doc’s gang,”
Whyte [1969 (1943), p. 3] highlights leadership, collective identity, and the grouped nature of the
gang, laying out topics that have interested gang scholars for decades.
Key to life in Cornerville are the “street corner groups”—what many might call street gangs.
Although Whyte does mention conflict and even a few fights, the bulk of his description of the
Nortons and its members centers on the group and its groupness. True to his anthropological
roots, Whyte describes how group structure orients life for street corner groups, his analysis
focusing on the groupness of the Nortons and the connections between individuals, groups, and
institutions in Cornerville that perpetuate and interact with these street corner groups. As such,
Whyte does not exclusively talk about Doc and the others that make up the Nortons; instead, he
situates the intragroup relationships of the street corner group within a broader conceptualization
of the Cornerville ecology, and he also describes the relationships of members to non-Norton
friends, romantic interests, family members, political actors, local institutions, and other street
corner groups. Though, like Thrasher, Whyte does not discuss the nodes, ties, or formal network
structural patterns that typify modern network studies, Whyte’s analysis of Doc’s gang nonetheless
yields a sociogram of the Nortons’ underlying structure (see Figure 1).
Whyte’s discussion and depiction of Doc as being at the top of the Nortons’ social hierarchy
provide one of the first academic descriptions of gang structure. Moreover, Whyte’s visualization
of the underlying structure of the Nortons—all based on his detailed observations of the group over
time—uses the position of members to explain their relative status in the group. By combining the
links between Norton members and showing how these connections relate to intragroup status,
Whyte emphasizes the interaction between group structure, status, and leadership. Far from that
of a democratically elected leader, Doc’s position as the Nortons’ leader is a product of his rela-
tionships with various group members, such as his physical domination of Nutsy to assume control
of the Nortons [Whyte 1969 (1943), p. 4]. Although he was not a network scientist in the modern
sense, Whyte’s anthropological approach and hand-drawn sociogram privilege group processes
and structure in his exploration of the gang, its members, and their relationship to other commu-
nity actors and institutions.2Coupled with his ethnographic analysis, this early network analysis
of the Nortons helped shape the way scholars to this day conceptualize and measure gang life.
2It should be noted that Whyte’s earlier anthropological predecessors pioneered early versions of network diagrams, though
not in the context of cities or street gangs (see White & Johansen 2005).
www.annualreviews.org •Social Networks and Gangs 19.5
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Doc
Mike Danny
Long John
Nutsy Angelo
Frank Fred
Joe Lou
Alec
Carl
Tommy
Corner boy
Line of inuence
Figure 1
Social network diagram of Nortons’ street gang from William Whyte’s (1943) Street Corner Society. Each box
represents a member of the Nortons’ gang, and each of the ties represents a channel through which
communication, resources, and influence travel between individuals. The position of the boxes indicates
relative status of the Nortons’ members as determined by Whyte, with higher-positioned boxes denoting
higher status. Adapted with permission from Whyte 1943.
THE 1950s AND 1960s: SUBCULTURE, GROUP PROCESSES,
AND STREET OUTREACH
Beyond his important contributions to the academic understanding of street gangs, Whyte’s study
of the Nortons and Cornerville parallels and informs subsequent developments in gang violence
prevention efforts. For example, Whyte documents how Doc was tapped by Mr. Smith, the head
of the Cornerville Settlement House—a neighborhood institution that housed social workers,
provided services, and offered a recreation area—to run one of the three recreation centers. With
Doc spending his time in the center, “[t]he remaining Nortons responded by spending some of
their time in the center, but this changed the nature of their activities” and pulled them off the
corner as well [Whyte 1969 (1943), p. 43]. Though we cannot conclusively know that Mr. Smith
planned to change the Nortons’ behavior by way of influencing Doc, it is clear that the same status
and respect afforded to Doc as leader of the gang were channels through which the settlement
house facilitated changes in the Nortons’ activities.
Whyte’s example is an early instance of an approach to gang intervention work that persists to
this day: leveraging individuals with insider or gang experience as potential avenues for intervention
(Spergel 2007). These insights into gang behavior and how to affect it became vitally important
to the gang research—as well as the popular fascination with gangs—that gathered steam in the
1950s as cities boomed and rates of juvenile offending rose in the postwar United States (Bursik &
Webb 1982, Monahan 1961, Teeters & Matza 1959). Looking for answers to growing public fears
over the future of America’s youth, pundits and politicians pointed to potential causes ranging
from the dissolution of the American family to comic books, television, and rock ‘n’ roll (Gilbert
1988, Wertham 1954). To make matters worse, horrific (and widely publicized) cases of gang
19.6 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
violence perpetrated by young men, such as the infamous Cape Man murders in New York City,
fueled public fear of young, violent street gangs (Schneider 2001).
Recognizing the need for concerted academic attention on rising rates of crime and delin-
quency, several gang scholars cooperated with locally embedded outreach workers to glean better
understanding of the group processes and structures that set the bounds of the gang and gang
life. Beginning with Clifford Shaw’s Chicago Area Project in the 1930s—credited by some as the
first time so-called indigenous workers were used to make connections with delinquent youth
(Kobrin 1959, p. 27)—the mid-twentieth century saw the rise of “detached worker” programs
(Klein & Maxson 2006, p. 164). These programs employed individuals from the neighborhoods
under study, often with previous gang involvement themselves, leveraging their position within
the community to forge relationships with delinquent boys and reintegrate them into prosocial
institutions (Spergel 1995).3
Recognizing the value of these individuals (also termed street or outreach workers) for gaining
entr´
ee into the lives of delinquent gang members, several noted gang scholars cooperated with
workers to rigorously study group structure, leadership, membership, and cohesion. For example,
research in Boston (Miller 1957, 1962), Chicago (Short & Strodtbeck 1963, 1965; Spergel 1966),
and New York (Cloward & Ohlin 1960, N.Y. City Youth Board 1960) was all done in cooperation
with outreach workers, representing rare exceptions to gang research that often takes place in a
“policy vacuum” (Decker & Van Winkle 1996, p. 9).4
At roughly the same time, the roots of what would eventually come to be formal SNA were
also taking hold in the social sciences with social psychologist Jacob Moreno’s (1941) development
of sociometry. Conveniently enough, Moreno’s first bit of sociometry analyzed runaway girls
from a juvenile detention facility, thus offering criminologists an easy extension to the study of
delinquency. Malcolm Klein, another central figure in gang research, was inspired to incorporate
network thinking into his own work on gangs in Los Angeles.
In cooperation with outreach workers who produced contact reports detailing where, when,
and with whom they contacted “576 male Negro gang members,” Klein & Crawford (1967,
p. 63) were able to, for the first time, measure and visualize variation in gang cohesion—the
connection between members and members’ connection to the group—using relational data.
This, to our knowledge, produced the second known sociogram of a street gang. Like Whyte’s
early sociogram of the Nortons’ group structure, Klein & Crawford’s use of relational data shows
that gangs are not homogeneous social groupings and are instead composed of smaller cliques and
subgroups operating within the larger clusters that street outreach workers observed. Further,
their analysis provides one of the earliest demonstrations of how variation in gang structure is
related to differences in offending: More cohesive gangs, according to Klein & Crawford (1967),
were more delinquent.
Unfortunately, although these cooperative endeavors between research and outreach workers
provided the means for important advances in the academic understanding of gangs and their
behavior, this did not ensure that contemporary gang outreach programs themselves benefited
from such insights. In fact, as Klein (1971) argues in a review of pre-1970s gang literature, out-
reach programs that depended on the relationships between street workers and gang members
3See Kobrin (1959) for a more thorough description of the theoretical and practical motivations behind the use of detached
workers for street outreach work.
4See Hughes (2015), Maxson (2015), and Moule (2015) for a more thorough discussion of these research projects, as well as
the legacies of these notable gang scholars more generally.
www.annualreviews.org •Social Networks and Gangs 19.7
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
instead of rigorous research showed little (if not counterproductive) effects on gang deviance (Klein
& Maxson 2006, pp. 90–91). And though stalwarts like Irving Spergel (1966, 1995; Spergel &
Grossman 1997) kept the flame of research-outreach collaborations alive, a lack of rigorous pro-
gram assessment has produced limited evidence that outreach workers can reduce gang violence
(Spergel 2007). This lack of rigorous evaluation of such outreach-driven interventions highlights
the gap between gang research and gang interventions that expanded throughout the coming
decades, and which extends to the present day (Klein 2011).
THE 1970s TO THE 1990s: DRUGS, LEGAL INTERVENTIONS,
AND THE WAR ON GANGS
Following the 1960s, negative perceptions of welfare programs (Gilens 2010), postindustrial
economic change (Wilson 1987), and rising gang violence (Block & Block 1993, Hagedorn
1998) combined to justify a punitive turn in criminal justice away from the “rehabilitative ideal”
of earlier decades (Garland 2002). In some cities, Chicago in particular, increasing levels of
postindustrial gang violence were accompanied by restructuring of some gang organizations
(see Venkatesh 1997 and Levitt & Venkatesh 2000), with post-1960s gangs that became heavily
involved in the underground drug economy becoming more corporatized and violent to control
lucrative drug-dealing turf.
In the wake of Nixon’s declaration of drugs and drug abuse as “public enemy number one”
(Friedersdorf 2011), approaches to gang intervention (and social ills more generally) took an
increasingly punitive turn. Rather than address root causes of poverty, crime, and addiction, aid
and outreach workers were replaced with police and prosecutors. This move to the truncheon
in lieu of assistance is part of the broader shift from the War on Poverty to the War on Crime
described by historian Elizabeth Hinton (2016), and it had profound effects on how gangs and drugs
were addressed. As the War on Drugs intensified throughout the 1970s and 1980s, intervention
evolved to attack a gang problem that quickly became a “gang and drug problem” (Coughlin &
Venkatesh 2003, p. 43; Rosen & Venkatesh 2007). At the street level, specialized gang units began
to form as early as the 1960s (Katz 2001), and curfew laws (Harvard Law Review 1994) and no-
knock search warrants (Allegro 1989, Balko 2013) became key weapons in what became a veritable
War on Gangs.
Legislatively, a popular method to combat the gang-drug problem was the use of civil gang
injunctions. Civil injunctions do not require that gangs be involved in serious crime or violence
and instead rest upon a definition of gangs as “public nuisances” (Rosen & Venkatesh 2007).
California led early implementations of such injunctions in cities like Los Angeles and Oakland
(Yoo 1994), making it an arrestable offense for gang members who were part of an injunction
to congregate in public, be in specified zones, or wear clothing or colors associated with gang
membership (Maxson et al. 2003). In much the same way, Chicago modified its anti-loitering laws
to keep known or suspected gang members off of street corners [City of Chicago v. Morales (1999)].5
Although ultimately deemed unconstitutional by the Supreme Court, Chicago’s gang loitering
laws epitomized the criminalization of a wide range of gang behaviors beyond those associated
with violence—and in many cases individuals not directly involved in gangs at all.
Federal agencies also innovated in the War on Gangs. In an effort to curtail gang criminality,
especially drug dealing (whether or not it was accompanied by violence), prosecutors turned to
5As Rosen & Venkatesh (2007) discussed, gang injunctions are rooted in logic drawn from Kelling & Wilson’s (1982) theory
of broken windows, which suggests that controlling low-level (gang) infractions can reduce more serious crime and violence.
19.8 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
the RICO (Racketeer Influenced and Corrupt Organizations) statute that was previously used
to prosecute organized crime families. In keeping with the legal requirements needed to win a
RICO conviction, police and prosecutors were required to show that a gang had a hierarchical
structure and that the gang collectively used criminal activity to stimulate profits (Gerber 1987).
However, most gangs are in fact not highly organized criminal conglomerates (Decker et al. 1998,
Hagedorn & Macon 1988, Huff 1998). As such, RICO-based prosecutions face great difficulties
when trying to fit stark legal frameworks to the much greater number of small, fluid gangs that
engage in nonspecialized, cafeteria-style offending (Klein 1968, 1995; Needle & Stapleton 1983;
Short & Strodtbeck 1965).
These enforcement-centric efforts demonstrate how the criminal justice system focused on
structure to the exclusion of group processes and superimposed a crude legal typology predicated
on hierarchical organization. Although there are indeed cases of successful RICO prosecutions of
sophisticated gang organizations (Bonney 1993, p. 604; Gibeaut 1998), the promulgation of the
largely inaccurate view of gangs as “groups with national power affecting the nation’s economic
and political structure” (Bonney 1993, p. 606) only fanned the moral panic around gangs and
violence, increasing political and public pressure to make arrests and increase sentences (Barkow
2005, Moore 1990). Under this pressure, criminal justice agents across the country continued their
punitive efforts without the nuance provided by social science research—much less a networked
understanding of gangs and gang violence.
POST-1990s: THE BIRTH OF NETWORKED GANG INTERVENTIONS
Despite the dramatic rise in enforcement-centric strategies after the 1960s, adult and juvenile
homicide trends continued to rise throughout the 1980s and early 1990s (Fox 1996, Fox & Zawitz
2007). Recognizing that said strategies were not a cure-all for gang violence, cities like Boston—
where homicides increased from 100 in 1990 to 152 in 1991 (Braga et al. 2001a)—looked for
new alternatives. Drawing on research that showed that street violence concentrated in particular
places and was driven largely by young males involved with gangs or drug crews (Braga 2003, Cook
& Laub 2002), a team of researchers and practitioners in Boston looked to reduce the breadth of
legal intervention while fostering cooperation between police and community stakeholders.
Instead of enforcement strategies that target gangs and gang behavior broadly, or RICO-
like statutes that necessitate hierarchical gang structure, the Boston effort developed a focused
deterrence framework that begins from the assumption that addressing gang violence depends
on targeting specific gangs and gang members (Kennedy 1997). This logic drew from a widely
recognized idea that the majority of violent crime in any particular community is usually driven
by a small population of active and repeat offenders (Kennedy 1997). For example, although a city
might have dozens of gangs or groups that engage in illegal activity, there is variation in any gang’s
involvement in street violence, and there are many nonviolent gangs as well (Kennedy et al. 1997,
Sierra-Ar´
evalo & Papachristos 2015b). Furthermore, only a small number of individuals within
any gang are likely to be actively engaged in violence. A focused deterrence approach narrows the
scope of gang violence reduction (Braga et al. 2001b, Kennedy et al. 1996) and funnels limited law
enforcement, social service, and community resources to those gangs and their members that are
actually involved in street violence.
The Boston effort also employed formal network analysis to help focus its intervention. To
identify violent groups, the Ceasefire team developed a process known as a group audit, a focus
group–style data collection session used to gather information on the location, membership, and
activities of gangs in the city from gang “experts” in each jurisdiction (Kennedy et al. 1997, Sierra-
Ar´
evalo & Papachristos 2015b). A key component of the audit records data on which groups are
www.annualreviews.org •Social Networks and Gangs 19.9
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
actively involved in violent crimes (especially those involving firearms), as well as information
on the alliances or violent relationships between gangs that undergird reciprocal gang violence
(Decker 1996, Papachristos 2009, Papachristos et al. 2013). In other words, the audit gathers data
on the precise network structure of gang conflict in a city: which gangs are involved in conflict
with which other gangs. These relational data support network analyses that can measure and
display neighborhood- or city-wide gang conflict networks, as well as the group-level structure of
individual gangs (Kennedy et al. 1997, Sierra-Ar´
evalo & Papachristos 2015b).
In turn, these network maps are used to guide which groups are to be included in the focused
deterrence intervention, termed the call-in. The call-in is an hour-long meeting between members
from violent groups and a collaborative group of law enforcement; social service providers; and
community stakeholders, such as clergy, ex-offenders, and family members of gun violence victims
(Braga et al. 2008, Pegram et al. 2016, Sierra-Ar´
evalo et al. 2016). At this meeting, invitees receive
a message that emphasizes a unified, moral voice against violence, offers connection to social
services, and makes clear the legal consequences for individuals and groups that continue to
engage in violence (Crandall & Wong 2012).
Focused deterrence strategies stand out not simply for their departure from the single-minded
application of criminal justice resources but also because of their explicit use of network data
and methods to guide interventions. Theoretically, this model reengages with the importance of
intragroup processes and leverages informal dynamics within the group to encourage behavioral
change. Specifically, the intervention focuses on group accountability for violence and looks to
engender group self-policing of violence to prevent law enforcement attention from being fo-
cused on the group and its members (Kennedy 2011, Natl. Netw. Safe Communities 2013). With
the support of formalized network analysis that can capture these processes and structures with
concrete network metrics like centrality, density, and cohesion, focused deterrence’s cooperative
involvement of law enforcement, social services, and the community has shown promising results
in small and large US cities (see review by Braga & Weisburd 2015).
NETWORKED GANG RESEARCH
The use of network data and methods in gang violence reduction efforts parallels the growing
use of SNA in gang research. Over the last decade, a handful of researchers began applying
formal network methods to the study of street gangs and criminal offending, such as in work by
Morselli (2009, 2010), which investigates the role Hells Angels members play in a drug distribution
network, or by McGloin (2007), who draws on relational data gathered in interviews to concretely
map the underlying structure of street gangs in Newark, New Jersey. Additionally, these methods
have been used to revisit and revise our understanding of central gang concepts around group
structure and processes.6For example, Hughes (2013) uses relational data gathered on Chicago
gangs by Short & Strodtbeck (1965) to test the relationship between gang cohesion and deviant
behavior. In contrast to the finding that more cohesive gangs engage in more delinquency and
violence (Klein 1995, Klein & Crawford 1967), Hughes finds that gang cohesion is unrelated
to delinquency, but that members of less cohesive gangs are more likely to engage in violence.
Similarly, Papachristos (2006) recodes Suttles’s (1968) ethnographic observations of the Erls gang
and uses network concepts like centrality, cohesion, and structural equivalence to nuance previous
depictions of leadership, status, and group structure.
6See Sierra-Ar´
evalo & Papachristos (2015a) and Papachristos (2011) for more thorough discussion of this network shift in
gang research and criminology more generally.
19.10 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Network methods are also being used to extend research on the spatial concentration of gang
violence that has previously been restricted to measurement at the neighborhood, street, or gang
turf level (Block 2000, Braga et al. 2010, Brantingham et al. 2012, Tita & Greenbaum 2009, Tita
et al. 2005). By reexamining how gangs are tied to one another through networks of violence and
conflict, this research demonstrates how group processes like contagion and reciprocity influence
gang violence that is simultaneously embedded in social and physical space (Braga et al. 2010,
Papachristos et al. 2013, Tita & Radil 2011). Though gang turf boundaries continue to be im-
portant predictors of gang violence, this line of research emphasizes that intergang conflict must
be understood as more than a collection of individual events—gang conflict creates an enduring
network structure that transcends the individual, with the dominance struggles between groups
culminating in individual acts of violence that reify the broader system of violence (Papachristos
2009).
In addition to gun violence being highly concentrated in geographic space, recent research
shows that violence is concentrated in social networks. In one high-crime Boston community, for
example, 85% of nonfatal shooting victims can be found in a co-offending network of less than 5%
of the community’s population (Papachristos et al. 2012); in Chicago, 70% of fatal and nonfatal
shootings can be found in a network that comprises less than 6% of the city’s total population
(Papachristos et al. 2015b). This work shows that individual victimization risk is a product of how
individuals are connected to victims of violence—someone directly or indirectly tied to a victim
(or multiple victims) is at an elevated risk of victimization compared with someone who is not.
These studies also extend the link between gang membership and violent victimization (Peterson
et al. 2004, Pyrooz et al. 2014) with the finding of a spill-over effect of gang membership: Being
directly or indirectly tied to a gang member in one’s network dramatically increases one’s risk of
victimization even if one is not a gang member oneself (Papachristos et al. 2015a).
NETWORKED GANG INTERVENTIONS
Just as many of the early insights on group processes helped shape some of the early outreach
and intervention work on gangs, so too are recent networked insights on gangs poised to make
meaningful contributions to intervention, prevention, and policy efforts. As McGloin & Rowan
(2015) argue, not only can SNA can be used to guide interventions through the identification of
gang conflicts and influential members, it can also provide metrics for assessing the efficacy of a
given intervention beyond a tally of shootings or arrests (see also McGloin 2005). In much the
same way, we maintain that the ability of SNA to formally measure cohesion, membership, and
structure in a relational way is invaluable for untangling what works, what does not, and what is
doing more harm than good.
The successful use of network methods and relational data in focused deterrence interven-
tions provides a useful case in point (Braga et al. 2013). We believe that other violence reduction
efforts are similarly well-suited to employ SNA, especially if they extend beyond the traditional
law enforcement focus and proceed in collaborative and transparent ways. For example, Spergel’s
(1995, 2007) long-running comprehensive gang model explicitly recognizes the importance of
gang members’ ties to social services and local institutions, and Cure Violence treats gang vi-
olence as a contagious phenomenon that can be strategically interrupted by outreach workers
(Slutkin 2013). In effect, both of these highly regarded programs are operating under an implicitly
networked framework: Spergel’s model looks to build prosocial links between individuals and local
institutions, and Cure Violence aims to disrupt negative ties between gangs. However, neither
program uses formal network methods to guide its interventions or evaluate its effectiveness. The
network foundation has already been set—what is needed now is a concerted effort to collect
www.annualreviews.org •Social Networks and Gangs 19.11
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
relational data on gangs and their members and then use said data to guide and assess these gang
interventions.
We suggest that data collection efforts like group audits can be used outside the focused
deterrence framework to gather relational data and guide other gang violence interventions (Sierra-
Ar´
evalo & Papachristos 2015b). Such audits need not be restricted to law enforcement experts;
instead, the audit process can (and should) be carried out with social service providers, street
workers, teachers, and other unofficial experts that can speak to the tangle of gang members’
overlapping relationships. Along similar lines, gang interventions would do well to take advantage
of existing relational data, such as police data on arrests that have been used to assess individual-
level risk of victimization (Papachristos et al. 2012, 2015a). These data are particularly useful in
that the network analyses they support can help direct limited resources to individuals at a high
risk of victimization, even if they are not themselves identified gang members.7
For instance, Figure 2 provides an example of how a network analysis can provide insight into
gang violence in the city of East Palo Alto, California. Using data on arrests, police field contacts,
homicides, and gang membership from 2008 through 2013, we constructed a network in which
each node represents a unique individual who was arrested or interviewed by police, and in which
each tie represents an instance of coarrest or cocontact. The larger nodes represent identified
members of a Norte ˜
no gang set, and darker nodes represent homicide victims. We restrict this
network to the single largest component (n=266) of the total network (n=4,370).
Immediately apparent from Figure 2 is the clustering of gang members in the network: Most
of the identified Norte ˜
nos are located in the large clusters near the center and lower-left areas
and are tied either directly with or within a few steps of other Norte ˜
nos. In practical terms, this
network structure reflects that Norte ˜
no members tend to be seen and arrested together by police.
That said, this network also makes it clear that Norte ˜
no members are not exclusively being seen or
arrested with fellow members, and instead are hanging out and committing crimes (in some cases,
exclusively) with non-Norte ˜
nos—an insight supporting the long-standing ethnographic evidence
that shows gang members are part of much larger social, familial, and neighborhood networks
(e.g., Fleisher 1998).
Figure 2 also highlights the concentration of gun violence in an exceedingly small segment of
the population. Out of the 36 homicides that occurred in East Palo Alto during the time period,
20% (n=7) can be found in a network component of 266 individuals, or less than 1% of the
city’s population. Interestingly, though nearly 43% of the victims in this network component are
gang members, the remaining victims are not. Much in the same way that the structure of this
network belies an understanding of gangs as negating relationships outside of the gang, so too
does it suggest that an approach that focuses exclusively on identified gang members is likely to
miss important contours of a city’s broader violence landscape.
THE FUTURE OF NETWORKS AND GANG VIOLENCE PREVENTION
Our example analysis of East Palo Alto demonstrates the possibilities of using police-generated
data, but the application of SNA to understanding and doing something about gang violence can
and should extend beyond criminal justice data and applications. Future research should explore
how SNA can be extended to other administrative data, such as school records, or else implement
7This type of analysis has been used recently in individual-level “custom notifications” as part of focused deterrence efforts
(Kennedy & Friedrich 2014), though to our knowledge there is no evaluation of this technique’s effectiveness above and
beyond that of program call-ins.
19.12 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Associate victim
Norteño victim
Associate
Norteño member
Figure 2
Co-arrest network of members of Norte ˜
no gang members, associates, and homicide victims in East Palo Alto, CA. Each of the nodes
represents a unique individual, and each tie represents an instance in which two individuals were arrested together. The co-arrest
network was constructed using East Palo Alto arrest data from 2008 to 2013.
survey- or interview-based network questions within schools, social service providers, and other
institutions to capture and analyze gang members’ myriad social ties. Additional administrative
data (e.g., foster care, social services, health services) might also provide information on system
use that might otherwise disrupt or impede gang behaviors and negative network ties.
The insights provided by the networked gang studies reviewed here notwithstanding, these
studies pay little attention to those individuals who, despite being embedded in gang networks,
manage to avoid the negative effects of gang membership and conflict. That is, although past
research clearly shows how gang membership or being tied to gang members is associated with a
host of negative outcomes, we know markedly less about why some individuals—though embedded
in the same high-risk networks as victims—appear to be resilient to victimization. For example,
though we know individuals in Figure 2 that are embedded in network clusters with more victims
www.annualreviews.org •Social Networks and Gangs 19.13
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
are at an elevated risk of being shot, it is unclear why some individuals who are at a statistical
disadvantage are not, in fact, shot or killed.
One way to examine such resiliency would be to gather additional data on non-gang ties—
such as those to family, school, employment, or church—that might inoculate individuals against
exposure to violence, or that otherwise provide mitigating influences from outside the gang. In
addition to learning more about how these ties might function as network conduits to resiliency-
enhancing social capital, gathering such relational data has the benefit of more fully capturing the
multiplex ties that link individuals embedded in overlapping social groups (Fischer 1982, Fleisher
1998, Pattillo 1998, Pyrooz et al. 2013), and which helps those at a high risk of victimization avoid
such a fate. To gather these data, existing surveys and data collection efforts might be augmented
with questions that gather data on individuals’ broader social networks, as well as with questions
pertaining to gang involvement and behaviors.
Gang scholarship and violence reduction efforts can also be enhanced by expanding the scope
of the networks to include legal agents and institutions. Just as gang members are tied to non-
gang community members and institutions outside of the gang, so too are they connected to
legal institutions and agents such as police and probation officers. However, networks constructed
from co-arrest or field contact information are largely restricted to those who have been ar-
rested or contacted and neglect the role of those whose implementation of social control produces
the data that underlies these networks. Future work might investigate how officers’ arrest deci-
sions link them to gang members and other officers, and how the pattern of arrests by particular
officers affects the structure of gang and violence networks. Similarly, how gang members are
tied to probation and parole officers stands to enhance our understanding of how criminal jus-
tice monitoring is patterned within gang networks, and how the constellation of legal actors
within these networks exacerbates or ameliorates the negative outcomes that concentrate in these
networks.
Furthermore, technological advances that have increasingly blurred the boundary between the
real and online world create new challenges. Namely, as the ability to communicate instantaneously
and across space has been enhanced by the internet and social media, gang life increasingly plays
out online (Lane 2016, Moule et al. 2014, Pyrooz et al. 2015). Where once taunts and insults had
to be delivered face-to-face, today they are sent by way of smartphones and laptops, invectives
posted publicly as fast as one can type or upload a video (Patton et al. 2016, Sierra-Ar´
evalo &
Papachristos 2015a, Stuart 2016). That these disputes now evolve by way of Facebook posts, tweets,
or YouTube videos makes them no less dire—in fact, these technologies might vastly expand the
audience that is privy to these disputes, in turn creating social pressure that demands real-world
bloodshed.8
For those who hope to reduce gang violence, the fact that these online interactions are both
public and recorded provides a potential source of relational data. Researchers and practitioners
alike would do well to consider how they can use nearly real-time data on when, how often, and
with whom individuals interact online to better understand how group processes like contagion,
mobilization, and recruitment manifest digitally. Just as network models can be used to detect
natural disasters and disease outbreaks through Twitter (Ashktorab et al. 2014, Christakis &
Fowler 2010, Eubank et al. 2004), these methods can help diagnose how virtual gang conflicts
move offline, as well as provide ways for online disputes to be monitored and interrupted long
before shots are fired.
8See Collins (2007) for a discussion of how audiences influence the likelihood of arguments escalating to violence.
19.14 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
As promising as these sources of data are, there are real risks to blindly using data, be it from
Twitter or the local police department. Though the advent of Big Data and predictive algorithms
in criminal justice might help reduce bias in decisions otherwise made by fallible humans, it also
raises grave ethical and constitutional concerns. What are the implications of arresting someone
before a crime is committed, or enhancing a sentence because an algorithm spits out a result
indicating someone is more likely to reoffend upon release? How are we to reconcile the legal
standards of probable cause or reasonable suspicion with algorithms and data that are not publicly
available (Barry-Jester et al. 2015, Ferguson 2015)?
Though we are certainly not the ones to definitely answer these crucial questions, we do believe
that the potential costs of new data and methods must be considered carefully alongside their
potential benefits, in both gang violence reduction and the criminal justice system more broadly.
In particular, we argue that although networked approaches can support offender-centered efforts,
they are best suited to support approaches that are victim centered (Green et al. 2017). Network
analyses like those shown in Figure 2 are frequently used in other public health interventions that
deal with high-risk populations, such as intravenous drug users and sex workers (Luke & Harris
2007). Within such a public health framework, efforts to reduce gang violence should focus on
victimization risk and harm reduction, not on apprehension and confinement. Such a victim-
centered or risk reduction approach toward gang violence can reorient existing narratives around
how to address gang violence: Saving lives and lowering levels of gun violence means saving the
lives of young people who are often viewed only as offenders or criminals.
To be sure, no data set or analytic approach is without its weaknesses; when decisions can cost
lives, it is all the more important for these weaknesses to be discussed openly and honestly. Any
use of new data or network analysis that extends to gang violence reduction should be done as
transparently as possible—information on what data are being collected and how they are being
used should be a matter of public record, accessible to police, social workers, school teachers, and
community members alike. Accomplishing this necessary transparency will require careful consid-
eration and the forging of relationships among these varied interests, all of which are embedded
in unique demographic, historical, and political contexts. As trying as this process may prove, the
time to wrestle with these issues and to have these crucial discussions is now, before interventions
are implemented in the real world.
Finally, we urge that the human element of addressing gang violence not be abandoned in
its entirety. There is great value in the “experiential assets” of those that are most familiar with
the gang members and the streets they spend time on (Kennedy et al. 1997). As sophisticated
as SNA analyses have become, the nodes in every network graph will always represent people.
Generating a network graph, circling a dense cluster of nodes, and going out to make arrests
or provide outreach is ill advised, if not blatantly dangerous. Instead, we would do well to take
heed of the example set by gang scholars that recognized the value of direct observation of gang
members, as well as the knowledge of street outreach workers and police officers. Human rela-
tionships are far from static, and it would be folly to believe that any network picture can capture
the totality of rapidly shifting, dissolving, and emerging relationships on the street. It is only
by tempering our analyses with the expertise of teachers, street workers, and police that nodes
and ties can become the individuals and relationships that can then be targeted to reduce gang
violence.
DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.
www.annualreviews.org •Social Networks and Gangs 19.15
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
LITERATURE CITED
Allegro DB. 1989. Police tactics, drug trafficking, and gang violence: why the no-knock warrant is an idea
whose time has come. Notre Dame Law Rev. 64:552–70
Angst CM, Agarwal R, Sambamurthy V, Kelley K. 2010. Social contagion and information technology diffu-
sion: the adoption of electronic medical records in U.S. hospitals. Manag. Sci. 56(8):1219–41
Ashktorab Z, Brown C, Nandi M, Culotta A. 2014. Tweedr: mining Twitter to inform disaster response. Proc.
11th Int. ISCRAM Conf., pp. 1–5. University Park, PA: ISCRAM
Balko R. 2013. Rise of the Warrior Cop: The Militarization of America’s Police Forces. New York: PublicAffairs.
1st trade pap. ed.
Ball B, Newman MEJ. 2013. Friendship networks and social status. Netw. Sci. 1(1):16–30
Barkow RE. 2005. Federalism and the politics of sentencing. Columbia Law Rev. 105(4):1276–314
Barry-Jester AM, Casselman B, Goldstein D. 2015. The new science of sentencing. Marshall Project,
Aug. 4. https://www.themarshallproject.org/2015/08/04/the-new-science-of-sentencing
Bearman PS, Moody J, Stovel K. 2004. Chains of affection: the structure of adolescent romantic and sexual
networks. Am. J. Sociol. 110(1):44–91
Block CR, Block R. 1993. Street Gang Crime in Chicago. Washington, DC: Natl. Inst. Justice
Block R. 2000. Gang activity and overall levels of crime: a new mapping tool for defining areas of gang activity
using police records. J. Quant. Criminol. 16(3):369–83
Bonney LS. 1993. The prosecution of sophisticated urban street gangs: a proper application of RICO. Cathol.
Univ. Law Rev. 42:579–613
Bouchard M, Spindler A. 2010. Groups, gangs, and delinquency: Does organization matter? J. Crim. Justice
38(5):921–33
Boushey G. 2010. Policy Diffusion Dynamics in America. Cambridge, UK: Cambridge Univ. Press
Braga AA. 2003. Serious youth gun offenders and the epidemic of youth violence in Boston. J. Quant. Criminol.
19(1):33–54
Braga AA, Apel R, Welsh BC. 2013. The spillover effects of focused deterrence on gang violence. Eval. Rev.
37(3–4):314–42
Braga AA, Hureau DM, Papachristos AV. 2014. Deterring gang-involved gun violence: measuring the impact
of Boston’s Operation Ceasefire on street gang behavior. J. Quant. Criminol. 30(1):113–39
Braga AA, Hureau D, Winship C. 2008. Losing faith?: Police, black churches, and the resurgence of youth
violence in Boston. Symposium: legitimacy and criminal justice. Ohio State J. Crim. Law 6:141–72
Braga AA, Kennedy DM, Piehl AM, Waring EJ. 2001a. Reducing Gun Violence: The Boston Gun Project’s Operation
Ceasefire. Washington, DC: Natl. Inst. Justice
Braga AA, Kennedy DM, Waring EJ, Piehl AM. 2001b. Problem-oriented policing, deterrence, and youth
violence: an evaluation of Boston’s Operation Ceasefire. J. Res. Crime Delinq. 38:195–225
Braga AA, Papachristos AV, Hureau D. 2010. The concentration and stability of gun violence at micro places
in Boston, 1980–2008. J. Quant. Criminol. 26:33–53
Braga AA, Weisburd DL. 2015. Focused deterrence and the prevention of violent gun injuries: practice,
theoretical principles, and scientific evidence. Annu. Rev. Public Health 36(1):55–68
Brantingham PJ, Tita GE, Short MB, Reid SE. 2012. The ecology of gang territorial boundaries. Criminology
50(3):851–85
Bullmore E, Sporns O. 2009. Complex brain networks: graph theoretical analysis of structural and functional
systems. Nat.Rev.Neurosci.10(3):186–98
Bursik RJ, Webb J. 1982. Community change and patterns of delinquency. Am. J. Sociol. 88(1):24–42
Carrington PJ, Scott J, Wasserman S, eds. 2005. Models and Methods in Social Network Analysis. Cambridge,
UK: Cambridge Univ. Press
Chic. Crime Comm. 1995. Gangs: Public Enemy Number One: 75 Years of Fighting Crime in Chicagoland. Chicago:
Chic. Crime Comm.
Cho Y, Hwang J, Lee D. 2012. Identification of effective opinion leaders in the diffusion of technological
innovation: a social network approach. Technol. Forecast. Soc. Change 79(1):97–106
Christakis NA, Fowler JH. 2007. The spread of obesity in a large social network over 32 years. N. Engl. J.
Med. 357(4):370–79
19.16 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Christakis NA, Fowler JH. 2010. Social network sensors for early detection of contagious outbreaks. PLOS
ONE 5(9):e12948
City of Chicago v. Morales, 527 U.S. 41 (1999)
Cloward RA, Ohlin LE. 1960. Delinquency and Opportunity: A Theory of Delinquent Gangs. Glencoe, IL: Free
Collins R. 2007. Violence: A Micro-Sociological Theory. Princeton, NJ: Princeton Univ. Press
Cook PJ, Laub JH. 2002. After the epidemic: recent trends in youth violence in the United States. Crime
Justice 29:1–37
Coughlin BC, Venkatesh SA. 2003. The urban street gang after 1970. Annu. Rev. Sociol. 29(1):41–64
Crandall V, Wong S-L. 2012. Group violence reduction strategy: call-in preparation and execution. Practice Brief,
Natl. Netw. Safe Communities, Cent. Crime Prev. Control, John Jay Coll. Crim. Justice, New York
Curry GD, Decker SH, Egley AJ. 2002. Gang involvement and delinquency in a middle school population.
Justice Q. 19:275–92
Decker SH. 1996. Collective and normative features of gang violence. Justice Q. 13(2):243–64
Decker SH, Bynum T, Weisel D. 1998. A tale of two cities: gangs as organized crime groups. Justice Q.
15:395–426
Decker SH, Van Winkle B. 1996. Life in the Gang: Family, Friends, and Violence. New York: Cambridge Univ.
Press
Elliott M, Golub B, Jackson MO. 2014. Financial networks and contagion. Am. Econ. Rev. 104(10):3115–53
Eubank S, Guclu H, Kumar VSA, Marathe MV, Srinivasan A, et al. 2004. Modelling disease outbreaks in
realistic urban social networks. Nature 429(6988):180–84
Ferguson AG. 2015. Big data and predictive reasonable suspicion. Univ. Pa. Law Rev. 163(2):327–410
Ferrier GD, Reyes J, Zhu Z. 2016. Technology diffusion on the international trade network. J. Public Econ.
Theory 18(2):291–312
Fischer CS. 1982. To Dwell Among Friends: Personal Networks in Town and City. Chicago: Univ. Chicago Press
Fleisher MS. 1998. Dead End Kids: Gang Girls and the Boys They Know.Madison:Univ.Wisc.Press
Fox JA. 1996. Trends in Juvenile Violence: A Report to the United States Attorney General on Current and Future
Rates of Juvenile Offending. Washington, DC: US Dep. Justice, Bur. Justice Stat.
Fox JA, Zawitz MW. 2007. Homicide Trends in the United States. Washington, DC: US Dep. Justice, Off. Justice
Progr., Off. Juv. Justice Delinq. Prev.
Friedersdorf C. 2011. The war on drugs turns 40. The Atlantic, June 15. http://www.theatlantic.com/politics/
archive/2011/06/the-war-on-drugs-turns-40/240472/
Garland D. 2002. The Culture of Control: Crime and Social Order in Contemporary Society. Chicago: Univ. Chic.
Press
Garot R. 2010. Who You Claim: Performing Gang Identity in School and on the Streets. New York: NYU Press
Gerber EB. 1987. A RICO you can’t refuse: New York’s Organized Crime Control Act. Brooklyn Law Rev.
53:979–1006
Gibeaut J. 1998. Gang busters. ABA J. 84:64–69
Gilbert J. 1988. A Cycle of Outrage: America’s Reaction to the Juvenile Delinquent in the 1950s. New York: Oxford
Univ. Press
Gilens M. 2010. How the poor became black: the racialization of American poverty in the mass media. In Race
and the Politics of Welfare Reform, ed. SF Schram, JB Soss, RC Fording. Ann Arbor: Univ. Mich. Press
Green B, Horel T, Papachristos AV. 2017. Modeling contagion through social networks to explain and predict
gunshot violence in Chicago, 2006 to 2014. JAMA Intern. Med. 177(3):1–8
Hagedorn JM. 1998. Gang violence in the postindustrial era. Crime Justice 24:365–419
Hagedorn JM, Macon P. 1988. People and Folks: Gangs, Crime, and the Underclass in a Rustbelt City. Chicago:
Lakeview
Harvard Law Review. 1994. Juvenile curfews and gang violence: exiled on Main Street. Harvard Law Rev.
107(7):1693–710
Hinton E. 2016. From the War on Poverty to the War on Crime: The Making of Mass Incarceration in America.
Cambridge, MA: Harvard Univ. Press
Howell JC. 2012. Gangs in America’s Communities. Thousand Oaks, CA: SAGE
Huff CR. 1998. Comparing the criminal behavior of youth gangs and at-risk youth. Res. Brief, Natl. Inst. Justice,
Washington, DC
www.annualreviews.org •Social Networks and Gangs 19.17
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Hughes LA. 2013. Group cohesiveness, gang member prestige, and delinquency and violence in Chicago,
1959–1962. Criminology 51(4):795–832
Hughes LA. 2015. The legacy of James F. Short, Jr. In The Handbook of Gangs, ed. SH Decker, DC Pyrooz,
pp. 440–57. Hoboken, NJ: John Wiley & Sons
Kahler M. 2015. Networked Politics: Agency, Power, and Governance. Ithaca, NY: Cornell Univ. Press
Katz CM. 2001. The establishment of a police gang unit: an examination of organizational and environmental
factors. Criminology 39(1):37–74
Kelling GL, Wilson JQ. 1982. Broken windows: the police and neighborhood safety. The Atlantic, March.
http://www.theatlantic.com/magazine/archive/1982/03/broken-windows/304465/
Kennedy DM. 1997. Pulling levers: chronic offenders, high-crime settings, and a theory of prevention. Valpso.
Univ. Law Rev. 31:449–84
Kennedy DM. 2011. Don’t Shoot: One Man, a Street Fellowship, and the End of Violence in Inner-City America.
London: Bloomsbury
Kennedy DM, Braga AA, Piehl AM. 1997. The (un)known universe: mapping gangs and gang violence in
Boston. In Crime Mapping and Crime Prevention, ed. D Weisburd, T McEwan, pp. 219–62. New York:
Crim. Justice
Kennedy DM, Friedrich MA. 2014. Custom Notifications: Individualized Communication in the Group Violence
Intervention. Washington, DC: Off. Community-Oriented Polic. Serv.
Kennedy DM, Piehl AM, Braga AA. 1996. Youth violence in Boston: gun markets, serious youth offenders,
and a use-reduction strategy. Law Contemp. Probl. 59:147–96
Klein MW. 1968. Impressions of juvenile gang members. Adolescence 3(9):53–78
Klein MW. 1971. Street Gangs and Street Workers. Englewood Cliffs, NJ: Prentice-Hall
Klein MW. 1995. The American Street Gang: Its Nature, Prevalence, and Control. Oxford, UK: Oxford Univ.
Press
Klein MW. 2011. Comprehensive gang and violence reduction programs. Criminol. Public Policy 10(4):1037–44
Klein MW, Crawford LY. 1967. Groups, gangs, and cohesiveness. J. Res. Crime Delinq. 4(1):63–75
Klein MW, Maxson CL. 2006. Street Gang Patterns and Policies. Oxford, UK: Oxford Univ. Press
Kobrin S. 1959. The Chicago Area Project—a 25-year assessment. Ann. Am. Acad. Political Soc. Sci. 322(1):19–
29
Lane J. 2016. The digital street: an ethnographic study of networked street life in Harlem. Am. Behav. Sci.
60(1):43–58
Levitt SD, Venkatesh SA. 2000. An economic analysis of a drug-selling gang’s finances. Q. J. Econ. 115(3):755–
89
Ludeke BW. 2006. Malibu locals only: “Boys will be boys,” or dangerous street gang?—Why the criminal
justice system’s failure to properly identify suburban gangs hurt efforts to fight gangs. Calif. West. Law
Rev. 43:309–62
Luke DA, Harris JK. 2007. Network analysis in public health: history, methods, and applications. Annu. Rev.
Public Health 28(1):69–93
Maxson CL. 2015. The legacy of Malcolm W. Klein. In The Handbook of Gangs, ed. SH Decker, DC Pyrooz,
pp. 406–23. Hoboken, NJ: John Wiley & Sons
Maxson CL, Gordon MA, Klein MW. 1985. Differences between gang and nongang homicides. Criminology
23:209–22
Maxson CL, Hennigan KM, Sloane D. 2003. For the sake of the neighborhood? Civil gang injunctions as
a gang intervention tool in Southern California. In Policing Gangs and Youth Violence, ed. SH Decker,
pp. 239–66. Belmont, CA: Wadsworth
McFarland DA, Moody J, Diehl D, Smith JA, Thomas RJ. 2014. Network ecology and adolescent social
structure. Am. Sociol. Rev. 79(6):1088–121
McGloin JM. 2005. Policy and intervention considerations of a network analysis of street gangs. Criminol.
Public Policy 4(3):607–35
McGloin JM. 2007. The organizational structure of street gangs in Newark, New Jersey: a network analysis
methodology. J. Gang Res. 15(1):1–34
19.18 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
McGloin JM, Rowan Z. 2015. Street gangs and co-offending networks. In Disrupting Criminal Networks:
Network Analysis in Crime Prevention, Vol. 28, ed. G Bichler, AE Malm, pp. 9–26. Boulder, CO: Lynne
Rienner
Miller WB. 1957. The impact of a community group work program on delinquent corner groups. Soc. Serv.
Rev. 31(4):390–406
Miller WB. 1962. The impact of a “total-community” delinquency control project. Soc. Probl. 10(2):168–91
Monahan TP. 1961. On the trend in delinquency. Soc. Forces 40(2):158–68
Moore J. 1990. Gangs, drugs, and violence. In Drugs and Violence: Causes, Correlates, and Consequences,ed.M
De la Rosa, EY Lambert, B Gropper, pp. 160–76. Rockville, MD: Natl. Inst. Drug Abuse
Moreno JL. 1941. Foundations of sociometry: an introduction. Sociometry 4(1):15–35
Morselli C. 2009. Street gang presence in drug-distribution operations. In Inside Criminal Networks,Vol.8,
pp. 139–57. New York: Springer
Morselli C. 2010. Assessing vulnerable and strategic positions in a criminal network. J. Contemp. Crim. Justice
26(4):382–92
Moule RK. 2015. The legacy of Walter B. Miller. In The Handbook of Gangs, ed. SH Decker, DC Pyrooz,
pp. 458–77. Hoboken, NJ: John Wiley & Sons
Moule RK, Pyrooz DC, Decker SH. 2014. Internet adoption and online behaviour among American street
gangs: integrating gangs and organizational theory. Br.J.Criminol.54(6):1186–206
Natl. Netw. Safe Communities. 2013. Group Violence Intervention: An Implementation Guide. Washington, DC:
US Dep. Justice, Off. Community-Oriented Polic. Serv.
Needle JA, Stapleton WV. 1983. Police handling of youth gangs. Rep. Natl. Juv. Justice Assess. Cent., Am. Justice
Inst., Washington, DC
N.Y. City Youth Board. 1960. Reaching the Fighting Gang. New York: N.Y. City Youth Board
Papachristos AV. 2006. Social network analysis and gang research: theory and methods. In Studying Youth
Gangs, ed. JF Short, LA Hughes, pp. 99–116. Lanham, MD: AltaMira
Papachristos AV. 2009. Murder by structure: dominance relations and the social structure of gang homicide.
Am. J. Sociol. 115:74–128
Papachristos AV. 2011. The coming of a networked criminology? In Measuring Crime and Criminality: Advances
in Criminological Theory, Vol. 17, ed. J MacDonald, pp. 101–40. New Brunswick, NJ: Transaction
Papachristos AV, Braga AA, Hureau DM. 2012. Social networks and the risk of gunshot injury. J. Urban Health
89(6):992–1003
Papachristos AV, Braga AA, Piza E, Grossman LS. 2015a. The company you keep? The spillover effects of gang
membership on individual gunshot victimization in a co-offending network. Criminology 53(4):624–49
Papachristos AV, Hughes LA. 2015. Neighborhoods and street gangs. In The Handbook of Gangs,ed.SH
Decker, DC Pyrooz, pp. 98–117. Hoboken, NJ: John Wiley & Sons
Papachristos AV, Hureau DM, Braga AA. 2013. The corner and the crew: the influence of geography and
social networks on gang violence. Am. Sociol. Rev. 78:417–47
Papachristos AV, Kirk DS. 2015. Changing the street dynamic: evaluating Chicago’s group violence reduction
strategy. Criminol. Public Policy 14(3):525–58
Papachristos AV, Wildeman C, Roberto E. 2015b. Tragic, but not random: the social contagion of nonfatal
gunshot injuries. Soc. Sci. Med. 125:139–50
Pattillo ME. 1998. Sweet mothers and gangbangers: managing crime in a black middle-class neighborhood.
Soc. Forces 76(3):747–74
Patton DU, Lane J, Leonard P, Macbeth J, Smith-Lee JR. 2016. Gang violence on the digital street:
case study of a South Side Chicago gang member’s Twitter communication. New Media Soc.
doi:10.1177/1461444815625949
Pegram K, Brunson RK, Braga AA. 2016. The doors of the church are now open: black clergy, collective
efficacy, and neighborhood violence. City Community 15(3):289–314
Penley J. 2011. Urban terrorists: addressing Chicago’s losing battle with gang violence. DePaul Law Rev.
61:1185–212
Petersen RD, Howell JC. 2013. Program approaches for girls in gangs: Female specific or gender neutral?
Crim. Justice Rev. 38(4):491–509
www.annualreviews.org •Social Networks and Gangs 19.19
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Peterson D, Taylor TJ, Esbensen F-A. 2004. Gang membership and violent victimization. Justice Q. 21:793
Pettit B, Western B. 2004. Mass imprisonment and the life course: race and class inequality in U.S. incarcer-
ation. Am. Sociol. Rev. 69(2):151–69
Pyrooz DC, Decker SH, Moule RK Jr. 2015. Criminal and routine activities in online settings: gangs,
offenders, and the Internet. Justice Q. 32(3):471–99
Pyrooz DC, Moule RK, Decker SH. 2014. The contribution of gang membership to the victim-offender
overlap. J. Res. Crime Delinq. 51(3):315–48
Pyrooz DC, Sweeten G, Piquero AR. 2013. Continuity and change in gang membership and gang embedded-
ness. J. Res. Crime Delinq. 50(2):239–71
Rosen E, Venkatesh S. 2007. Legal innovation and the control of gang behavior. Annu. Rev. Law Soc. Sci.
3(1):255–70
Schneider EC. 2001. Vampires, Dragons, and Egyptian Kings: Youth Gangs in Postwar New York.Princeton,NJ:
Princeton Univ. Press
Scott J, Carrington PJ, eds. 2011. The SAGE Handbook of Social Network Analysis. London: SAGE
Short JF, Strodtbeck FL. 1963. The response of gang leaders to status threats: an observation on group process
and delinquent behavior. Am. J. Sociol. 68(5):571–79
Short JF, Strodtbeck FL. 1965. Group Process and Gang Delinquency. Chicago: Univ. Chic. Press
Sierra-Ar´
evalo M, Charette Y, Papachristos AV. 2016. Evaluating the effect of project longevity on group-
involved shootings and homicides in New Haven, Connecticut. Crime Delinq. 63(4):446–67
Sierra-Ar´
evalo M, Papachristos AV. 2015a. Social network analysis and gangs. In The Handbook of Gangs,ed.
SH Decker, DC Pyrooz, pp. 157–77. West Sussex, UK: John Wiley & Sons
Sierra-Ar´
evalo M, Papachristos AV. 2015b. Applying group audits to problem-oriented policing. In Disrupting
Criminal Networks: Network Analysis in Crime Prevention, ed. G Bichler, A Malm, pp. 27–46. Boulder, CO:
Lynne Rienner
Slutkin G. 2013. Violence Is a Contagious Disease. Chicago: Cure Violence
Spergel IA. 1966. Street Gang Work: Theory and Practice. Reading, MA: Addison-Wesley
Spergel IA. 1995. The Youth Gang Problem: A Community Approach. New York: Oxford Univ. Press
Spergel IA. 2007. Reducing Youth Gang Violence: The Little Village Gang Project in Chicago. Lanham, MD:
AltaMira
Spergel IA, Grossman SF. 1997. The little village project: a community approach to the gang problem. Soc.
Work 42(5):456–70
Spindler A, Bouchard M. 2011. Structure or behavior? Revisiting gang typologies. Int. Crim. Justice Rev.
21(3):263–82
Stuart F. 2016. Dispatches from the rap wars: my 18 months inside one of Chicago’s most notori-
ous gangs. Chicago Magazine, Sept. 19. http://www.chicagomag.com/Chicago-Magazine/October-
2016/Chicago-Gangs/
Suttles GD. 1968. The Social Order of the Slum: Ethnicity and Territory in the Inner City. Chicago: Univ. Chic.
Press
Teeters NK, Matza D. 1959. The extent of delinquency in the United States. J. Negro Educ. 28(3):200–13
Thrasher FM. 2013 (1927). The Gang: A Study of 1,313 Gangs in Chicago. Chicago: Univ. Chic. Press
Tita GE, Cohen J, Engberg J. 2005. An ecological study of the location of gang “set space.” Soc. Probl.
52(2):272–99
Tita GE, Greenbaum RT. 2009. Crime, neighborhoods, and units of analysis: putting space in its place. In
Putting Crime in Its Place, ed. D Weisburd, W Bernasco, GJN Bruinsma, pp. 145–70. New York: Springer
Tita GE, Radil SM. 2011. Spatializing the social networks of gangs to explore patterns of violence. J. Quant.
Criminol. 27(4):521–45
Tita GE, Riley KJ, Ridgeway G, Grammich CA, Abrahamse A. 2010. Reducing Gun Violence: Results from an
Intervention in East Los Angeles. Santa Monica, CA: RAND Corp.
Venkatesh SA. 1997. The social organization of street gang activity in an urban ghetto. Am. J. Sociol. 103(1):82–
111
Wakefield S, Wildeman C. 2016. Children of the Prison Boom: Mass Incarceration and the Future of American
Inequality. New York: Oxford Univ. Press
19.20 Sierra-Ar´
evalo ·Papachristos
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.

LS13CH19-Papachristos ARI 14 July 2017 10:7
Wasserman S, Faust K. 1994. Social Network Analysis Methods and Applications. Cambridge, UK: Cambridge
Univ. Press
Wertham F. 1954. Seduction of the Innocent: The Influence of Comic Books on Today’s Youth. New York: Rinehard
&Co.
Western B, Wildeman C. 2009. The black family and mass incarceration. Ann. Am. Acad. Political Soc. Sci.
621(1):221–42
White D, Johansen U. 2005. Network Analysis and Ethnographic Problems: Process Models of a Turkish Nomad
Clan. Lanham, MD: Lexington Books
Whyte WF. 1969 (1943). Street Corner Society. Chicago: Univ. Chic. Press
Wilson WJ. 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago: Univ.
Chic. Press
Yoo CS. 1994. The constitutionality of enjoining criminal street gangs as public nuisances. Northwest. Univ.
Law Rev. 89:212–67
www.annualreviews.org •Social Networks and Gangs 19.21
Changes may still occur before final publication
Annu. Rev. Law. Soc. Sci. 2017.13. Downloaded from www.annualreviews.org
Access provided by Yale University - Medical Library on 09/12/17. For personal use only.