Gangs, Methodology and Ethical Protocols: Ethnographic Challenges in Researching Youth Street Groups

Gangs have been described as an episodic phenomenon comparable across diverse geographical sites, with the US gang stereotype often acting as the archetype. Mirroring this trend, academic researchers have increasingly sought to survey the global topography of gangs through positivist methodologies that seek out universal characteristics of gangs in different cultural contexts. So, research about youth street groups requires an innovative methodological approach to develop a renewed approach for the twenty-first century’s youth street groups, different from the local, coetaneous, male and face-to-face model, used to understand the twentieth century’s gangs. How can complex social forms such as street gangs be researched in the twenty-first century? Can a single ethnographic approach be shared by researchers working in entirely different cultural contexts? What novel methodological and ethical challenges emerge from such a task and how might they be resolved? This article examines the methodological perspectives of the TRANSGANG project.


Introduction
Gangs 1 have been described as an episodic phenomenon comparable across diverse geographical sites. Academic researchers have increasingly sought to survey the global topography of gangs in order to define the "universal characteristics" of groups that operate in different cultural contexts (Klein 1971;Miller 1992;Esbensen and Maxson 2012). The use of quantitative data and positivist methodologies has tended to result in rather Eurocentric accounts in which the "US gang stereotype" acts as a kind of global "gang archetype". 2 In contrast, ethnographic work has revealed that contemporary gang formations diverge significantly from this normative model. Modern gangs are not strictly territorial, nor do they have compact structures. Instead, gangs today are structurally fluid, have significant geographic mobility and, due to patterns of human migration and globalization, organise and have a strong presence on social media (Reguillo 1995;Brotherton and Barrios 2004;Perea 2007). Gang identities in the global era are best understood as hybrid clusters of elements taken from the respective countries of origin of gang members; they are nomadic identities that, just like other contemporary "youth cultures", appropriate and reproduce styles and trends as they circulate around the globe (Nilan and Feixa 2006). This paper presents preliminary findings from a large-scale, multi-sited ethnographic study (Marcus 1995) of transnational youth gangs in 12 different cities around the globe (Barcelona, Madrid, Marseille and Milan in Southern Europe; Casablanca, Tunis, Algiers and Djendel in Northern Africa; Medellin, San Salvador, Santiago de Cuba and Chicago in the Americas). The study began in 2018 and is due to finish in 2022. This paper explores the methodological and ethical challenges of developing an ethnographic project on such a large scale and with such a high degree of cultural difference between the field sites and communities of study. The project is based on an experimental approach that combines "extended case method" (Burawoy 2009) with "relational ethnography" (Desmond 2010) and departs from the twentieth century's model of studying youth street groups (a model that privileged local, coetaneous, male and face-to-face gangs). Our aim is to situate the experiences of youths at the centre of the project, unveiling the positive aspects of youth street sociability and how marginalized position within social structure is resisted and remade as a consequence. Some research focuses on proactive experiences in gang behaviour and policies (Leinfelt and Rostami 2011;Venkatesh 2009), but very few studies systematically compare such aspects in order to find variants and invariants in the evolution or in the reversal of the criminal gang model, use a transnational comparative methodology or focus on a group rarely included in gang studies (Young Arabs) along with another overstudied group (Young Latinos). Both groups face big challenges regarding new generations in their homelands and in their diasporic new land where their collective forms of behaviour have been seen as barriers to their social inclusion. Our standpoint combines postsubcultural studies and decolonial theoretical perspectives with critical criminology focusing on challenging traditional understandings and uncovering false beliefs about youth street groups. As a result, the combination of these viewpoints facilitates look at the field within the social structure of class and status inequalities and considers law and punishment of crime as connected to a system of social inequality and as the means of producing and perpetuating this inequality. Beyond this, we highlight the difficulties to apply the traditional criminological perspective in several different cultural contexts. Accordingly, the third section presents the process of elaboration of a transnational operational definition of "gang" in several academic discussions among all the local researchers of the project. Moreover, the discussion is oriented to a key point marked by several authors in the conceptualization of the meaning of gang: the question of labelling. How can such complex contemporary social formations be researched? Can a singular methodological approach be applied across very different cultural contexts? What new methodological and ethical challenges emerge from such a task and how might they be resolved? This article examines the methodological perspectives of the TRANSGANG project. 3 The paper focus in particular on our attempts to accommodate cultural difference in the research design and what happened as we put on plans in practice. The first section of the paper shows the theoretical perspective adopted contrasting with current literature on gangs and youth street groups, highlighting the ethnocentric tendency to use experiences from the Americas as a heuristic for experiences elsewhere in the world. We then set our how our plans for fieldwork sought to overcome this limitation by framing our understanding of gangs and youth street groups through debates about post-colonialism and critical subcultural studies. In particular, we focus on the definition of youth street groups in a transnational perspective and the need for out working definitions to avoid trite Western centrist viewpoints and the coloniality of knowledge (Mignolo 2010). 4 In "Methodological Perspectives on Gang Research in the Twenty-First Century", we explore what happened in the first stage of fieldwork and the advantages and limitations of our approach. In "Methodology in Motion: Defining "Gang Field"", we discuss our unique approach to research ethics and safety, our attempts to tailor the highly bureaucratic ethical guidelines of the EU's European Research Council for use around the globe and some of the ethical issues that have emerged from fieldwork thus far. The article is closed by a conclusion section summarizing the main conclusions of the different sections and its general implications for youth street group research.

Methodological Perspectives on Gang Research in the Twenty-First Century
As it is mentioned, our methodological perspective aims to reverse the traditional criminological approach. If we analyse these groups, in the first place, we find that these adolescents and young adults have a feeling of union and group belonging in a structure of sociability that resembles a second family. The use of the word hermanito (brother) by Latino groups shows the dimension of fraternity in the organization, whose main objective is not to commit crimes, but to offer solidarity by sharing their difficult daily life in terms of protection, identity construction and feelings of affection (Brotherton and Barrios 2004;Nilan and Feixa 2006;Feixa et al. 2008). As it is evidenced in previous research, gangs are diverse in ethnic composition, criminal (or not) activities, age of members, propensity towards violence and stable organization (Feixa et al. 2008;Feixa et al. 2019). Gangs experience changes due to direct factors and indirect factors, such as demographic shifts, economic conditions or the influence of the media, and their reactions vary according to community understanding, representation and policies; effective responses are diverse too: prevention, intervention and suppression or enforcement. A decade later, this situation is still "in process" in the Latin diaspora and the same (in) definition affects the Arabic and Muslim diasporic youth worlds (Camozzi et al. 2014;Feixa and Romaní 2014;Queirolo Palmas 2014). The challenge is to build a framework focused in the background of personal and social narratives, subjectivities and identities of group members. Our proposal is a unique mixture of views coming from subcultural and post-subcultural youth studies combined with a decolonial perspective that applies intersectional frame analyses.
The first tradition is rooted in subcultural studies elaborated at the Centre for Contemporary Cultural Studies (CCCS) at the University of Birmingham in the second half of the 1970s and the early 1980s. This university produced a series of influential works on youth and popular culture in the British context that opened up this new theoretical perspective from empirical research. One of the most insightful elements of the Birmingham school's approach is its aim to take youth subcultures seriously and on their own terms, without dismissing them as ephemeral expressions of non-conformism youth or as forms of "juvenile deviance" like most of the previous studies on youth cultural practices and behaviours. As noted by Griffin: "The youth subculture project treated (…) working class youth cultural practices as imbued with meaning and political significance, as worthy to be studied in their own terms, and as potentially creative rather than inherently destructive or of minimal cultural value" (Griffin 2011, p. 4). It provided youth research with crucial keys for understanding contemporary youth cultural practices such as street sociability; it is articulated more around expressive behaviour and less around direct and explicit political commitment. So, a comprehensive understanding of youth cultural worlds and production, drawing on ethnographic methods and on semiotic analytical tool, is essential for understanding the mediation processes at the heart of gangs.
However, we consider that there are several gaps in subcultural studies that need to be covered to refine this perspective according to our objectives. First, the attention paid to gender, sexual, ethnic and geographical differences among young people will be studied through an intersectional analysis (Yuval-Davis 2006). Moreover, we consider that youth subcultures are not clearly delimitated entities, but rather entities with blurred limits and crossbreed cultural references. Because the analysis of youth street groups' cultural practices is no longer confined to spectacular styles, it rather tends to encompass the everyday life experiences and cultural practices of members or ex-members of youth groups. Consequently, the consideration of the limitations of the Birmingham conceptual framework and the new tendencies in conceiving youth cultural practices constitute the core elements of our "post-subcultural" approach (Bennett 1999;Bennett and Kahn-Harris 2004;Hodkinson and Diecke 2007).
On the other hand, this perspective considers identity construction as key variable in the research. Therefore, we refer to identifications rather than understand identity as a finished thing, as something in continuous construction and strategically negotiated. In recent descriptions of the identity creation processes among youth groups in the West, the treatment of the body (its construction, its treatment, its restructuring, deconstruction), the influence of an alleged global culture centred on the creation of transnational communities and the influence of music, specifically pop, rock, rap and local hybrid scenes, have emerged as major axes for young populations. These cultural elements are setting the primary reference markers for identity negotiation that some authors reflect upon in relation to global youth. Their importance to youth street groups is determined essentially by influencing choices, they invent new ways of understanding the body and diversifying transnational relations and the possibility of participating in solidarity groups related to similar cultural practices and identification artefacts coming from specific backgrounds differentiated from Western traditions. So, the identitarian processes emerge in an interface where, in addition to the hegemonic host culture and the traditional culture of their parents, several other subcultural traditions come together (Brotherton and Barrios 2004;Feixa and López 2015;Klein and Maxson 2006;Matza 1961;Venkatesh 2009). We can define five basic axes according to our subject, which are used as identification sources of cultural devices: (1) North American street gang tradition (Klein 1995;Thrasher 1927Thrasher /2013Whyte 1943); Latin American gang traditions: pandillas and naciones (Feixa 1998;Perea 2007;Ramoset al. 2013;Reguillo 2000); Arab youth subcultural traditions (Bayat 2012;Camozzi et al. 2014;Sánchez García 2016, 2019; European subcultural traditions (Esbensen and Maxson 2012;Klein et al. 2001;Leccardi 2016;Queirolo Palmas 2016;van Gemert et al. 2008); and finally, virtual global tradition represented by youth identity models that circulate through the Internet (Fernández Planells et al. 2020).
Beyond this, post-subcultural studies meet critical criminology challenging traditional understandings about gangs. This perspective examines the gang field within the social structure of class and status inequalities and considers law and punishment of crime as connected to a system of social inequality and as the means of producing and perpetuating this inequality. As a result, crime is seen as a product of oppression of subaltern groups within society, such as women and ethnic minorities. According to Brotherton (2015) to research gangs as subaltern groups, it is necessary to have a critical anti-colonial ethnography, as youth members have "little option but to resist this relationship of domination" (Brotherton 2015, p. 80). This position emphasizes the creative and agency capacity of the gang members, their cultural productions and their forms of sociability as resistance practices, of course contradictory and ambiguous, against a set of discrimination processes in relation to culture, class, race and ethnicity. On the one hand, these groupings are seen as places of production and social transformation; on the other hand, the reproduction dynamics are also evident, that is, the homologies between their functioning logics and their symbolisms (masculinity, strength, authority, hierarchy) and the global functioning of society.
Finally, post-colonial studies, critical criminology links with decolonial epistemologies introducing the concept of border as both a symbolic and a physical space joining gang members' perspectives with stakeholders' and academic studies to produce border thinking in gangs. 5 In short, border thinking is a tool that allows us to discard Western conceptions and seek to accumulate other visions of the world that have been previously dismissed as invalid or backward. Border thinking arises in those populations that neither want to accept the humiliation of being relegated to an inferior position nor assimilate the imposed model. It is in these border spaces that other possible ways of seeing arise, which do seek not only to get rid of what is imposed but also to empower other ways of thinking, being and living (Mignolo 2010). With all this, we understand the subjects and groups studied as agents (with their own agency) that negotiate their situation in migrant societies and that, in that displacement (physical or social), adapt varied cosmovisions that are situated in what we will call the border.
Beyond studying gangs, we are interpreting social processes with blurred boundaries in different locations with very different social, political and economic conditions. So, more than construct "subjects" of research in an inductive manner, the objective is to follow configurations of relations. The focus of fieldwork becomes to describe a system of relations, "to show how things hang together in a web of mutual influence or support or interdependence or what have you, to describe the connections between the specifics the ethnographer knows by virtue of being there" (Becker 1996, p. 56). Thus, more than to construct "subjects", "youth street groups" or "gangs" in an inductive or deductive manner, the objective is to follow configurations of relations. Methodology constructs the pitch, in our case, "youth street group micro-cosmos" encompassing those agents that are part of it (state, academia, media, the gang, themselves among others) to understand how this field works what positions each of these agents occupies (although positions are variable) and see what dynamics are generated.
In summary, to research youth street groups as subaltern groups, it is necessary to have a critical de-colonial ethnography, as youth members have "little option but to resist this relationship of domination" (Brotherton 2015, p. 80). The final objective is not to know "what it is" but "what it could be". This perspective emphasizes the creative and agency capacity of the gang members, their cultural productions and their forms of sociability as resistance practices, of course contradictory and ambiguous, against a set of discrimination processes in relation to culture, class, race and ethnicity. On the one hand, these groupings are seen as places of production and social transformation; on the other hand, the reproduction dynamics are also evident, that is, the homologies between their functioning logics and their symbolisms (masculinity, strength, authority, hierarchy) and the global functioning of society.
Methodology in Motion: Defining "Gang Field" However, the term gang exists, although it is usually associated mainly with organized crime, delinquency and illegal businesses with leadership and hierarchies similar to gangs in North America; terms and meanings may vary according to geographic locations and subcultural traditions, but using the concept, the danger of stigmatization of although all youth street practices is on the table.
From an "emic" point of view, in the three regions in which our study will be carried out, the use of the term is far from homogeneous. In Europe (as in the USA), the term "gang" tends to have a pejorative sense associated with crime, so it is juxtaposed with other terms of local use. This is the case in Marseille, where youth street groups are considered actors of the small banditry (drug trafficking, deal), while gangs "play in the major leagues" (they have guns and appear once or twice a year on television). Students between 12 and 16 years old use the term gang-to simply refer to their group of friends, showing that young people can call themselves gangs, despite their involvement in criminal activities (Mansilla, forthcoming). In other contexts, as in peripheral neighbourhoods of Paris, young people perceive the term gang as an extrastigmatization to the fact that their neighbourhoods are marginalized and prefer to call themselves team, clique or crou (Moignard 2007). In Spain, the term gang-banda in Spanish-evokes the tradition of banditry of ancient origin and opposes the term pandilla, which does not have criminal connotations but refers to a group of peers. In Italy, when maras appeared in Milan, newspapers started to write about gangs and pandillas to distinguish them from local youth street groups.
In Latin America, there are a lot of local terms to name youth street groups: gangas, clicas and vatos on the border between Mexico and the USA, chavos banda in Mexico, maras in Central America, combos, parches and galladas in Colombia, coros in the Dominican Republic, pibes choros in Argentina, etc. In Cuba, despite the ignorance of the government about the phenomena, the term banda is associated with a musical group, while the term pandilla designates a criminal group.
By his side, in Standard Arabic, the term gang is ignored by its colonial background. The general term used to refer to "criminal youth groups" is iṣhāba while the term shila is used to designate a youth street group. However, there are other related terms coming from the national and local contexts and expressed in colloquial Arabic, such as hittistes (Algeria), tcharmils (Morocco) and baltagiyya (Egypt), which designate different criminalized street groups from paramilitaries to organized drug clans.
On the other hand, each youth group can use different categories to define itself. In Barcelona, the Latin Kings define themselves as a "nation" or "organization", while the Ñetas define themselves as an "association". In San Salvador, the Salvatrucha is a mara while the 18 is a pandilla or a barrio. In the case of the North African region, young people do not use a specific name, but rather are identified with the neighbourhood. In addition, some youth street groups propose using the term "street family" to avoid the term "gang" and to denote the horizontal fraternity and vertical authority relationships that occur among them.
So, how can we be sure that we will give the same meaning to the key concepts? This section examines two main methodological challenges that emerged during the operationalization discussions with different local researcher teams of the central concept of the project: what does "gang" mean? 6 According to Thrasher's definition, a gang is "an interstitial group originally formed spontaneously, and then integrated through conflict" (Thrasher 1927(Thrasher /2013. Thrasher also points out that gangs, as forms of sociability, are characterized by a behaviour guided by face-to-face encounters, fights, urban spatial movement as a unit, conflicts with other agents and the planning of their actions. Thus, "the result of this collective behaviour is the development of a tradition, unreflective internal structure, esprit-de-corps, solidarity, morale, group awareness, and attachment to a local territory" (ibid.). In line with this approach, a gang can be characterized as an informal group of peers who are attached to a territory, in conflict with other peer groups and sometimes with adult institutions, an obsolete definition.
Although crime is not the main reason why gangs form, the police and political approaches in the USA have reinforced the criminal conceptualisation of gangs. When delinquency is not considered as a fundamental attribute of youth street behaviour, other concepts are used, such as peer groups, street groups, subcultures, countercultures or lifestyles, among others, and the term gang is reserved for youth street groups with members from mainly minority, migrant or ethnic backgrounds. However, the criminological tradition has tended to use the term gang as a synonym of youth street group more or less linked to criminal activities. Consequently, offering a gang definition with which all social actors (gang members, researchers, social workers, institutions, among others) can agree has always been a difficult challenge. Hence, during the twenty-first century, scholars have faced different challenges and have provided different approaches when trying to offer a conceptualisation of gangs.
First, the way in which we define "youth gang" determine the number and composition of what it is that we are talking about regarding the conceptualization of the term. In this question, two kinds of approaches can be found: (1) those offering wide definitions that gather more young people into the gangs' conceptual net; and, on the other side and (2) those offering narrow definitions that are more exclusive conceptualizations that include fewer young people in gangs. Should more youth groups be integrated in the conceptual field defined as gangs or should the definition be narrowed to include only those groups engaged in illegal activities? Choosing a narrow approach usually involves focusing more on the illegal activities of the group and, consequently, being a member of a gang is seen as a criminalized behaviour. This perspective is represented by academic researchers who apply Klein's definition, developed in the 6 The different meanings of "gang" emerge in two different seminars. The first in the kick off Meeting of the project celebrated in Barcelona in October of 2018 where,all the local researchers engaged in the project and significant scholars in the field of "gang" studies participated and gave consensus to the operative definition of "gang". The second one is a seminar with all the Spain team researchers where we discussed the operativa final definition. seventies in Los Angeles: "a gang is a group of young people that can be identified by: a) being perceived as an aggregation different from the others in the neighbourhood, b) recognizing themselves as a defined group, c) being involved in various criminal episodes that generate a constant negative reaction of the neighbours and/or of the services in charge of the application of the law" (Klein 1971, p. 13). In this direction, the Eurogang network of researchers has defined a gang as "a street gang (or troublesome youth group) is any durable, street-oriented youth group whose involvement in illegal activity is part of their group identity" (Esbensen and Maxson 2012, p. 5). These broad definitions focus the core criteria on durability, street-orientation, youth, identity and, most importantly, illegal activity.
The second challenge is about the naming process associated with "youth gangs" research. Although the process of assigning characteristics to gang groups helps to determine how conflicts and social problems are framed, if we focus on gangs only as a "social problem", we ignore fundamental structural issues like racism, poverty and social inequality. Scholars who fail to capture the fluidity and contradiction inherent in gang identification create an artificial sense of similarity between diverse cultural contexts. The image of "the gang" as socially dangerous or damaging prevents gangs from developing into pro-social organizations or more organized criminal entities, often leading to intervention by state agents of control. This is the case in Algeria, where the term gang or its French version bande is ignored for colonial reasons; the researchers do not accept using the label gangs to youth street groups because the remembrances of the concept are very far from the realities of street young people selling gold or looking for a precarious job walking the streets and prevent criminalization and stigmatization.
Another question arising from the conceptualisation of gangs is related to the problem of how to study collective behaviour and group commitments while integrating personal experience and individual behaviour as well. The gang can be understood as an analysis frame about group status and relationships with other social subjects as individuals, criminalizing all the members. Here the focus is on collective behaviour and group engagements, and the personal experience is ignored. A good example of this dichotomy is Miller's definition: "A self-formed association of peers, united by mutual interests, with identifiable leadership and internal organization, who act collectively or as individuals to achieve specific purposes, including the conduct of illegal activity and control of a particular territory, facility, or enterprise" (Miller 1992, p. 21).
The last critical challenge we would like to point out when trying to offer a gang definition is how to integrate and emphasize the creative and agency capacities of members of youth street organizations. The definitions that line up into this issue take gang's cultural productions and forms of sociability as resistance practices, contradictory and ambiguous, against a set of discrimination processes by culture, class, race and ethnicity. In this sense, Queirolo Palmas (2014, p. 23) define gangs as "urban youth groups that take shape in the interstices of a post-migration society, with their cultural practices and sometimes cooperative interactions that are sometimes conflicting, and which are designated by the thinking of the institutions and the media as gangs, a signifier associated with violence, crime and social danger". A definition that attempts to collect all of these attributes is that offered by Brotherton and Barrios (2004, p. 23): "groups formed in large part by young people and adults from marginalized classes, whose objective is to provide their members with an identity of resistance, an opportunity for empowerment both individually and collectively, of a possible 'voice' capable of challenging the dominant culture, of a refuge with respect to the tensions and sufferings of daily life in the ghetto and, finally, of a spiritual enclave in which practices and rituals considered sacred can be developed". In this perspective, we find the Latin American tradition of gang studies understanding gangs as social formations that attempt to build a cultural citizenship from the margins.
Based on the evidence established from ethnographic research in diasporic situations, as in the case of the Latin Kings in Barcelona (Nilan and Feixa 2006), the definition focused on group remarks the identitarian capacity of the crowd and describes it as street-oriented youth groups, with names, symbols and long-time traditions, composed by youths from deprived social backgrounds. Some of their members have connections with illegal activities, even if these activities are not part of the core group identity (Feixa et al. 2019). Adding the society-network context and the potential role of gang members as mediators to the Thrasher's classic definition, we propose to use the generic term "youth street group" to refer to any gathering of young people, according to the definition of youth that exists in each context, who recognize themselves as a group and who use the public space, physical or virtual, to meet and find ways to be respected.
Thus, our perspective challenges the traditional criminological perspective on "gang", considering the youth street groups not as a sole model but as a "continuum". At one extreme, we would find, always ideally, the classic gangs based on illegal activities and not only formed by young people-like the bacrim in Colombia, the maras in El Salvador, the tcharmil in Marocco and the quinquis in Spain. At the other extreme, we would find youth subcultures based on leisure and economic activities-like the vatos locos in the Mexican American border, the rappers in north Africa and the tribus urbanas in Spain. And in the middle, there are a variety of hybrid groups that combine both strategies-like the naciones in Latin America, the hittistes in North Africa and the bandas latinas in Spain. The proposed conceptualization and operationalization make it possible to differentiate street gangs not only from organized crime or from transnational criminal organizations, including terrorist cells, but also from informal groups without stable organization, grouped exclusively around leisure. In short, we consider a gang as a dynamic cultural formation in a context of exclusion and social transformation. Youth street groups can both evolve towards more associative, cultural or sports forms and specialize in some kind of crime.

Ethics and Ethnography
In each of the project's twelve field sites, local researchers are working with young, materially impoverished and social marginalised people who were possibly involved in criminality and may have little access to familial and social support. With a research topic of such ethical sensitivity, it was essential for us to develop project protocols that would ensure the interests of all researchers and research participants were safeguarded at all times. It was agreed with the European Research Council (ERC), the projects funder, that an internal Ethics Advisory Board (EAB) would be appointed to advise on the ongoing conduct of the project and that a post-doctoral researcher with experience of managing research ethics in the context of ethnographic research would be recruited.
Formalised research ethics review processes have been the subject to several bruising critiques from anthropologists and sociologists for their inherent Eurocentrism and general inapplicability to ethnographic research. In our project, the greatest difficulty was ensuring that we abided by our ethical agreement with the ERC while not blindly following a set of standards that were inappropriate for our subject of study and chosen methodology. We therefore worked to the principle that our agreement with the ERC was necessary but not sufficient to ensure the safe and ethical conduct of the research, or put differently, we sought to go beyond the agreement we made with the ERC by adapting our protocols to the ongoing, open-ended and culturally contingent nature of the ethnographic research process. In May 2019, work began on a "handbook" on how to conduct the aspects of the research that required special ethical consideration. The handbook was intended to be a reference document for fieldworkers and was based on broadly accepted ethics principles and norms of sociology and anthropology (ASA 2011), our agreement with the ERC and input and feedback from the local researchers and ERB. Before beginning fieldwork, each local researcher received training about how to use the handbook and conduct data collection activities in a manner that the research team considered safe and ethical.
Among the areas that required most adaptation was the process for gaining consent from research participants. Other than standard practices such as translating all consent documents into Spanish, English, Italian, French and Arabic to ensure that all research participants could understand the agreement they were signing, special consideration was given to the lower age limit of consent. The handbook states that laws and local customs and norms regarding the age of consent for participation in research must be respected by the researchers. However, in several of the locations where the research is based (including Spain, France and Italy), there exists no clear regulation regarding the age at which people can legally give consent for participation in research. 7 Moreover, the age at which people are considered mature adults is culturally contingent and varies in relation to class, gender, nationality and so on. In Algeria, for example, children are considered legally mature at 21. However, it is not clear whether legal maturity is required to give consent to research participation.
We decided that in countries where no law concerning research participation existed, only people over 14 years of age would be invited to participate in the research. This decision was based on the judgement that potential participants who were over the age of 14 were likely be able to (1) understand the information relevant to participating in the research, (2) understand the consequences of participating in the research and (3) give informed consent to participate in the research. Local researchers were instructed to exclude 14-year-olds who they believed lacked these capacities.
A second related point concerned parental permission. As potential participants over the age of 14 were considered able to give informed consent for themselves, we decided not to make the consent of parents and/ or legal guardians a mandatory precondition for participation. However, the handbook stated that under normal circumstances, parents and legal guardians should be informed by the local researchers about their child's or ward's participation in the project. This presented some difficulties, however. The team was concerned that some children may be put at risk of harm if their patents or legal guardians were informed about their association with a "gang". Local researchers were therefore advised to only inform a participant's legal guardians or parents about their child's or ward's participation in the project if they (1) had the participant's permission and (2) believed that it was appropriate and safe to do so. If the local researchers judged it to be unsafe, they were advised not to tell the parents and to record their reasoning for not doing so in their next bimonthly ethics report. Our evocation of the researcher's judgement was consciously intended as safeguard against ethnocentrism of the ethical principles on which ethical guidelines are based. The local researchers were recruited because of their local knowledge and expertise, and it was by positioning them as the principle decision maker that we hoped to avoid inappropriately applying our measures and definitions of safety.
A similar tension lay at the heart of the decisions we made regarding the process for recording consent. Among the trickiest aspect of the handbook that the team developed was the process for recording consent. The team was aware that gang members and exmembers may be reluctant to sign consent forms for fear of such documents being used against them in legal proceedings. It was decided that while written consent was ideal, in some circumstances, two alternative processes for recording consent could be used. The first was oral consent recorded via the researcher's audio recorder. The second was a process called "single-party testimonial consent". This is where the researcher writes a short description of the fact that the participant has given consent and explains their reasons for not recording consent via more conventional means. The local researchers were advised that if they used an alternative means of consent, they must include a description of their reasoning in their bimonthly ethics report for fieldworkers.
In Algeria, during an ethnographic visit, two local researchers arranged a meeting with a gold seller in a peripheral city about ten kilometres from Algiers. When the local researchers arrived to the informal market, they were met with a 19-year-old gold seller. Before starting the conversation, the local researchers asked him to sign a consent form, and because he was a minor, they asked for the permission of his parents as well. The gold seller refused both requests for different reasons. The gold sellers explained that he attended law classes, and having his name on a document that linked him to illicit economic activity could cause significate problems. The local researcher suggested that the gold seller sign the consent form for young members without seeking parental permission and explained again the nature of confidentially and anonymity in the project, but again, the gold seller refused, explaining that he would not sign any document for fear of the local authorities. In the end, another local researcher suggested that the consent could be recorded at the very beginning of the interview. The gold seller agreed, and the interview was conducted and audio-recorded. What this common situation demonstrates is that when working with at risk populations in repressive contexts, it is necessary to be flexible with the ethical requirements of the project design. By building this flexibility into out project protocols, we were able to avoid a situation where our efforts to be ethical, to ensure consent was properly recorded, ended up jeopardising the interests of our research participants.
Another area of particular concern was researcher and participant safety. All data collection for the project is being carried out under conditions of confidentiality. However, ethics committees usually require researchers to define the circumstances in which confidentiality may have to be broken. The general principle is that confidentiality and secrecy are not the same thing, and if a research participant discloses information that given the researcher course for concern, the researcher has an obligation to involve other professionals or agencies. Some studies work to the principle that any illegal activity uncovered during the course of the research should be reported to the relevant authority. The difficulty with this approach for our research was that it assumes that (1) criminal activity is unlikely to be revealed during the course of the research and that (2) state authorities can be trusted to safeguard the interests of those involved in the research (Scheper-Hughes 2004).
Neither assumption was safe in the context of our project, and if such an approach were applied, it could mean that local researchers would be obliged to report everyday forms of criminality such as drug use or vandalism to national authorities in whom they had no trust whatsoever. At the same time, however, the research team did not want local researchers to feel that they were never permitted to break confidentiality. The research team was concerned that local researchers may feel, for example, unable to warn one informant about a threat to their safety because of the need to protect the confidentiality of another. The team therefore developed a reporting framework that included examples of events that may occur and actions that should be taken in response. In the most severe cases, when the researcher believes someone involved in the project faces a credible threat of physical violence or abuse, they are instructed to report the incident immediately to regional coordinator, their local legal advisor, the project's ethics supervisor and/or the ethics advisory board (EAB).
A discussion will then take place between all the interested parties, and a decision will be make about the appropriate action to take. Local researchers were told that in extremis, this may mean involving the police or some other state agency, but only if it is safe, beneficial and appropriate to do so. The handbook states that in such an event, the safety of the researcher will be treated as paramount and external reporting will not happen if it is likely to jeopardise the safety of the researcher. So far, in practice, only one serious ethical event has occurred during the course of the research. In our field site in Colombia, a group of armed men threatened one of the researcher participants, causing her and the local researcher to leave the neighbourhood where research was being conducted. By leaving the site when she believed she faced a credible threat of physical danger, our local researcher followed the protocols as set out in the project's handbook. Thankfully, the situation has now been resolved and both the local researcher and research participant are not safe. The local researchers' response to the threat reinforced the appropriateness of our approach.
The last example occurred in Madrid, Spain, during a course on mediation for members of youth street groups organized at the end of 2018-before writing the ethical handbook-in which we conducted a focus group. Three of the participants refused to fill the consent form and became silent. We interpreted that decision by the fact that their group suffered police prosecution even if it was not involved in criminal activities. One of the other participants was also reticent to sign the form, although he finally did, and he managed to finish the course and obtain the certificate. Time later, that same reticent participant was arrested by the police because his legal situation in Spain was irregular, but thanks to the certificate of the course provided by us, he was able to prove his integration and his social rooting in the community and avoided deportation. Another of the participants who had refused to sign the consent form ended up in prison, asked for our help and accepted to participate in the project.
In conclusion, while it is crucial to ensure that standardised protocols such as leaving field sites when threats are encountered are in place prior to the beginning of ethnographic fieldwork, it is equally important to allow people with local expertise to judge for themselves the severity of the threat and take action that they feel is appropriate.

Conclusion
The examples found during our fieldwork in Algeria, Colombia and Spain demonstrate that methodological tools, scientific categories and bureaucratic standards, such as ethical protocols, should be developed with flexibility in mind and should be adapted to the social and cultural context in which the research is conducted. As we have shown in the last section, flexibility built into the design of our ethical protocols means that the local researchers were able to respond appropriately and in a manner that was consistent with the cultural specificities of the field site. More than to construct "subjects" based on standardised categories and methodologies, our objective is to follow configurations of relations that came from "local knowledge" (Geertz 1983)-in this case relations in, relations out and relations between youth street groups. The "macro-cosmos" of global gang's policies and representations constitutes the general frame, but the pitch is formed by a myriad of local "youth street group micro-cosmos".
What unites the three sections of the article is our commitment to emphasising the voices and perspectives of "others" is a basic ethical and methodological commitment that guides all aspects of the project including its methodology, its design, the data being collected and the ethical standards to which we adhere. This relates back to the point about "Eurocentrism" and post-colonial approaches. It is by providing researchers with autonomy and by having confidence in their expertise and judgement that we have been able to avoid reproducing the notion that "European" standards and ideas of risk and danger can be applied elsewhere. The fact that we are not "exporting" researchers from "the centre" (Europe) to the (semi)periphery (North Africa and Latin America) but rather relying on local expertise is the reason why we are able to work in this way. We celebrate the fact that since its inception, the project has made use of local researchers as a means of amplifying their voices, and this is what we mean by a "post-colonial" approach to research.
As seen in this article, traditionally, a youth gang has been typically understood as a small delinquent group of young men based in a locality. The focus has been on crime and violence. Where there has been acknowledgement of larger-sized gangs with a greater geographical range, the emphasis has still been primarily on violence and crime. Less attention has been paid to migration (rural-urban, transnational) and to the economies of gangs, that is, how members and local communities gain a variety of benefits. Gangs have also shown specific cultural practices and creative outputs. These, too, require recognition and highlight the needed of new ways of talking about transnational youth gangs in the global.
This article sets out to fill the gaps detected in gang conceptualization, and we expect to have helped to move forward thanks to the theoretical perspective proposed. The definition we have developed in this article is being implemented in the TRANSGANG project and has strong implications for practitioners and professionals working in law enforcement, public policy or with at-risk youth/young adults and for academic disciplines as criminology, social work, sociology or anthropology interested in youth street groups. The definition sets criminalization views aside and deals with inclusive and positive aspects of gang membership, trying to positivize the marginalized position of gangs within the social structure. Some research focuses on proactive experiences in gang behaviour and policies (Leinfelt and Rostami 2011; Venkatesh 2009), but very few studies systematically compare such aspects in order to find variants and invariants in the evolution or in the reversal of the criminal gang model.
Our perspective aims to recognize youth street groups as forms of youth culture to resist hegemonic discourses and practices and as social institutions to deal with and fight against stigmatization. Gangs have been examined as forms of youth culture used to resist hegemonic discourses and practices and as social resilience institutions to deal with and to fight against stigmatization. Normally, they are perceived as young, delinquent, depressed school drop-outs, jobless, marginalised, as well as aggressed by the lifestyle of the rich. However, the way of life they have chosen allows them to create distance from their sordid reality and everyday life, which is made even unliveable by contempt, exclusion and rejection.

Compliance with Ethical Standards
Conflict of Interest The authors declare that there is no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.