¡A Nosotros, nos Tienen que Respetar! (They Have to Respect us!): Gangs, Inter-Generational Conflict, and Graduated Governance in Urban Nicaragua

Gangs are often associated with violence and chaos, but they can also be institutional vectors for the imposition of particular forms of social order, based on their members’ status as locally hegemonic “violence experts”. This “gang governance” is often exclusive and volatile, however, and its underlying logic can easily change. Instances of the latter are frequently connected to generational turnovers within gangs. An obvious question in this regard is how old and new gang members interact with each other, especially in circumstances where the rules and norms upheld by the latter become detrimental to the former. Drawing on longitudinal ethnographic research in barrio Luis Fanor Hernández, a poor neighbourhood in Managua, Nicaragua, this article explores the conflicts that emerged between different generations of gang members following the gang’s transformation from a vigilante self-defence group to a predatory drug-dealing organisation, and what these might mean for the notion of gang governance.


Introduction
There exists a long-standing tradition of research highlighting how gangs can be institutional vectors for the imposition and promotion of different forms of local security orders, or what might be termed forms of "gang governance". The basis on which these are enacted is generally considered to be the fact that gang members constitute locally hegemonic "violence experts", enabling them to dominate the local communities within which they are embedded. At the same time, however, one of the hallmarks of gangs the world over is that they generally undergo regular generational turnover, due to the fact that being a (youth) gang member is a finite social role. Although the overwhelming majority of exiting gang members tend to reduce their engagement in violent activities-indeed, this is seen as a primary characteristic of "desistance"-this does not mean that they lose their violence expertise. Considering that many former gang members remain within their local communities after leaving the gang, this means that they can potentially challenge the violent hegemony of new generations of gang members. While the fact that former gang members are often held in high standing by new ones-and even sometimes considered as enduring but inactive members of the gang-means that this potential conundrum is frequently not an issue, it can become one when the nature of a gang governance regime changes, and when former gang members become the targets of a new generation of gang members and actively resist their victimisation, something that can complicate the social ordering actions of new generation gang members.
This article explores the emergence of such a situation as it occurred in barrio Luis Fanor Hernández, 1 a poor neighbourhood in Managua, the capital city of Nicaragua, and how it subsequently evolved. It begins by reviewing the literature on gang governance, desistance, and inter-generational relations between different generations of gang members, before then providing a brief overview of my research methods. The main body of the article then presents different empirical examples to illustrate the conflicts and tensions that emerged between two generations of local gang members following the gang's transformation from a vigilante gang that protected the local community in the mid-1990s, to a drugdealing gang that preyed on it in the early 2000s. In particular, it highlights how former gang members challenged the rules and norms of new gang members-albeit in a limited way-and how this led to the emergence, for a while, of a local regime of "graduated gang governance". This, however, was superseded with the increasing professionalisation of the drugs trade in barrio Luis Fanor Hernández, which led firstly to the re-institutionalisation a more unitary form of gang governance, before the emergence of a new actor in the form of an organised criminal group known locally as the cartelito (little cartel) incited a wholesale change in security governance arrangements in the neighbourhood.

Gangs, Desistance, and Graduated Governance
Numerous scholars around the world have noted how gangs can often be at the centre of local-level forms of social structuration, imposing rules, norms, and particular patterns of behaviour over the territories they control and the communities over which they hold sway (Arias 2006(Arias , 2017Brotherton and Barrios 2004;Jensen 2008;Lepoutre 1997;Mohammed 2011;Rocha 2007;Rodgers 2006a;Sen 2014;Stephenson 2015;Suttles 1968;Thrasher 1927;Whyte 1943). Such forms of what we might term "gang governance"-see Rodgers (2021a)-are generally presented as being predicated on three basic elements, including: (1) the systemic absence of, or collusion with, other forms of governance such as the state; (2) a gang's ability to be violent in a locally hegemonic manner, due to its members being "violence experts" (Tilly 2003); as well as-although this is arguably the least important, and is not always present-(3) a gang and its members being imbued with a measure of localised symbolic legitimacy, in the manner of "social bandits" (see Hobsbawm 1959). Although these three elements can be observed time and again in a variety of different contexts, the nature of gang governance can vary tremendously, and gangs can act as neighbourhood vigilantes, self-defence organisations, dispensers of community justice, or arbitrators of disputes, but also as oppressors, predators, or facilitators of exclusive forms of economic extraction, for example. The nature of gang governance obviously affects the relationship between the gang and the local community within which it is embedded-that is to say, between the governing and the governed-as can the fact that a fourth hallmark of gang governance is arguably that it is inherently volatile (see Ayling 2011;Rodgers 2006bRodgers , 2007aSanchez-Jankowski 2003), and "today's youth gang might become a drug posse tomorrow, even transform into an ethnic militia or a vigilante group the next day" (Hagedorn 2008: xxv).
This volatility is clearly at least partly due to the fact that being a gang member is generally a finite social role all over the world (Hazen and Rodgers 2014). Certainly, as Scott Decker and David Pyrooz (2011: 16) have pointed out, most youth who join a gang will eventually leave that gang through what they term a "natural desistance process" (see also Carson and Vecchio 2015;LeBel et al. 2008;Maruna 2001;Pyrooz and Decker 2011). Indeed, early studies of gang dynamics, such as Thrasher (1927) or Whyte (1943), viewed gang membership as a phase within a broader developmental life-cycle. Joining the gang was effectively seen as a form of youth rebellion, and leaving the gang was a natural consequence of growing up. Such quixotic assumptions tended to predominate within gang research until Martin Sánchez-Jankowski's Islands in the Streets (1991), a unique multisited, decade-long ethnographic investigation of gangs in Boston, Los Angeles, and New York, that was among the first to consider gang desistance in a more systematic manner. In particular, Sánchez-Jankowski (1991: 62) identifies six basic reasons for leaving a gang: (1) moving to individual delinquency (due to the high transaction costs of collective action); (2) joining another group (such as a "social club, political party, [or] organised crime [group]"); (3) the gang falling apart (because collective action is not easy to maintain in the long run); (4) imprisonment; (5) dying; and (6) getting a job. Other studies have subsequently also highlighted the emotional costs of regular exposure to violence (Decker and van Winkle 1996), having a family (Decker and Lauritsen 2002), disillusion (Padilla 1992), going to prison (Weegels 2018), or migration (Bolden 2013), as important factors leading to desistance.
These factors are still all indirectly related in the literature to ageing and life course transformation, however, and gangs can therefore be said to be characterised by regular generational turnovers. Although this does not necessarily transform the nature of a gang's governance-indeed, a measure of a gang's institutionalisation can be the degree to which its dynamics do not change across generations-it can frequently be a contributing factor (Rodgers 2007a). Moreover, generational turnover also raises important hypothetical questions with regard to gang-based security governance arrangements. Although one of the hallmarks of desistance is that individuals who leave the gang generally reduce their levels of engagement in violence (Pyrooz and Decker 2011), this does not mean that they cease to be violence experts, continuing to possess the skills that critically distinguished them as gang members from the rest of the local population. Generally, most former gang members continue to live within their local communities, something which can clearly potentially complicate a new generation of gang members' ability to be hegemonically dominant. Inter-generational tensions between former and present gang members are rarely explored in the literature, however. The gang research that has considered what are variably called "old guys", "old heads", or "veterans", either portrays them as "inactive" or "retired" (Fagan 1989;Vigil 1988), as "advisers" to new gang members (Brotherton 1999), or else have focused on former gang members who have to completely changed their persona explicitly to leave the gang way of life behind, for example through the performance of evangelical piety (Brennemann 2012;Flores 2014;Wolseth 2011). 2 But what happens in cases where former gang members act in ways that challenge the governance regime of new gang members within a given locality? How does this impact on the operations of gang governance? It is perhaps useful in this respect to turn to Aihwa Ong's (2000) notion of "graduated sovereignty", which she developed to characterise how under conditions of globalisation, sovereignty generally involves variegated modalities of government rather than being a monolithic phenomenon. Ong proposes two different ways in which sovereignty can be "graduated": firstly, with regard to the differential treatment of segments of the population by the sovereign, and secondly in relation to the emergence of "shared" sovereignty arrangements. Both of these forms of graduation could clearly apply to forms of gang governance in the face of complicated inter-generational relationships between old and new gang members. Former gang members might be treated differently from non-gang members within a local community, for example. As such, they would constitute "segments of the population [that] are differently disciplined and given differential privileges and protections" (Ong 2000: 65). This might happen due to their enduring violence expertise, or for more symbolic reasons relating to a continuing sense of respect from new gang members. On the other hand, former gang members also be able to ignore or flout certain gang-imposed rules as a result of their ability to be violent, and impose alternative norms, which could be seen to correspond to a situation of shared sovereignty. A key question in relation to both situations, however, is how stable such forms of what might be termed "graduated gang governance" might be considering that they inevitably constitute compromises, and compromises rarely endure, and this is what this article will endeavour to explore, after first offering some detail about its methodological underpinnings.

Methodological Approach
The empirical materials presented in this article draw on twenty-five years of longitudinal ethnographic research carried out in barrio Luis Fanor Hernández, a poor neighbourhood in Managua, the capital city of Nicaragua. Drawing on regular extended visits to this neighbourhood begun in 1996, I have followed the evolution of successive gang iterations, exploring the dynamics of the neighbourhood gang's mutation from being a vigilante group in the early and mid-1990s to a predatory drug-dealing organisation in the early 2000s, its replacement by a more professional and lethal drug trafficking "cartelito" in the mid and late 2000s, its revival as a classic street corner gang in the early 2010s, its transformation from a territorial group to a virtually connected peer network in the mid-and late 2010s, and finally, the co-optation of gang members by the Nicaraguan state to repress dissent against the current government in the post-2018 period (see Rodgers 2021a). I have written about my methodological approach to studying the gang and its activities in detail elsewhere (Rodgers 2007b(Rodgers , 2019, but in practical terms, this has involved nine bouts of ethnographic fieldwork totalling 23 months over 1996-97, 2002, 2002-03, 2007, 2009, 2012, 2014, 2016, and 2020, during which I engaged in a mixture of participant observation, in-depth interviewing-both one-off and repeatedly with the same individuals over an extended period of time-as well as conducting mapping exercises and focus groups. I have also carried out a household survey of barrio Luis Fanor Hernández, repeatedly photo-documented its infrastructural transformation over time, and I have also researched issues beyond gang dynamics including poverty, gender relations, labour, migration, remittances, history, memory, urban development, politics, and spatiality. Overall, I have had some form of direct interaction with almost every single individual who has been a member of the barrio Luis Fanor Hernández gang since 1996, and I have carried out formal, one-off interviews with about half of the total number. 3 This subset of 70 individuals was selected through a combination of serendipity, convenience, and purposeful sampling, and can be said to be generally representative of the population of gang members in barrio Luis Fanor Hernández. Having said this, one of the hallmarks of ethnographic research is that it aims to collect "theoretically representative" as opposed to a "statistically representative" data, 4 partly due to the fact that it is rarely replicable, because it is extremely contingent and highly dependent on the ethnographer's particular skills set (although this is arguably true-but not generally acknowledged-of all research involving the collection of primary data). At the same time, however, among the great strengths of ethnographic research is that it allows for the collection of particularly fine-grained and often difficult to uncover empirical material that frequently cannot be obtained through other research methods, and also due to its participatory nature, it explicitly allows for a contextually validated apprehension of causal mechanisms and processes that generally offer greater analytical insight than the simple correlation of phenomena observed independently and in isolation. The longitudinal nature of my research has furthermore allowed 1 3 for chronological reflexivity, and the periodic re-assessment of events and information obtained previously from a more considered and less immediate vantagepoint, allowing me to see processes from a longue durée-and therefore more consequential-perspective (see Rodgers 2021b).

Graduated Gang Governance in Barrio Luis Fanor Hernández
Although I have witnessed many different transformations of the barrio Luis Fanor Hernández over the years (see Rodgers 2021a), none was more important than the local gang's evolution from a vigilante gang that protected the local community in the mid-1990s to a predatory drug-dealing gang in the early 2000s. Although both iterations clearly constituted forms of gang governance, its nature was quite different in the two cases. During the 1990s, although gangs were a major generic source of insecurity to the inhabitants of barrio Luis Fanor Hernández, it was notable that local gang members never attacked local neighbourhood inhabitants but rather actively sought to protect them, both individually, for example frequently acting as 'bodyguards' when they went on errands in surrounding areas, as well as collectively, by engaging in semi-ritualised forms of gang warfare with outside gangs that not only aimed at "defending" the local neighbourhood territory, but also due to its regular and ritualised nature limited the scope of violence and provided a measure of predictability to the local neighbourhood community (see Rodgers 2006a). Most local neighbourhood inhabitants did not fear but clearly held gang members in high esteem, as was obvious from the way that they almost always cheerfully greeted and bantered with them on street corners, were happy to give them glasses of cold water during hot spells, or offered them shelter in their homes during flash downpours in the rainy season (none of which they necessarily did for non-gang members). 5 Perhaps more obviously, the quasi-symbiotic relationship between the gang and local neighbourhood inhabitants was also evident in the fact the latter never called the police during gang wars, and nor did they ever denounce gang members, even often going so far as actively hiding them and providing false information to any figure of officialdom on the rare occasions any came round the neighbourhood asking questions about local gang members. 6 Gang members, in other words, were viewed extremely positively, even celebrated (see Rodgers 2015).
Around 1999-2000, however, the nature of gang governance in barrio Luis Fanor Hernández shifted from being about territorial and community protection to ensuring the proper functioning of an emergent drug economy which gang members integrated individually as street dealers and collectively as the drug business' local security apparatus (see Rodgers 2018, for a detailed analysis of the rise of drug trafficking in barrio Luis Fanor Hernández). In stark contrast with the past, the gang began to impose a localised regime of terror by repeatedly threatening and instrumentally committing arbitrary acts of violence against the inhabitants of barrio Luis Fanor Hernández, in order to reduce the risk of denunciation and facilitate a drug-based process of "primitive accumulation" for themselves and local drug dealers (see Rodgers 2007a). Fear of local gang members was widespread among neighbourhood inhabitants, who now actively avoided them in the neighbourhood streets, keeping their heads down whenever they walked past them, and did not engage in conversation with them, contrarily to the past. Gang members were an overtly menacing presence, sitting on street corners in small groups of two or three, openly playing with their guns and machetes, glowering at passers-by, shouting out threats and insults, and often asking them to hand over small sums of money or something that they were carrying, which people did without comment or hesitation. Local inhabitants were also frequently randomly beaten up and assaulted in the streets, as part of a general "disciplining" of the neighbourhood population.
Despite the stark differences in the nature of the gang governance imposed by the two iterations of the gang, in both the 1990s and the 2000s this was arguably based on the fact that gang members constituted locally hegemonic "violence experts" (Tilly 2003). In the 1990s, their violence was turned against outsiders, while in the 2000s it was turned inwards the neighbourhood, but in both instances, the effectiveness of the violence that they deployed very much depended on gang members being the locally dominant actors of violence. At the same time, one of the immediately striking elements of this radical transformation in the nature of the gang's governance regime was a generational turnover in the gang's membership (see Rodgers 2007a). Although there had been a generational turnover between the first iteration of the gang that emerged in the early 1990s and the iteration that I encountered in the mid-1990s, this had not been significant as the gang's governance regime had not so much changed as institutionalised, and "retired" gang members from the early 1990s supported and aided the new generation in the mid-1990s. The change of direction of the violence underpinning the gang's governance regime in the 2000s led to a different situation, however.
In the first instance, the new generation of gang members tended to treat former gang members differently to non-gang member inhabitants of barrio Luis Fanor Hernández. To a certain extent, this was due to the particular symbolic cosmology of the gang, insofar as former gang members were "respected" by the new generation, almost in the same way that an ancestor might be, for example. This aura also extended to the family of former gang members, who were less likely to be assaulted, mugged, or threatened by new gang members in the early 2000s. At the same time, there were clearly also some more "practical" reasons for the new generation's "respect" for former gang members, as was highlighted by an episode in March 2002 when a couple of new gang members, la Juana and Chucki, assaulted a former gang member called Felix one night as he came back into barrio Luis Fanor Hernández after a night out, and he defended himself more than efficiently, breaking one's arm, and stabbing the other in the stomach. During an interview just after the event with la Juana, whose arm had been broken, he complained "it was fucking dark, maje, how were we supposed to know that it was Felix?".
"Would you have attacked him had you known?", I asked. "¡Claro que no! (Of course not!) I mean the guy is one of the first gang members here in the barrio, and he also served in the army and is seriously dañino (badass)…" This of course rather starkly illustrates how former gang members did not cease to be violence experts on leaving the gang, and as such, could challenge the new generation's hegemony over violence. Indeed, former gang members from the 1990s possibly even had a comparative advantage compared to the new generation in the 2000s, insofar as their violence expertise came from either being trained as military conscripts at the end of the 1980s, or learning directly from these conscripts, rather than third or fourth hand as was the case of the new generation (see Rodgers 2017). Certain parallels can be made here with former military servicemen from the 1980s during the 1990s, when the latter were the only non-gang members to stand up to gang members, as an episode that occurred in June 1997 highlights well. I had just sat down with Pablo Alvarez and a few members of the Gomez household in barrio Luis Fanor Hernández to share a few drinks when suddenly, Pablo's wife, Carola Gomez, came running into the house, screaming "lo mataron, lo mataron, mataron a Pablito (they've killed him, they've killed him, they've killed Pablito)." It rapidly turned out that her son Pablito had not actually been killed but rather attacked and beaten up by the Comemuertos gang in the nearby barrio Pablo Quintero, where she and Pablo lived with their son and daughter. The gang had recently invaded and occupied barrio Pablo Quintero, after defeating the local gang. The Comemuertos had taken Pablito's shirt and shoes, but he'd managed to escape when a group of his friends intervened, but some of the Comemuertos gang members had followed him, and had him surrounded in his aunt Manuela's house, where he had taken refuge. Carola and Pablo immediately ran off and came back an hour or so later with Pablito, who had been beaten up and stabbed in the arm. As I bandaged him up, Carola told me what had happened: "I've got a crazy husband, Dennis," she began, with half a smile on her face. "Loco, loco, loco! (Crazy, crazy, crazy!) I don't know what I'm going to do with him he's so crazy. You should have seen him, he was completely encachimbado (mad), and when we arrived at Manuela's, where the Comemuertos gang members were still trying to break in, he ran straight into them. We'd stopped off to get his gun, and he began shooting at them. ¡Pum, por este lado, pum por el otro! (Bang this way, bang that way!)-he's mad I tell you! He still thinks that he's in the army, in the mountains… They were well scared and so ran away… But man did he get even angrier when he saw Pablito! He immediately went off to barrio Chinandega, where the Comemuertos come from, to kill them!" "And did you?", I asked Pablo, who was listening to his wife's tale with a Cheshire cat-like grin on his face. "No," he answered. "But I got the shoes back, as well as a shirt. I found those hijuéputas celebrating, near the entrance to their barrio, already high. No lookouts, nothing, so it was easy, easier than anything I ever had to do in the mountains. Stupid fucks! I came up from behind, and put my gun to the head of one of them and said 'return what you stole from my son.' They gave me the shoes, but they didn't have Pablito's shirt anymore, so I told the one with the cleanest shirt to give it to me, and then backed off. Gave them a warning shot, and then I ran. Just like when we raided the Contras, during the war." "I told you, he's crazy! He thinks it's the war all over again," Carola interjected. "The Comemuertos have been terrorizing our barrio for weeks now. They've disbanded our pandilla, and they've assaulted God knows how many people. Everybody is terrified, scared to leave their houses, and even in their houses, because the pandilleros throw stones at them, and even shoot their mortars at the houses for no good reason... It's hell, how are we supposed to live? It's like it was during the insurgency against Somoza, when we lived in fear of the National Guard. Oh God, what if they find out who you are and where we live, Pablo, what are we going to do? They'll want revenge for what you did to them-they'll kill Pablito, they'll kill us all!" "No they won't," Pablo said grimly. "No jodás, if they come again, I'll kill those sons of bitches, for real this time. If they fuck with us, I'll really take those amateurs out…" Such violent confrontations of gang members by non-gang members were relatively rare in the 1990s, however, and Pablo's reaction was clearly a function of his particular background as a revolutionary guerrillero during the 1970s and a career military during the 1980s, both of which meant that he had certain martial skills and experiences that provided him with the practical means to stand up to the gang members who had attacked his son. He was not the only person in barrio Pablo Quintero or barrio Luis Fanor Hernández with such a background, and while I heard of several others behaving in analogous ways, more often than not, it was stories of individuals successfully defending themselves when being mugged rather than engaging in the kind of action that Pablo did. This was clearly partly because acting individually against the gang group was unlikely to be successful most of the time (as the Comte de Bussy-Rabutin once famously put it, "God is usually on the side of the big squadrons against the small"). 7 As individuals such as Pablo aged, however, their ability to resist gang members waned. In 2007, for example-10 years after the episode with the Comemuertos gang members-Pablo tried to defend himself when a local barrio Pablo Quintero gang member mugged him, and he ended up being shot in the arm and badly beaten up, when had he not resisted, he would likely have just been roughed up a little and then left alone. "Ya no puedo (I can't anymore), Dennis," he told me in a subsequent interview, "soy viejo (I'm old), and I can't defend myself anymore…".
The temporal distance between having been a gang member in the 1990s and the early and mid-2000s was not as great as the temporal distance between Pablo's guerrillero and military experience in the 1980s and his unsuccessful attempt to resist being mugged in 2007, but the dynamics of inter-generational gang member relations changed significantly between the early and the mid-2000s as a result of the barrio Luis Fanor Hernández drug economy's institutionalisation (see Rodgers 2018). This led to a transformation in the way that former gang members engaged with the new generation; in particular, while in the early 2000s, inter-generational conflicts had almost always occurred on an individual basis, by the mid-2000s the actions of former gang members against the new generation had become more collective, because the current gang was becoming less and less discriminating in who they assaulted. Take for example an episode that occurred in 2004, when a couple of current gang members assaulted the sister of a former gang member called Norman. As he explained subsequently during an interview in 2007: "Ok, so what happened was this," Norman began. "You know my sister Lisette, yes?" "Sure," I answered. "I've seen her around -she's what, 16, 17 years old, right?" "Seventeen, that's right, so she was fourteen when this all happened… So anyway, she was running an errand for our mother, you know, buying some leche agria (sour milk) from a pulpería (cornerstore), when some of the gang members starting dissing her." "Why?". 1 3 "Pfff, I don't know… They were drunk, high, or bored -all three, perhaps? You know how those guys were." "Yeah, sure…". "OK, so they were dissing her, and if that's all they'd done, it would have been fine, you kind of expected that to happen to you with gang members then. I mean, not to me, because I was a member of the gang before, and you know, ¡a nosotros, nos tienen que respetar! (they have to respect us), although by then the little shits increasingly didn't and thought they could get away with anything… Anyway, Lisette knew that she just had to ignore them and go about her business, but then two of them, Lino and Giovanni, followed her and trapped her in an alleyway, where they took turns finger-fucking her (metiéndole el dedo)." "Fucking hijuéputas! How was Lisette?". "She was fine, I mean it wasn't anything she hadn't done before, and they didn't do anything else, and what scared her more was that they also told her that if she ever said anything or did anything to fuck with them or the gang, that they'd gang rape her, so she came back terrified… I mean, I knew as soon as she came back to the house that something had happened, and my mother and I got it out of her. I was so fucking mad, I tell you, I mean, what the fuck were those assholes thinking? How dare they touch my sister?" "Didn't she tell them that she was your sister?" "Sure she did, but they just laughed at her and told her to tell me that I could just fuck off (andarse por la quinta verga)." "Wow, so they really didn't care?". "No, they didn't. You've got to understand, Dennis, things weren't the same as when you were here before. At the beginning the new guys respected us old-timers, but as they got more and more into la piedra (crack cocaine), they stopped giving a shit, and they began to give us shit just like everybody else." "So you decided to get revenge?". "Yeah, the fuckers needed to be taught a lesson, I mean it was bad enough to see them and the rest of the gang tirándoselo del culo, fucking off everybody in the barrio, and I couldn't care less if they insulted me, but I couldn't let them get away with fucking with my sister, so I got several of the old guys together…" "Who?". "Julio, Bayardo, Milton, Felix, Jasmil, and Wilmer -and then we went looking for the two hijuéputas." "But weren't you afraid that you'd be taking on the whole gang?" "Sure, that's why I got the others together, and a couple of us who still had our old animales (guns) brought them along too, but by the time we found Lino and Giovanni, what had happened to Lisette had gone round the barrio, and it was clear that the others weren't going to interfere, and so we were able to beat the shit out of them and teach them a lesson no problem." "Why would the others care? I mean, those guys were in control of the barrio, no? They could do whatever they wanted, like the cartelito today, no?" "No jodás, Dennis, they were nothing like the cartelito today, those guys are really bad-ass, you can't talk to them, they've got big guns, AKs, Uzis, that kind of thing, so you can't fight them, they'll just take you out without thinking. Sure, the gang was in control of the barrio, but they knew that us old-timers would be difficult to take down, especially as there were several of us and we were armed, and even those hijos de la setenta mil putas (assholes) knew that I was justified in seeking revenge, so they didn't even try to stop us from beating Lino and Giovanni…" On the one hand, Norman's narrative highlights how by the mid-2000s former gang members had lost much of their kudos with the new generation of gang members. The latter did not hesitate to pick on members of the families of former gang members, for example, and he notes how they were acting increasingly disrespectful of him and other former members of the gang. On the other hand, it is also clear that former gang members were clearly still considered a potential challenge by the new generation, who were clearly being selective in terms of how they dealt with them. The central focus of the gang's gang governance at the time was ensuring that the neighbourhood drugs trade operated smoothly, and a conflict with a group of former gang members would likely have been highly disruptive. Indeed, as Lino put it in an interview in 2007, "sure, Norman was allowed to beat me and Giovanni up, because it wasn't important, it didn't touch our bisnes, and it kept him from becoming disruptive (joder)". The latter was very much the bottom line, however, and when an ex-gang member from the 1990s called Wilmer tried to hijack a drug deal in 2005, the gang simply killed him, as I learnt during a conversation with a middle-level drug dealer called Bismarck in 2007: "Hey, Bismarck," I asked, "where's Wilmer, I haven't seen him for ages, not since my visit in 2003 I think…" "He's dead, Dennis, didn't you know?", Bismarck answered. "No jodas, he's dead? What happened to him?" "Ay, this is old stuff, maje, it's not important anymore… Pero bueno, he tried to fuck up one of Chucki's drug deals, and so the gang took him out, de uno solo…" "Just like that? No warning?". "Just like that." "When was this?". "Pffff, four-five years ago, something like that? But who cares? Everything's changed now, the gang doesn't exist anymore, and the drug dealers are now the cartelito…" The latter development intimated by Bismarck clearly suggests that graduated security governance arrangements only last for a while, probably because ultimately, they are compromises. In this regard, a much more unitary form of security governance was established in barrio Luis Fanor Hernández from 2006 onwards by the cartelito. This was linked to the way that the local drug economy developed. When this emerged in 1999-2000, it had been an activity initially organised in a rather ad hoc and informal manner, and at first extensively involved gang members, but as it steadily professionalised, the latter began to be excluded (see Rodgers 2018). By the end of 2005, the barrio Luis Fanor Hernández drug economy was run by a group referred to locally as the cartelito, or "little cartel", which was principally made up of individuals who were not from the neighbourhood. The cartelito actively sought to develop its own security infrastructure and rapidly clashed with the barrio Luis Fanor Hernández gang, at least partly in order to eliminate them as potential challengers in the neighbourhood for the local monopoly over violence. Members of the cartelito would wander around the neighbourhood openly bearing arms, intimidating and sometimes shooting at any gang members they saw hanging around in the streets, to scare them and to warn them "not to get uppity", as local inhabitants put it. After a few months of this, the gang decided to retaliate and one evening launched an attack on the house of the leader of the cartelito, which led to a shoot-out between the gang and members of the cartelito, during which a gang member called Charola was badly wounded.

3
The other gang members fled, leaving him behind, and a member of the cartelito called Mayuyu went up to Charola and killed him, shooting him in the head, execution-style, "as a warning to the others", as he put it in an interview in July 2012, and the gang effectively ceased to exist as a collective unit in barrio Luis Fanor Hernández (see Rodgers 2015), and the cartelito became the sole security arbitrator in the neighbourhood.
Paradoxically, and very much against Hobbesian "Leviathan"-type analyses, the unitary form of security governance imposed by the cartelito was by no means easier for local neighbourhood inhabitants. Indeed, the period 2006-2009 was probably the period of greatest insecurity during the 25 years that I have been going back and forth to barrio Luis Fanor Hernández, partly because the cartelito's membership included individuals who were not from the neighbourhood, and therefore people had no idea who they were (see Rodgers 2019). Unknown, armed individuals that people associated with the cartelito randomly patrolled the neighbourhood on motorbikes, arbitrarily intimidating the neighbourhood population-"to train us", as an inhabitant called Doña Yolanda put it during an interview-including in particular violently preventing local youth from congregating on street corners, very obviously in order to avoid them potentially coalescing into a gang that might eventually challenge them. The cartelito also fought against other equivalent organisations located both in Managua and beyond, in order to secure an increased share in the drug market. Shoot-outs in and around the neighbourhood became commonplace albeit unpredictable occurrences. In 2009, however, the cartelito began to reduce its involvement in local drug-dealing activities, refocusing instead on drug trafficking, mainly because this is intrinsically a much more lucrative activity. This significantly changed local violence dynamics, as the cartelito no longer sought territorial control in order to manage sales, but generally avoided attracting attention, using the neighbourhood only as a residence and organisational base of operation. This meant that violence declined, and as a result, life in barrio Luis Fanor Hernández became much safer, even if a certain diffuse fear of the cartelito remained, and it was clearly seen as the dominant authority in the neighbourhood.
In 2011, however, the leader of the cartelito, along with a number of other members, were arrested-reportedly because a rival whose group had developed close links to certain members of the Nicaraguan government was consequently able to supersede the cartelito's bribes to the Police-and this opened up what might be termed a "sociological space" for new actors to emerge, and new security governance arrangements to develop. By mid-2012, a group of twelve 14-to 15-year-olds was regularly coming together, hanging out on local street corners. All were involved individually in a variety of petty criminal activities, although most local inhabitants considered them more or less innocuous as these were generally low-level and they did not carry out their delinquent activities in the neighbourhood, albeit less due to any consideration for local neighbourhood inhabitants, and more for reasons related to preserving their anonymity when carrying out delinquent acts, as well as former gang members coming together to "explain" to these new gang members that they were not to target local inhabitants following an "unfortunate mistake" shortly after the new gang began to coalesce as a group. In July 2012, they acted for the first time collectively, attacking the local gang in a nearby neighbourhood as a group. Although they were repelled, with several individuals being injured-two of them critically-this event led to the beginning of a perception in barrio Luis Fanor Hernández that "the gang is back", as Doña Yolanda put it.
This impulse for this collective action had been related to the group learning about the neighbourhood's gang history from former gang members, and deciding after a heavy drinking session to emulate their predecessors, rather than any form of identification with the local neighbourhood population or territorial. Collective gang wars nevertheless became a significant practice of this new iteration of the gang and marked the beginning of a new cycle of evolving gang governance, whereby initially graduated and imitative iterations gave way to progressively more unitary forms, fuelled by ever-increasing levels of violence that were motivated by a desire for hegemonic territorial control, first in order to increase the profitability of low-level drug dealing, but subsequently as part of a campaign of state repression that followed a popular uprising in Nicaragua in April 2018, as gangs all over the country were brought under the aegis of the authoritarian government of Daniel Ortega as it sought to pacify and re-establish control over the country's population and territory (see Rodgers 2021a). This arguably effectively constituted a scaling up of gang governance from the local to the national level, but its motivation was a challenge to the government's authority, something that suggests that graduated forms of authoritarian governance, whether by gangs or otherwise, are unsustainable, and there is a tendency for them to become more unitary over time, just as occurred previous in barrio Luis Fanor Hernández in the 2000s.

Conclusion
It is widely recognised that across the world, gangs are not just sources of indiscriminate violence and chaos, but can also, under certain circumstances, impose forms of local social order. The nature of this gang governance can clearly vary, but it is generally predicated first and foremost on gang members' ability to be locally hegemonic violence experts. At the same time, however, gang membership is generally a finite social status, and individuals generally mature out of gangs, for a variety of reasons. Although some former gang members leave their local contexts (for example migrating or due to being imprisoned), most continue to reside locally after leaving the gang. They do not however lose the skills that made them "violence experts" when they leave the gang, which begs the question of the nature of the relationship between retired and new gang members, since the former could clearly potentially challenge the hegemony of the latter. Drawing on empirical material gathered over the course of 25 years of longitudinal ethnographic research in barrio Luis Fanor Hernández, a poor neighbourhood in Managua, the capital of Nicaragua, this article has endeavoured to highlight one particular way the co-existence of different generations of gang members can impact on the nature of a gang's governance, namely how it led to the emergence of forms of "graduated gang governance" in the early 2000s whereby former gang members from the 1990s were treated differently from the rest of the local population, to the extent that they in many ways can be said to have "co-governed" with the new generation.
At the same time, the article also shows how gang governance is highly volatile, and that its underlying logic can change, moreover in ways that mean that graduated forms of gang governance can progressively become untenable. The barrio Luis Fanor Hernández gang's increasingly professionalised involvement in drug trafficking in the mid-2000s, the consequent intensification of gang members' violence, and their development of a more exclusive mercantile outlook all meant that "co-governing" with the previous generation became an inviable proposition, and these began to be treated like any other local neighbourhood inhabitants. In other words, a more unitary form of gang governance was (re-) imposed, something that became even more evident when the drugs trade was taken over by the more exclusive, violent, and socially disembedded cartelito, and it muscled the barrio Luis Fanor Hernández gang out of security governance altogether. This suggests that there may be an inherent tendency for gang governance to tend towards becoming more unitary in nature, something that is also observable in the re-emergence and longitudinal evolution of the gang governance associated with successive iterations of the barrio Luis Fanor Hernández gang following the cartelito's fall in the early 2010s. Seen from this perspective, graduated forms of gang governance likely constitute rather exceptional instances of gang-imposed social order.
Funding Open access funding provided by Graduate Institute of International and Development Studies. The research leading to the results presented in this article received partial funding from the Royal Anthropological Institute of Great Britain and Northern Ireland's Emslie Horniman Scholarship Fund, and the London School of Economics' Crisis States Research Centre.
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:// creat iveco mmons. org/ licen ses/ by/4. 0/.