Shadowing: Warrants for Intersituational Variation in Ethnography

First Warrant: Shoring up Intersituational Claims

As ethnographers intermittently point out, one of the strengths of participant observation is that rather than accessing what people say in experience-distant situations (such as surveys or interviews), the ethnographer witnesses what people actually do. In its latest incarnation, the problem of extrapolating action from such contexts of talk has been dubbed “the attitudinal fallacy” (Jerolmack and Khan 2014). Interviews and surveys, the authors argue, provide researchers with access to vocabularies of motives (see Mills 1940) or, at best, general attitudes. But, following writers such as LaPiere (1934) and Deutscher (1966, 1973), Jerolmack and Khan note that it is often tricky to make inferences from talk to action. The problem is that interviews or survey situations are specific, experientially distant, and well bounded. To generalize from one situation (and a relatively odd and reflexive one, at that) to other situations and arenas requires a leap of faith that is often unwarranted. People can say (both to others and perhaps to themselves) that they do not care about race, yet discriminate in job interviews (see Pager and Quillian 2005); conversely, people may profess racist attitudes and jettison their intolerance when encountering the face of the other (LaPiere 1934; see also Levinas 1981).

Although some sociologists argue that the notion of the attitudinal fallacy has been overstated (see, e.g., Dimaggio 2014; Lamont and Swidler 2014; Vaisey 2014), it raises an important set of questions. If we follow the logic of Jerolmack and Khan’s argument, it is not simply an “attitudinal fallacy” that they identify. Rather, it is an instance of a much wider empirical challenge—that of inferring from what people say and do in one situation something about what they say and do in another. If ethnographers are seriously in doubt about the move from circumscribed interviews or survey situations to moments of “action,” they should be equally worried about their own craft.

This is because ethnographers quite often extrapolate from one set of situations to a very different set of contexts. And yet, theoretically, there is no reason to think that the very same arguments that sow the seeds of doubt for the relationship between interviews and situated action would not be equally relevant for inferring from one context of action and interaction to another. Contexts of action are highly varied. Action itself is a vague sociological term, both unclearly demarcated from “talk” (which, in many situations, is action) and capturing anything between a bar fight to a wink. And, for all this, ethnographers routinely assume that readers would understand something general about the ethnographer’s interlocutors’ lives through the window the ethnographer provides to a relatively limited set of (usually) public situations. For example, studies that have shown how seemingly disorganized gatherings of poor men on “the corner” are actually socially organized assume that such organization says something general about other contexts of poverty, unemployment, or crime; ethnographers of religion have often looked at religious services to make general claims about religion, and the list goes on.

We can see how pervasive this challenge is by looking at the study of sports and poverty, a site we will return to in the ethnographic section below. Thus, a common trope in such work is that by engaging in sports—even violent sports—young poor men pull themselves out of the violent context of the neighborhood. Thus, for example, Wacquant (2004) quotes a number of young boxers from the gym he worked in who all say how boxing in the gym allowed them to move away from criminal forms of violence, quotes at length a conversation with the trainer who claims this to be the case, and notes that “Many professionals willingly confess that they would likely have turned to a life of crime if not for the discovery of boxing” (p. 27).

And yet, of course, Wacquant is not conducting an ethnography of “the street.” And although the people he spends time with go in and out of the street and the gym, he sees them when they are performing a particular self, within a particular context. This, we should emphasize, does not mean that the context of the boxing gym is not an interesting one, that the men he talks to are being untruthful, or that if Wacquant were to shadow them he wouldn’t find that boxing affords some of the boxers a way out of other forms of violence. And yet, exactly because the ethos of the gym stresses control and changes in rhythms and lifestyle that should take place outside the gym (e.g., the ban on sex or junk food before a fight), it gives rise to predictable narratives of redemption and of strong intersituational coherence.

Moving among contexts would have likely made Wacquant’s story more complex both ethnographically and morally. If some boxers would avoid violence through the gym, others may simply be getting better at it. Thus, for example, in the epilogue to Bourgois’ (2003) In search of respect, he tells of an encounter with a baseball-bearing young man on his way to collect money from someone who owes him cash, bragging about going to play baseball “with a nigga’s head.” As they talk, the young man begins speaking enthusiastically about his boxing training regimen, and how he stopped smoking and drinking sweetened drinks on orders from his trainer. This, however, has little to do with the avoidance of violence, as “with a mischevious wink, he tells us not to tell his trainer, but that he is trying to build his strength and punch with enough force and precision to break someone’s neck with one punch” (Bourgois 2003:344-45).

This is, again, not to call into question Wacquant’s remarkable work on the development of the pugilistic habitus or on the centrality of the gym in many boxers’ lives. But understanding how the discipline engendered in boxing and the very craft of boxing itself is intertwined with the street would have required a slightly different research design. Seen through the lenses of this first warrant, shadowing buttresses the study’s validity in cases in which the specific situation is leveraged to make claims about other situations—either in claiming that the situation affects these other moments or in claiming that a specific situation can serve as a synecdoche for a whole fabric of situations.

There are certain forms of intersituational variation that the ethnographer cannot access—the movement from school years to the job market, for instance, just takes too long to make intersituational variation a viable strategy (see, e.g., Khan 2011; Lareau 2003; cf. Black 2009; Smith 2006, 2014). Moving among some other situations simply takes extra work. If ethnographers truly care about the relationship between what people say they do and what they do—between one set of situations and another—then being able to empirically trace whether what happens in Situation A is transferable to Situation B is a warrant for shadowing.

Second Warrant: Meaning and Variation (or a Wink and a Blink)

Perhaps the most celebrated warrant for ethnographic work is the ability of the ethnographer to explicate local meaning-making processes—to provide “the native’s point of view.” Precursors to this warrant appear already in Weber’s (2011) Verstehen method of historical sociology as well as in Kenneth Pike’s (1967) differentiation between “emic” and “Etic” perspectives. It was then developed and popularized by Clifford Geertz’s (1973) celebrated borrowing of the term thick description from philosopher Gilbert Ryle. Ryle’s (1971) example of the meaning of a wink, a parody of a wink, a practicing of a wink, and so on, is still instructive. To figure out whether someone winks or blinks, whether they wink or parody somebody else’s blink, whether they are practicing winking, or actually winking, is a very different endeavor than to thinly explain “what happens” when they perform the action. Thick description requires us to explicate how people intend their actions will be understood and the shades of meanings with which they imbue the act.

Providing such “thick” description compels us to take what people say about what they do quite seriously. Whether we study winks or Berber sheep trading, we need to understand how people learn to make sense of their worlds and how they creatively enact meaning. As ethnographers enter the contexts of meaning making and witness how such meanings are enacted and how they crystallize, they are uniquely situated to provide thick description. Thus, in some quarters, thick description became practically synonymous with the craft of ethnography. And yet—whether or not we agree that ethnography is synonymous with the explication of local meaning—there is a sense in which Geertz may inspire ethnographers with false confidence. The notion that if we just hang out “there” for long enough, we will get what people mean by their actions is slightly misleading, at least if we do not carefully specify where there is and how this there is related to other sites and meanings.

The problem here is not so much that people may not tell the researcher the “truth” or that they will opt for a facile vocabulary of motive when pressed to tell themselves stories about themselves. The potential limitation we would like to point out is rather that meaning is always “oversaturated” (see Marion 2002; Tavory and Timmermans 2014; Timmermans and Tavory 2012). Although meanings are made in situations, they shift as actors move through time and contexts, so that what something “means” is far from simple. And, much as any sign is defined against a background of its relationships to other signs (de Saussure [1916] 1986), it is also located in relation to the same sign in diverse situations (Tavory 2014).

The relationship among situations is important for ethnographers studying meaning for a few reasons. First, by moving among situations, ethnographers can attend to axes of variation they would otherwise likely ignore. For example, Robert Smith’s (2006) understanding of how second generation Mexican American boys and girls negotiate gender shifted as he went with the families to Mexico. Among other considerations, he observed how transnational life generated conflict over gendered roles and expectations in his interlocutors’ lives. The axis of “Mexican” and “American” notions of masculinity and femininity would not have fully occurred to him had he only followed the young people in New York City or their parents’ birthplace in Mexico. Once he realized this axis of variation existed, however, he could see how they attended to it in multiple settings.

Second, there is a temporal dimension to meaning. Is, for example, a flirtation an interlude to a romantic involvement, or is it a way to pass time that plays with never-fulfilled possibilities? In such an ambiguous context, the answer depends not so much on the situation itself but on what happens later on. In moments of “local action” (Leifer 1988), meaning is often held in abeyance. Moreover, even in unambiguous situations, meanings often change with the passage of time, and such changes can come to be themselves expected: A wink that parodies another party’s wink (as Ryle describes) can be considered a light hearted spoof in the moment but later understood as a cruel put-down. As action is propelled toward possible futures, the question of the “true”-situated meaning becomes obviously fraught.

Lastly, meanings are not locked in specific situations but are regularly transposed. As theorists of practice and meaning have argued (see, e.g., Bourdieu 1977; Giddens 1984; Sewell 1992), creative change often occurs precisely through such transpositions. And although these theorists stressed the transpositions of different meaning structures, it is also the case for ostensibly similar meanings among situations. Indeed, such situational transpositions are the necessary condition for humor and metaphor (Tavory 2014), as they are for everyday problem solving and creativity.

In short, meanings shift among situations, are defined by their relationship to these other situations, are often made retroactively, and can always be potentially transposed. If one warrant for ethnographic work is to provide an analysis of local meaning, then missing the meanings of transpositions and metaphor, hidden variation, or the way that possible future meanings invade the present impoverishes the understanding of meaning. Ethnographers would miss the ways in which different forms of meaning making bleed into one another in surprising ways (see, e.g., Wagner Pacifici 1994), the subtleties of humor and metaphor, as well as sets of meaningful differences that they simply wouldn’t otherwise conjure. While the construction of situated meaning is important, these intersituational hues give meaning making much of its dynamism and multivocality.

Third Warrant: Shadowing and the Fabric of Social Worlds

In the previous warrant, we assumed that ethnographers trace meaning as an end in itself. Yet when ethnographers think about thick description, they often blur meaning and personhood. They describe particular meaning-making contexts and situations, but also particular actors who transcend these situations. But for a handful of strict methodological situationists inspired by ethnomethodology and talk-in-interaction (see, e.g., Brubaker et al. 2006; see also Sacks 1979), ethnographers seldom assume that the self emerges only in the specificities of the situation in which it is elicited. In the words of Dr. Seuss, “a people house has people too.”

As a way to flesh out this assumption, it is instructive to return to the work of Clifford Geertz. Thus, when Geertz describes the meanings of the Balinese cockfight in his (1973) essay “Deep Play,” he purports to tell readers not only something about the cockfight but also something about the Balinese themselves. They are, after all, the ones who “read” their cultural texts in particular ways. The meanings that people creatively construct and evoke thus provide the ethnographer with a window into their lives: whether it is the tension between politeness and “animal nature” or the structure of loyalties within and among villages. It is in this sense that much of ethnography has a humanistic streak: Because meaning making and personhood are intertwined, when ethnographers study meaning, they study it in relation to a specific social world and the people who inhabit it.

Shifting the analytic lens from meaning to personhood may seem to be relatively minor. After all, since people are the ones who make meaning, it is obvious that if we trace meaning making, we already know something about the people making it. And yet, in the move from meanings to people, we necessarily move from the ways in which meanings are actually made and understood to the ways in which social worlds are structured. It has become somewhat of a truism to note that as people move among situations, they also shift their identifications. It is not only meanings that changes among situations but also—if often subtly—actors themselves (see, e.g., Simmel 1964; White 1992).

If part of the warrant of ethnography is to gain an appreciation for the lives of the interlocutors the ethnographer tells the reader of, then moving among situations becomes crucial. This is because, in most cases, the social lives of the people the ethnographer writes of rhythmically shift among situations and identifications. Thus, for example, Tavory (2016) found that Orthodox men in Los Angeles who worked secular jobs had to juggle very different selves—during their working hours they often had non-Orthodox friends and interests and were defined by their job and work history as much as (if not more than) by their religion. When they came home, however, they often had to erase this working world, alongside its camaraderie and meaning, and re-erect the symbolic boundary between Jew and non-Jew.

As this example demonstrates, shifts among selves and situations are predictable. And, as actors move among these situations, they construct a patterned fabric of situations and selves. To take another example, Duneier (1999) shows how one unhoused man he worked with, Mudrick, juggled his life on the streets and his relationship to his granddaughter who did not see him as a “homeless person.” Understanding Mudrick requires understanding him as he is on the street, as he acts when his granddaughter sometimes comes to the corner, and as he acts when he goes to see her. Thus, even a relatively minor foray in a book that is generally situation based provides a depth that recasts the way that the reader encounters the men Duneier describes.

In short, then, shadowing reveals the way that people move through social life, and how such movements are followed by shifts in meaning and selfhood. Instead of assuming that the situation provides a window into a life, ethnographers can opt for the much more conservative (and much more interesting) assumption: that it is not a single situation, but the predictable rhythm of situations and selves that defines the lives of the actors they follow.

Fourth Warrant: The Wider Picture and Causal Narratives

The last warrant we highlight moves from ethnographers’ units of analysis to the general claims they often attempt to make. Whereas interactions, situations, meanings, and people are often the units of analysis that ethnographers use, ethnographers (like other social researchers) are often after a different prize: a wider social pattern that emerges through the ethnographic detail.

Thus, for most ethnographers, the specific situation and social world they study is portrayed as a synecdoche for a general condition (see Small [2009] for the challenges such a position entails). This form of rhetorical induction (Edmondson 1984) is not the only form of generalization possible: Some ethnographers use the empirical case to construct theory on a more abstract level; still others really are interested in telling the story of a particular social world they have been privy to. However, the majority of ethnographers (at least in sociology) investigate a set of local conditions that they see as standing for a more general empirical set of determinants. Through the ethnographic, they can show the causal power of the general condition in the minutiae of everyday life.

This causal relationship between specific observations and general social forces may seem like a truism—and one that is repeatedly attacked by methodologists at that. But seen from this angle, what shadowing potentially allows ethnographers to do is twofold. First, shadowing allows them to make a stronger causal case by showing how the same sets of pressures become consequential in different ways in different situations. Thus, for example, one of the key rhetorical moves in Alice Goffman’s (2014) recent On the Run occurs when she shows how police surveillance affects romantic relationships in the neighborhood she studied. The reader could have easily imagined that in a situation of intense police surveillance, young black men learn to run from the police or that they routinely expect to be profiled. It is far more jarring to see how young women use the warrants for young men’s arrest both as leverage against the men they are involved with and as ways to show commitment and love (by, e.g., attending their trials or visiting them in jail).

If the condition the ethnographer studies shapes their subjects, showing the effects of the same condition across situations buttresses their claim for its general power to shape the social world. Rather than being only able to make a case that the general social pressure affects a specific set of situations, the ethnographer creates variation within the study. And while the “scope conditions” of the study are still limited, the proliferation of effects strengthens the plausibility of the ethnographers’ claims within the study.

Second, intersituational variation can also show that a given social condition has not only surprising but contradictory effects. Thus, David Halle’s (1984) America’s Working Men, an ethnography of chemical factory workers in the late 1970s, owes much of its power to its ability to move between the work and the household. As Halle shows, although the work is blue-collar factory work, the people he met lived a very different life outside the factory. Working in a relatively well-paid unionized job, their salaries allowed them to buy houses in neighborhoods that were mixed working class and lower white-collar. The wives of the factory workers often worked office jobs and saw themselves as occupying the middle class; their pastimes were both “working-class” and “suburban middle-class” pastimes, blurring white- and blue-collar identifications and patterns of consumption. This is clearly a case in which shadowing provides a richer picture of both meaning and personhood (as stressed above). However, by moving from the workplace situation to the home, Halle simultaneously provides a portrait of a point in American history of class relations, the tail end of a time in which workers could maintain a “dual life”—hovering between working-class and middle-class identifications. This dual life, as Halle shows, is precisely what defined these families—staying with one situation would not have simply been partial but distorted.

In short, then, by opening up novel axes of variation that are hidden from any one vantage point, shadowing allows ethnographers to provide a more compelling account of the general condition they attempt to describe. By showing variation in the “effects of a cause” (see, e.g., Morgan and Winship 2007), the ethnographer ends up shoring up the causal narrative they construct at the most general level that the ethnography often attempts to address.

Park Life: Shadowing in Action

To illustrate the utility of shadowing, we draw upon the first author’s ethnographic work. The project investigated the social world of a long-standing pickup soccer game and accompanying sociable interactions in a Los Angeles public park—and through it the experience of immigration of the Latino working-class men who mostly played in the park. Inspired by the tradition of streetcorner ethnography, Trouille conducted initial fieldwork primarily in the park. And, over five years of participant observation, he generated a wealth of data about the meaning and use of the public park.

As we show below, although Trouille spent years in the park, moving beyond the situation deepened his understanding of park life and its place in his interlocutors’ social worlds. Although it quickly became obvious that park interactions were much more than an isolated world of leisure, moving beyond the situation allowed the author to see how the park was connected to other aspects of the men’s lives, thus also shifting his understanding of the site itself. Moving beyond the situation also presented new axes of variation, which further deepened his understanding of local constructions of meaning in action that were invisible from one point of view.

First, however, a brief note on what shadowing involved. Over time, after he had spent almost two years in the park, Trouille gradually accompanied the men away from the park. This mainly involved following the men at work on home improvement jobs and when socializing away from the park, primarily in bars, restaurants, and other public settings. He also accompanied the men on their daily rounds: commuting to work, buying supplies, going to court, and picking up children from school. Shadowing opportunities grew out of the park but also required Trouille to develop new relationships and points of entry. Some attempts were more successful than others, and the author was unable to observe men at work in restaurants, another principal source of employment, or spend extended periods of time with them at home or in their neighborhoods. As in every new overture in an ethnographic site, some “moves” were blocked—some men simply didn’t want him around in other situations, even if they were happy to spend time with him in the park, some situations were off-bounds. And yet, despite these challenges and limitations, shadowing proved important both to understand park life itself and to understand the men who came to the park.

Work at Play and Play at Work

First, when shadowing men at work on home improvement jobs, Trouille began to see how the park provided a setting in which men developed employment opportunities. Enmeshed in the informal economy and without access to formal credentials, credit checks, or insurance policies, the men depended on networks and resource exchanges to make ends meet. Park life thus served not only to bring the men together but provided a staging ground where they built reputations and gauged the trustworthiness of others—a pressing concern when referring and recommending in the informal labor market.

As the men learned to play together, they learned other things about one another. For example, beer money was a recurrent source of drama and disagreement at the park. In some cases, the men used levels of contribution as proxies for work character, even if the facts were not always clear. Araña, for example, was frequently criticized for never putting in money for beer, despite being one of the first to request one. Many men, including those without firsthand experience working, hiring, or recommending him, drew from his park demeanor to speculate about his professional integrity. As one example, when the author asked someone why he did not take up Araña’s offer of employment, he responded: “How am I going to work for him, he doesn’t have one cent for beer,” a comment that reflected less on Araña’s financial state than on his willingness to contribute to the group. The presumed relationship between the park and work also operated in other ways. In contrast to Araña, Motor was often cited for his generosity and commitment to park life when the men endorsed him as a boss or recommended him for work. For example, when Mike asked around about Motor’s carpentry skills—his landlord was looking for someone to replace a fence—several men referenced his dedicated management of a park-based soccer team in supporting his hire.

Much of the practical resources that emerged, however, were more a by-product of being together than the result of calculated behavior or strategic tests of character. For example, someone learned about a job opening after a restaurant worker was playfully urged to “go wash dishes” when he misfired a shot during the soccer games. The men also learned about each other’s work in everyday conversations and from a range of professional indicators, from work clothes to nicknames like “Locksmith” and “Policía.”

This, however, did not mean that leisure time was simply a preparation for work, a future resource for interaction and camaraderie in another situation. The park also served as something for coworkers to talk about while on the job, a point of connection and a source of entertainment that guarded against the drudgeries of workaday life. To take one example, the author joined two men in lively conversation about a recent arrest at the park for alcohol consumption, as they painted the exterior of an apartment building in the scorching sun. The men even called men they anticipated were in the park to see if the police had returned. As was often the case, the men also went to the park for their lunch break to see if they had missed any of the action, returning to work with fresh materials to pass the time.

The men explicitly anticipated these sociable connections on the job. For example, when Muñeco asked Cabro if he would help him install a sign, Cabro, who was sitting with a group of men at the park, joked: “You must be getting lonely.” As the two men headed to the parking lot, one of the men told Cabro to tell Muñeco about a goal he scored during the midday soccer games; another playfully requested that they bring back beer when they finished. In another case, a group of amused men speculated that Valderama and Mi Chavo would have a lot to talk about the following day at work, as they play wrestled on the grass. That the men would often drive out of their way to pick someone up at the park to help on a job—rather than hiring someone at a more proximate day laborer site—was also revealing of park-based ties at work.

Interactions in the park thus had a transcendent meaning that was usually not spoken of. Although players interacted with and assessed each other as players, they also interacted with each other as potential job providers and job seekers, as coworkers as well as co-players. A reputation, then, transcended its situation. As we noted in the third warrant above, people expect to move among situations and, as we show here, they also expect others to move with them. This, then, affects the way they understand each other’s actions. Moreover, it is not that “play” is simply a testing ground for work. Rather, people in both situations talk about, anticipate, and act in relation to the other. The swirl of fun and competition at the park, in particular, took on new meanings in the context of employment exigencies.

The intertwining of work and leisure was, in retrospect (though only in retrospect), to be expected. But connecting the two worlds allowed Trouille to say something more general about the men’s lives. As opposed to seeing networks as innocently there or passively reproduced, he discovered how migrants manufacture a basis and foundation for networking and making new ties in their everyday lives. Contrary to scholarly and popular images, many of the men did not arrive in Los Angeles with dense social networks at their disposal. Categorical attributes assumed to generate ties (such as ethnicity, class, or even hometown) were usually insufficient bases for sustaining old connections or for forming new relationships. By charting the conditions and contexts of interaction that gave rise to the men’s social ties and resource exchanges, a general set of pressures and processes across situations came into clearer focus, something that became crucial in the ethnographic work as a whole. The different situations, then—as we argue in the fourth warrant—became indicative of this wider emphasis on the making and sustaining of social networks in immigrants’ lives.

These considerations already show that action in the park had implicit meanings that were not always obvious—reputations transcended the park, being a “good participator” (or a bad one) had reverberations for other situations. But shadowing also shed light on the meaning of activities in the park by providing new, and unexpected, axes of variation that would have been invisible if the author remained in the park. Below we discuss one such set of meanings—how an intersituational analysis of violence contextualizes the meaning and organization of park fights.

The Park as a Place to Fight

That Nelson and Tico were arguing during the midday soccer games was not surprising, as the two men frequently quarreled on the field. Despite a history of problems, the two men had never traded punches, preferring to spar verbally. On this day, Nelson, who was almost a foot smaller and a hundred pounds lighter than Tico, was determined to settle the simmering conflict with physical violence. Yet, as soon as he moved aggressively toward Tico, several men blocked his path, urging him in Spanish to “let it go.” Tico, who at first seemed uninterested in fighting, soon demanded that the men “let him go,” adding that he was going to teach the “little boy” a lesson.

As expected, Tico’s taunts further enraged Nelson who aggressively tried to push past the men’s clutches. With growing calls from onlookers to “let them” fight, the men’s resolve weakened and Nelson burst toward Tico. The crowd quickly circled around the two men as they exchanged a messy flurry of punches and kicks. At first willing to let them fight freely, three men physically intervened when Tico picked Nelson up and prepared to slam him against a fence lining the field. In the commotion, Nelson fell to the ground and Tico delivered a quick succession of kicks to his head. As Tico moved to inflict more damage, several men forcefully pushed him away. Tico did not resist the intervention but loudly warned: “Nobody get involved, let him fight, he provoked me.” Nelson, dazed but back on his feet, made a halfhearted attempt to counterattack, but relented when his path was also blocked. As his demeanor suggested, the short-lived fight was over. Several men crowded around him, offering their support and prognoses. A few even congratulated him for his efforts, patting him on the chest and back, despite the defeat. Valderama, with his arm around Nelson’s, commented: “Don’t worry, you fought like a man.” Then, pointing to his bloody nose, he added: “That’s nothing, it’s over.”

Tico, for his part, walked in a circle, fuming yet triumphant. He was particularly angry at the men who stopped the fight and criticized his aggressiveness. Sensing that the crowd was turning against him, he reminded them that Nelson was the instigator: “He wanted to fight, everyone saw it. I told him not to.” When someone informed him that he had probably broken Nelson’s nose, Tico replied: “That’s what he gets for getting involved in the affairs of men (‘cosas de hombre’).” Tico’s summation, however, was far from the final word on the matter, as the men debated and dissected the latest park fight for weeks to come.

The fight between Nelson and Tico was one of forty-eight violent confrontations the author witnessed during his five years in the park. In addition, he heard about twice as many fights, including some that occurred during fieldwork and others from years before he arrived. He also witnessed and was told about numerous “near” fights that approached but did not cross the threshold into physical violence. Approximately two-thirds of park fights occurred in the context of the soccer games, the remaining third within the swirl of off-field interactions in the park. Some physical altercations were primarily the product of the moment—a hard clash on the soccer field or spontaneous disagreement—while others revealed a deeper history, the culmination of brewing tensions and past problems.

Interpretations of what constituted a fight varied, but the men frequently defined physical altercations in terms of “meter manos” (putting hands on someone), which the men acknowledged by warning others not to “touch me” and in seeking retribution for being hit. In contrast, the men rarely described an unanswered blow as a fight and the author never observed combatants attacking unwilling known opponents beyond an initial shove at the park. The vast majority of antagonistic altercations did not escalate into physical combat but hovered at the zone of face-saving bluster and bravado. For example, men would put the onus on others to “hit me” (pégame) or find respectable ways not to fight in the heat of the moment.

At first, the author found the sporadic violence frightening. And yet, to his surprise, the men talked about the park as a relatively “safe” place to fight, at least in terms of physical damage. This is not to say the men did not occasionally hurt each other—black eyes, scratched faces, and chipped teeth happened—but nothing so serious as to require an emergency room visit. Over time, the author came to appreciate the performative quality of occasional park fights. As in Nelson and Tico’s fight, he observed how violence in the park was controlled, monitored, and usually broken up before long. The men also anticipated a degree of restraint by their opponent, as their long-standing use of the park generated a regular community of interaction that shaped the form that violence took. That is, the men’s repeated interactions in the park—whether when they played soccer or when they socialized over beers—created an ordered and meaningful form of local life. Players knew each other well and were held responsible for their actions if they planned on returning.

Thus, although fight-based reputations were rarely unequivocal signs of strength or weakness, fights were strung into a history that the men used along with other information to evaluate themselves and others (which was especially important in light of their employment-based ties and anticipations described above). For example, some men were considered “hot heads” too dangerous to bring on the job because of past conflicts, whereas others forged deeper connections on the battlefield and in how they negotiated their postfight relationships. While it is hard to tease out the direct relationship between conflict and collaboration at the park, it was certainly not a barrier, as many prior foes subsequently socialized and worked together. That is, the same context that allowed for new ties to form at the park facilitated a relatively “constructive” kind of violence to boil over on occasion.

As a way into the intersituational analysis of violence, it is worth noting that the men themselves talked about park violence in relation to other contexts. For example, the men talked about the park as relatively “nice” and safe. This characterization contrasted with their depictions of decrepit and underresourced city parks, especially those closer to their places of work and residence. The park certainly did not resemble the street, another public setting that was occasionally the stage of violent interactions in the men’s lives, especially for those who frequented rougher sections of the region. Not only did park violence contrast with the cutthroat world of fighting described in studies of inner-city violence (see Anderson 1999; Contreras 2013; Jones 2010), but also with what the men experienced or anticipated away from the park. Many of the men shared stories of vicious beatings and robberies they fell victim to or heard about in other sections of the city, including where they lived, and claimed to avoid certain areas and situations because of the threat of violence.

However, it was only after the author followed the men on their nightly rounds, especially in bars and walking the streets late at night, that he observed firsthand how threats and behaviors actually differed outside the park—sometimes in ways that confirmed the men’s mythic tales of Los Angeles violence, but also in other, more surprising ways. As the author gradually discovered, park fights were one of a series of situations in which the men encountered and enacted violence.

For example, after hours of drinking in an infamous “cantina,” one of the park players, Locksmith, became embroiled in an argument with someone sitting at the bar. Although it wasn’t clear why, Locksmith had clearly angered the stranger who stepped off his bar stool and threatened to “matarte” (kill you) if he did not immediately leave. In the commotion, a large and increasingly impatient bouncer, who had already intervened on Locksmith’s behalf several times that evening, grabbed him by the back of his jacket and forcibly expelled him from the bar. As the author helped him up from the sidewalk, Locksmith told him that the man by the bar had a knife on him, which is why he did not meet his challenge. Locksmith also informed him that the same bouncer had pepper sprayed him the last time he was at the bar, which would have further decreased his odds of success. When the author naively recounted the story at the park, many of the men castigated him for going to the bar, especially with Locksmith, a known brawler. The message was clear: He was safe in the park, but not elsewhere, and he was fortunate to have escaped unscathed. Even more, as the people in the park put it, he was a fool to believe that bar fights resembled park fights.

In another instructive incident, the author was walking out of a liquor store with four men from the park around two in the morning when they encountered approximately ten young Latino men they later identified as “cholos” (gang members). The area was dark and there was limited room to pass along the crowded sidewalk. Tough stares and defensive postures were exchanged as they crossed paths. “Accidental bumps,” the trigger of many fights (Katz 1988), were narrowly averted. Immediately after passing them, the young men turned in their direction and shouted explosive warnings and insults in Spanish and English. The park men never broke stride and hurried to Chepon’s house around the corner. In the safety of his backyard, they nervously laughed about the incident.

In these two incidents, the men avoided confrontation. Faced with uncertain odds, at least in the two cases above, players assumed the worst and avoided potential harm.

The peril and uncertainty of these events began to illustrate what was different about park fights. In these tense instances of stranger interactions, it was unclear how violence would unfold. Even if it was not clear that the man in the bar confrontation really had a knife on him or whether the young men were in fact cholos—maybe the men were simply saving face when dodging conflict—anonymity made it all the more dangerous. The men simply did not know the risks, especially as the odds were not clearly stacked in their favor. There was not the same expectation of restraint and control that the men could count on in the park.

Some men even mocked park fights or local tough guys within a broader context of violence. For example, Brujas, who lived in a notoriously rough area of Los Angeles by MacArthur Park, loved to talk about the time Polo and Pisa Muerto visited him at his home. In contrast to their bold park demeanors, Brujas painted a very different picture of their behavior in his neighborhood: “You see Pisa here all big and bad but he was so scared he didn’t want to go to a [local] bar,” preferring to drink beers in his home. As for the usually verbose Polo: “He barely said a word, nothing but silence.” These stories were especially instructive for men who had less experience in settings and situations Brujas and others insinuated were more “real” in their violence than park fights and helped to confirm what the author suspected about the men’s shifting behaviors he observed firsthand.

Anonymity, however, cuts several ways. While it can be a source of danger and fear, it can also prove liberating and intoxicating. In the park, combatants usually negotiated the new meaning of their postfight ties within a context of joint pasts and presumed futures. More generally, conflict created a history of adjustments and attachments that sustained the men’s commitment to each other and the group. For example, when recounting their park friendships, the men often pointed to past problems, usually a fight, to explain how much they had been through and overcome with particular men. Recurrent stories and debates recycled and reinvigorated the men’s fighting pasts as well. Tico brought this starkly home to the author when he casually put his arm around him and joked: “You know I wanted to kick your ass when you first started playing here.” Fight stories and legacies also covered incidents from years in the past, providing “narrative payoff” for those who had not fought in years. Descriptions and evaluations of fights, however, were often a source of debate and division, which was precisely what made park fights so compelling and distinctive. The men anticipated their fights to be talked about, thereby infusing their decisions and actions with greater meaning and longevity. Yet, despite the drama that inevitably ensued, most men appeared willing to overlook past problems after they fought. For example, while Nelson and Tico publicly feuded for years, the author found that they managed to relate much more amicably after the fight described at the beginning of this section.

In contrast, a key seduction of anonymity, such as in bar fights or crowd violence, is the relative freedom and spontaneity that comes with fighting someone who is unknown, as risky as it may be (Conley 1999). While strangers also have to work together to create the type of fight they are seeking (Jackson-Jacobs 2013), consequences are generally not rooted in how they relate going forward. Without the prospects of future interaction, the perceptions of one’s combatant are far less important, whether on the winning or losing end of the exchange.

The author thus also witnessed and heard about the flipside of fighting strangers when shadowing the men away from the park—when they were the aggressors rather than those avoiding conflict. For example, when a group of men stepped out of an unmarked bar in the early morning hours, Motor confronted a man walking by them for seemingly unprovoked reasons. Motor repeated the provocative accusation, “que pedo?” (what’s the problem?), as he urged the unsuspecting stranger to make a move. That he was drunk and five of his friends stood behind him presumably encouraged him. As he probably anticipated, the man hurried past them without reacting to the challenge. While Motor’s predatory behavior would have likely received censure at the park, the men laughed and encouraged his abuse in this situation. Motor’s actions were especially striking, since he was usually a “peacemaker,” stopping fights in the park rather than provoking them.

Another incident Trouille heard about from several sources also illustrates what is potentially different about fighting strangers. Half a dozen men went to a bar after a steady afternoon of drinking beer at the park. Soon after their arrival, they picked a fight with two unfamiliar men. The confrontation spilled out onto the street. In a scene that was vividly recounted for months at the park, Beavis pulled one of the men out of a car—as, they claimed, he was trying to retrieve a gun—and repeatedly kicked him in the head, leaving him bloody and apparently unconscious on the pavement. Like the individual Motor taunted, the men remained strangers before, during, and after the fight. ¹

In these incidents, the men revealed an enthusiasm for violent behavior and a disregard for its possible consequences that the author did not observe or hear about at the park. While park fights started for seemingly silly reasons, they were not as arbitrary and aggressive as these attacks. For example, the author never witnessed someone at the park randomly pick a fight, as Motor did, much less an adversary who was clearly outmatched and uninterested in fighting. And, in any case, onlookers would have certainly intervened and not allowed such unilateral aggression at the park. Strangers, in contrast, promised a noncommittal relationship.

In fact, the only carnival-like brawl the author witnessed at the park involved a group of unfamiliar Egyptian men in town for an event. In this explosive confrontation, which spilled out of problems during an impromptu soccer game between the two groups, park regulars exhibited an uncharacteristic eagerness and abandon in using violence at the park. For example, one park regular struck an unsuspecting foe in the head with his clenched fist, an unprecedented “sucker punch” that continued past the point of “defeat” with a flurry of kicks. While at first treating this episode as an outlier to be ignored, given the violence he observed when shadowing the men outside the park, it suddenly fell into a familiar pattern. As the men recognized when they talked about it retrospectively, these contexts of action allowed for, even demanded, different behaviors and identifications than what the author generally observed in the park. Under the right circumstances, it appeared, peacemakers were capable of remorseless and unprovoked violence.

Park fights generally differed from these scenarios, whether anonymity was experienced as a source of fear or freedom. This is not to suggest that the men were perpetual victims or perpetrators of violence as they traveled through the city—these incidents were also rare—but to explain what was contextually unique about fighting in the park.

Shadowing revealed the uniqueness of park fights in two ways. First, the author’s understanding of the meaning of violence in the park changed as he moved among situations. Some fights were more anonymous than others; some violence initiated by the men was completely unprovoked, whereas in other settings conflict was avoided at all costs. These sets of semiotic relationships were far from obvious in the park.

Indeed, the axis of provoked/unprovoked violence did not even occur to the author while he stayed in the park, and the importance of anonymity, while obvious in hindsight, came to the fore much more powerfully as he witnessed intersituational variation. And, once he attended to these axes of variation, he realized that these shaped the ways in which men in the park talked about violence. As men moved among situations, these hidden axes of variation framed the meaning of violence by contextualizing it for the actors themselves.

Second, remaining within the situation, the author could have easily assumed that men in the park always controlled their violence and that the park provided men with the opportunity to enact a “safe masculinity”—where they could be both violent and physically safe. And although this was undoubtedly the way violence operated in the park, this wasn’t the case elsewhere. In contrast to such an image of sports and violence (see, e.g., Brooks 2009; Wacquant 2004), the examples above show how precarious such assumptions regarding the effects of one set of situations on another truly are. As situations changed, both the possibilities of violence and the selves that “park men” occupied changed with them. Blustering fighters suddenly became tame; peacemakers initiated unprovoked violence.

Conclusions

As we have argued, producing intersituational variation by shadowing interlocutors among situations holds some promise. While the warrants provide theoretical reasons for this to be the case, the final cash value of this approach can be seen in the empirical. This is evident both in the way such intersituational forays play out in previous ethnographic work, as it is in the case of immigrants’ park life and violence. It is not simply that shadowing provides a more extensive picture but that it provided a different one. To draw an analogy from painting, shadowing not only enlarges the canvas but also changes the details we draw.

Like every deepening relationship, shadowing is hard work. Initial contact is one of the hardest emotional challenges of ethnographic work. It is a point rife with insecurities and rejections. Once an ethnographer plants him or herself in a situation, it is easier to simply “return” to the scene.

Shadowing requires ethnographers to ask whether they can move to new situations, something that puts them up, once again, for rejection. It is probably for this reason that shadowing usually happens at the tail end of an ethnographic project, when a relationship becomes deep enough to transcend its situational underpinnings.

The challenges—both in terms of time and in terms of emotional vulnerability—are thus substantial, and in many cases, the payoff is not immediately apparent. Taken ad absurdum, it may seem that we simply push ethnography to give a “complete” picture of the social world the ethnographer sets out to describe—something that is both impossible and analytically nonsensical. In a short story, Borges tells of an attempt to make a “perfect” map, as big as the area it describes. The results are dismal: “The following generations (…) saw that that vast map was useless, and not without some pitilessness was it, that they delivered it up to the inclemencies of sun and winters. In the deserts of the west, still today, there are tattered ruins of that map …” (Borges 1999:325). There are no perfect maps, and there cannot be a perfect ethnography. In order to tell a coherent narrative, ethnographers must shave off aspects of the world they are enmeshed in.

For all that, we argue that the form of intersituational variation we emphasize here is important for diverse theoretical reasons and for a variety of ethnographic projects. It is important methodologically, as a way to temper assumptions about the ways in which one situation can be transposed upon others; it is important for semiotic considerations, as a way to understand local meaning; it is important as a way to understand people and their social worlds; and it is often a powerful way to construct and buttress overarching causal narratives. As such, it both provides ethnographers a way to buttress claims of the study’s validity and offers fruitful ways to think beyond the specific case. Like other forms of intersituational variation evidenced in the work of life-course ethnographers who follow actors for long stretches of time (Black 2009; Smith 2006, 2014), or ethnographers who follow objects over multiple sites (see, e.g., Haugerud et al. 2000; Tsing 2004), shadowing is a way to generate new forms of variation, and thus deepen our understanding of the social worlds we study.

Not all projects require shadowing. Some questions are neatly contained within the narrative arc of a set of unfolding situations; other questions span temporalities that make intersituational variation irrelevant for all but a handful of long-haul ethnographers. And yet, many projects would be deeply enriched by shadowing and its possibility should thus be, at the very least, a methodological and theoretical question ethnographers grapple with.