Seeing like a state athletic commission: Multi-case ethnography and the making of ‘underground’ combat sports

Introduction

How can ethnographers access and assess macro-sociological influences on everyday life? This article examines one strategy, Michael Burawoy’s (2009) multi-case Extended Case Method (ECM), which uses contrasting cases to bring analytic focus to states and economic systems. The turn to multi-case and comparative methods offers a promising way to ‘extend’ from micro processes to macro forces beyond Burawoy’s original formulation. Yet this development has also added fuel to longstanding criticisms that the ECM privileges structural explanation (Katz, 2004), imputes macro forces without empirical observation (Sanchez-Jankowski, 2002; Smith, 2005), and locks practitioners into theoretical frameworks at the moment of case selection (Tavory and Timmermans, 2014). The most recent attack comes from Desmond’s (2014) call for Bourdieusian ‘relational ethnography,’ where he accuses ECM practitioners of reifying places and groups when generating comparisons, and refusing to break with preconception because of the primacy of analytic theory.

In a fierce rebuttal, Burawoy (2017) claims that Desmond’s failure to look beyond the observable world means his supposedly ‘relational’ approach to eviction and exploitation (Desmond, 2014) cannot understand crucial structural relations with the state and larger markets, and his abandonment of comparative logic means he cannot comprehend causality. For Burawoy, others fail to break with preconception by eschewing theory and macro comparison, instead reproducing their subjects’ worldviews. Thus, Desmond’s is a ‘spontaneous sociology’ out of touch with the Marxist theory that could actually explain the ‘exploitation’ he sees on the ground. Yet is this impasse of position-taking the most productive use of these insights? Like similar debates over comparative versus narrative history, such struggles may be turf battles as much as intellectual innovations (see Brubaker, 2003; Steinmetz, 2004). I suggest that we are better off putting these stances in dialogue, drawing on the strengths of the multi-case ECM but addressing the dangers identified by critics.

In the spirit of Burawoy’s research program, I aim to amend and extend his multi-case approach – first through critical exegesis, and then by using the method and working through points of tension. I offer my own comparative analysis of non-sanctioned combat sports in different US states, each existing under different regulatory and market conditions. I leverage comparison at the unit of state regime to better see ‘the macro’ shaping everyday interaction. The comparative puzzle is why one ‘underground’ fight group utilizes a referee-imposed rule set and the other opts for a self-governance and honor system. Here I engage macro and meso-level theories of how divergent state regulation creates niche markets, definitions for organizational identity construction, and degrees of autonomy from state and market.

A secondary puzzle emerges, however, in my state and market-centric explanation of these contrasting combative groups. Why does each organization emphasize the presence of ‘the state,’ moving fight events through different locales to avoid regulators, when neither is actually subject to investigation? Here criticisms of the ECM seem relevant: a ‘structural’ comparative case design and theoretical frame might make it too easy to impute supposed macro effects. Yet far from suggesting there are no ‘structural relations,’ the absence of the state athletic commission requires a different understanding of the micro and macro connection, one better suited to tools outside of the ECM orientation. Drawing on ethnomethodology and recent culturalist interpretations of the state, I show how a discourse of ‘the state’ acts as a resource for self-definition and problem solving in the two organizations, shaping action and local relations.

Ultimately, I show how different approaches to ethnography disclose different puzzles and highlight different insights from the same corpus of data. Embedded in these methodological distinctions are theoretical ones, such as different ontologies of ‘the state,’ but these need not be incommensurable. With a gestalt switch between multiple approaches, ethnographers can see crucial micro-macro relations in their data that are deemphasized in competing framings. Engaging in a series of pivots and re-casings (Timmermans and Tavory, 2014), my empirical contribution is to show how macro-sociological factors set conditions of possibility for relatively autonomous local practice. My methodological contribution is to demonstrate that the apparent binaries – comparison versus narrative, theory versus empirics, structural relations versus local relations – are most productively addressed not through a priori position taking, but by trial and error in the practice of research.

Seeing ‘social forces’ through macro comparison

Reflecting on his career, Michael Burawoy (2009) proposed a ‘multi-case ethnography’ as an alternative to the ‘multi-sited ethnography’ developed by anthropologists of globalization. Multiple sites could illustrate the flows and connections between locales that earlier anthropological forebears had ignored by taking villages or cultures as bounded and isolated entities, but this did not offer much room for causal analysis of differing outcomes. Burawoy, responding to criticism that his own claims about capitalism required data beyond a single Chicago shop floor, began practicing a distinctly comparative approach to multi-sited ethnography. To facilitate comparison he suggested reconstructing sites as analytic cases, bringing theoretically relevant distinctions into relief. Regarding his initial forays into Hungarian factories as a comparative case for his Chicago work, he wrote:

Constituting distinct sites as cases of something leads us to thematize their difference rather than their connection, which, then, poses questions of how that difference is produced and reproduced… Instead of the connection of sites to examine networks or flows, we have the comparison of cases constituted with a view to understanding and explaining their difference. (2009: 202)

Desmond (2014: 559) claims this means a privileging of ‘the macro over the micro, seeing the macro as that which is global, causal and historical and the micro as that which is local, descriptive and current.’ Burawoy responds that he does not privilege the macro, but his ‘structural ethnography’ (2017: 3) does point to the ‘underlying limits of transactions, limits set by forces beyond the field site that can only be explored with theoretical frameworks and comparative logic.’ ECM practitioners ‘highlight some aspect of the situation under study as being anomalous and then proceed to rebuild (rather than reject) that theory by reference to wider forces at work, be they the state, the economy, or even the world system’ (Burawoy, 1999: 6). Thus, although ‘structure’ is a many-sided concept (e.g. Porpora, 1989) ranging from objective positions in relation to the means of production and other fields of social space, to localized structures of ‘negotiated order’ (Strauss, 1978), and cultural structures, Burawoy’s orientation is to structural features of political economic arrangements, e.g. command versus market economies, and variations in state policy.

Thus for Burawoy ‘structure’ is beyond the field site, constraining on action, and often invisible without theory and comparison. What his ‘structural ethnography,’ macro-oriented theory, and comparative logics allow is for ethnographers to perceive the apparent impact emanating from states and economic systems. Consider again the analytic ‘casing’ of different empirical micro-interactional worlds as types, such as ‘hegemonic’ versus ‘despotic’ factory organization. If the cases share all major features but differ in, say, the type of state in which they reside, the researcher can begin by investigating whether there are mechanisms linking the macro-comparative factor to the observed micro-difference.

In simplified terms, this multi-case ethnography creates a formula for generating puzzles that begins with a how question: how does a given social process work in these two different settings? This leads to a why question that becomes the empirical puzzle: why do these two sites differ? The theoretical puzzle is derived in conjunction with this by pointing to singular existing accounts of a social process, and showing that there is variance in how it functions across cases. Thus solving the empirical why of difference sets up a reformulation of the existing theory. This opens myriad analytical possibilities and constitutes the innovation of the multi-case ECM.

Consider Ching Kwan Lee’s (1995) work on gendered labor regimes in two factories in Southern China, separated by the Hong Kong/Mainland border but otherwise similar in ownership, management team, and the actual products produced. Lee asks the interactional question of how women workers are controlled in the two sites, but does this primarily to set up her why question: ‘Why do distinctive regimes, “localistic despotism” in Shenzen and “familial hegemony” in Hong Kong, emerge in the two factories?’ (1995: 378). Notice the shift in focus from the shop floor world as a thing that has explanatory power, to interactional patterns as the thing to be explained. Lee’s work is exemplary, and demonstrates the ECM’s use of empirical anomaly to extend theory. She shows that Burawoy’s (1985) account of hegemonic factory regimes, which posits that state welfare provision and industrial regulation discourage coercion, cannot explain hegemony in the Hong Kong factory. Instead, she shows how Hong Kong women’s structural position in a labor market with declining industrial job options necessitates compliance.

The multi-case logic becomes more pronounced in the projects of later Burawoy protégés, such as Sallaz’s (2009) work on a Las Vegas and Johannesburg casino. The Labor of Luck begins with an apt criticism of Erving Goffman’s writings on gambling, arguing that the interactionist’s universalistic proclamations on card playing do not hold up to comparative scrutiny: card dealers and players in the two casinos Sallaz studies in fact act quite differently. Like Lee (1995) and Burawoy (1985), he identifies a contrasting hegemonic and a despotic labor regime, and argues that

… differences in micro-interaction orders must be traced to differences in social structures. Goffman, the quintessential ‘sociologist of small-scale entities,’ never ventured beyond the casino floor to make sense of his ethnographic findings. But to explain the differences between Silver State Casino and Gold City Casino, I had to extend out to the larger societal context of the United States and South Africa. Drawing upon new approaches in economic and political sociology, I show how state systems for regulating service industries structure the production and consumption of the gambling ‘product.’ (2009: 5, emphasis in original)

In Burawoy’s (2017) response to Desmond (2014), he highlights Sallaz’s sustained engagement with local-processes, showing that ‘structural ethnography’ need not privilege the macro over the micro. Yet while Sallaz’s presentation of data is multi-level and nuanced, his analytic summation is quite literally top-down: consider this graphic with the arrows in his model pointing only in one direction, from political field, to market, to managerial habitus, to position-taking, and finally experience (from Sallaz, 2009: 200; permissions from University of California Press).

This points to a risk of prioritizing structure in a political economic key. As Eliasoph and Lichterman (1999) have argued, the ECM often neglects how culture interpenetrates structure, or how cultural dispositions impact structure. The multi-case vision potentially exacerbates this by selecting for economic or state ‘structural’ explanations in the case design. Despite Sallaz’s interest in gambling consumption alongside production, it takes a backseat. What is the role of local cultures of risks and Goffman’s players? Sallaz could reconstruct Goffman’s theory of player risk as it diverges in cultural context, extending out to cultural structures. Instead, he looks to reconstruct Burawoy’s production regime theory. This pushes him to look ‘up’ the chain of actors and to the political and economic fields, offering a major contribution in one direction while simultaneously sidelining Goffmanian theory.

What would Sallaz’s book have looked like if he’d focused more on ‘structure’ in the micro-interactional sense of continually negotiated order or ‘cultural structures’ a la Eliasoph and Lichterman (1999)? Might he account for the way player habitus, local cultures of risk-taking, and shifting consumption and demand dynamically interact with dealer practice, management, and capitalists who then seek lobbyists to respond to regulators? Might he have filled in his graphic with upward arrows, or deconstructed ‘substantial’ places like his casinos or actors like ‘the state?’ Timmermans and Tavory (2014) write that ECM comparative case design ‘tethers researchers to the theoretical framework they began with… sacrific[ing] the flexibility of qualitative research’ (p. 19). It is notable, then, that both Sallaz and Lee end up extending Burawoy’s theory of production regimes, rather than numerous other potential options. This points to the advantages and drawbacks of working within a strongly delimited research program.

My claim is not that the multi-case ECM prevents Sallaz from extending a Goffmanian theory of player risk, that his analysis is wrong, or that top-down explanation is endemic to the method. What is worth noting, however, given Sallaz’s professed admiration of Goffman and the ECM’s concern with structuration, is that his conclusions regarding ‘casino capitalism’ ultimately privilege ‘structural relations.’ In the next section I offer my own comparison of underground fight sports to show how the multi-case formulation can be complemented with subsequent pivots that do not so much undermine but nuance the goal of the ECM: a better theorization of how the macro and micro actually connect.

Building the multi-case empirical puzzle: Methods and logic of case selection

Here I compare two non-sanctioned sport-fighting events that differ in how they regulate fight interactions and ascribe status. Why do distinctive regimes, ‘rationalized rule enforcement’ in New York and ‘honorable self-regulation’ in Los Angeles, emerge in the two events? ‘Rationalized rule enforcement’ refers to a system of order that grants an officially recognized referee the power to enforce fighter compliance with a rule set, known to all participants beforehand. ‘Honorable self-regulation’ refers to a system that removes referees and official rules, instead using an unwritten code of proper comportment to limit injury. Each event is a non-sanctioned, quasi-legal combat circuit where participants compete without pay, organizers claim to make little money, and group discourse proclaims the activities more ‘extreme’ than professional cage-fighting. I selected these cases as they are two prominent and respected ‘underground’ fight events, known to fight fans as an alternative product for consumption, and to fighters as an alternative venue for participation. The apparently crucial structural difference is that one operates in California, where the State Athletic Commission sanctions thriving amateur and professional cage-fighting scenes, and the other is in New York, a state that until recently banned the sport.

As articulated by some practitioners of ECM multi-case ethnography, analysis begins by identifying similar cases that vary in regards to their legal-institutional-economic environments, and also vary in their interactional outcomes. I begin by detailing how the interactional differences in the fights appear to be tied to the ‘social forces’ of the differently regulated markets, with each event defining itself in relation to the kinds of combat sport entertainment available in its respective state. Following Sallaz (2009: 5), ‘differences in micro interaction orders must be traced to differences in social structures.’ I therefore sought to explain the different interactional fight orders in light of ‘how state systems for regulating (combat sports) structure the production and consumption of the (fighting) “product”.’

To do this, I observed the New York events and trained with participants, watched and fought in the California events, and theorized their relevant interactional differences as analytic cases. Then I interviewed event organizers in each case, and investigated the history of sport regulation in the varying regions to try and understand why the cases looked so different. I read widely across literatures throughout the project, first on theories of how state regulation affects markets and organizations, second in the sociology of sport, social control, and embodiment, and third on theories of the state. By approaching my data with the ‘broadest theoretical base possible’ (Timmermans and Tavory, 2012: 180), I opened myself to both theoretical ‘anomalies’ within a research program (Burawoy, 1998) and ‘surprises’ against a backdrop of disparate knowledge (Timmermans and Tavory, 2014). All names of events, groups, and individual people are pseudonyms.

The extreme combat tournament: Rationalized rule enforcement

Beginning in the mid-2000’s, the Extreme Combat Tournament (ECT) started hosting what they describe as ‘underground’ cage-fighting matches, eventually gathering a considerable local and internet following. Rotating through the neighborhoods of New York City, fans only learn the location through word of mouth. In the following I describe my initial observations of the ECT and specifically the interactional content of the fights themselves.

An informant texted me the address for the secret location and told me not to post it online. I complied, informing only one of my training partners who expressed curiosity about the level of talent. We arrived at a small boxing gym in an outlying neighborhood, and as we paid our entrance fee a crowd milled about in the street. My training buddy recognized one of that night’s fighters, a large West African heavyweight named Kwame, as a member of his Brazilian Jiu-Jitsu gym and went over to say hello. Kwame had plans to fight professionally. I spoke with another young competitor who worked maintenance for the city, but wanted to pursue fighting. As we settled into our plastic seats around the ring, a referee in a black and white outfit and a promoter came out to announce the fights. One of the first matches was between Kwame and an enormous Latino man with a shaved head, who stared intensely across the ring.

At the opening bell they circled each other, and Kwame quickly struck the man with a clean punch, dropping him to the floor. As the man lunged for a wrestling takedown Kwame sprawled his legs back, dropping his full weight on him and raining down blows to the man’s face. The referee intervened, ending the fight and declaring Kwame the winner. Kwame stood, hands over his head in victory. My friend, a high level Brazilian Jiu-Jitsu competitor, told me that Kwame’s grappling needed considerable work. Given how easily he evaded the man’s takedown, he opined, neither had much of a chance against elite competition.

Monroe explained that he allowed fighters to agree either to ‘old school rules’ or the Unified Rules of Mixed-Martial-Arts, the rule set typically used by sanctioning bodies. As a fighter, Monroe explained that he prefers ‘old-school’ rules, as the Unified Rules has eliminated many effective techniques. As described in the above vignette, fighters appeared to know and follow the agreed upon rule set, such as eschewing effective yet illegal blows like strikes to the back of the head. When Kwame began to overwhelm his opponent, and the man was not ‘intelligently defending himself,’ the referee intervened. For this reason I refer to this form of practice as ‘rationalized rule enforcement,’ akin to sportive MMA.

Indeed, when I attended these ‘underground’ fights, I was struck by the fact that they looked and felt essentially identical to sanctioned MMA. Fighters divide by weight class, may use the bouts in their amateur record, and the referee enforces rules. Fighters aim to finish the opponent, or win based on points, and typically cheer for themselves. What made these fights ‘underground’ was that observers talked about the illicit nature, and organizers stated the need to elude the authorities by moving the event to different venues throughout the city.

The Reality Fighter Association: Honorable self-regulation

Beginning in the mid-1980s, the Reality Fighters began a ‘no rules’ weapons fighting tournament that now occurs biannually in Los Angeles. After seeing the fights online I traveled to California to conduct interviews, not knowing that I would later become a member of ‘the tribe.’ In the following field note summary I recount my first time observing a ‘Reality Fighter Assembly.’

On a sleepy Sunday afternoon I entered an athletic gym in a suburb of Los Angeles. I milled about with a group of fans, each of us paying a small fee to view the fights. Several fighters consented to interviews, and told me of their martial arts backgrounds. I first watched in disbelief as fighters walked to the mats holding wooden sticks, metal chains, and what appeared to be bladed weapons. Robert Canyon, the mutton-chopped man who acted as both announcer and ‘ringmaster,’ was dressed in cutoff jean shorts and a matching denim jacket. He walked around menacingly with a large, knotted stick doubling as a cane, staring around the room before saying, ‘the ritual is about to begin.’ The crowd immediately fell silent, and he explained the rules of order, or more specifically, the lack of rules. ‘No judges, no referees, no trophies. There is only one rule… be friends at the end of the day.’

As fighters sought partners I noted couplings that disturbed my sportive disposition. Heavyweights with lightweights, old versus young, and even fighters wielding different kinds of weapons left me with a slight feeling of nausea. While I stared around the room nervously all the participants seemed to be engaging in friendly small talk, forming a line that moved them up to the competitive space. When one of my first interviewees came to the front for a stick on stick fight, I was excited to see what his years of empty hand and weapons training would amount to. Justin was a martial arts instructor from the South, who loved kickboxing but believed most sport fighting was insufficiently realistic.

Blitzing his opponent with kicks and stick attacks, Justin eventually clinched and hurled the man to the ground with a Judo hip throw. He reared back with his stick in hand, as if to deliver a crushing blow to the opponent’s skull, but then slowed his movement and whizzed his weapon a few inches away from the man’s head. The audience cheered, and as the opponent acknowledged defeat, the men embraced.

In the 1990s it seemed the RFA’s fate might be tethered to the pro-circuit, as the group was approached by a major MMA sports company that wanted to promote weapons fighting. The RFA was eventually rejected as ‘too extreme’ for television, however, and with the institutionalization of the Unified Rules, the practices became more explicitly separate. Lacking an official referee and rules, the events are predicated on a shared understanding of a ‘code’ summed up in the phrase ‘friends at the end of the day.’ Participants prefer not to ‘finish’ fights, for instance, and will refrain from harming an already incapacitated opponent.

In my fieldwork and participation I observed fighters inquire after opponents’ wellbeing mid-fight, and debate whether certain attacks should be avoided, although few actions were outright banned. Strikes to the back of the head were considered ‘unfriendly,’ for instance, leading to an informal ban. Unlike in professional fighting, where athletes risk disqualification and the loss of money for banned attacks, RFA fighters do not stand to suffer any material losses from violating this ‘rule,’ but generally still avoid it. That said, I also observed attacks that are strictly illegal in Unified Rules MMA, such as knees to the head of a downed opponent and intentional strikes to the groin.

Fighters I spoke with further confirmed that while they could participate in standard MMA, and some did, many came to the RFA for something more ‘real,’ because MMA now had ‘too many rules.’ Others, especially older fighters, explained that it was unrealistic to continue participation in high-level sport fighting, but the RFA presented an event where they could still hone their self-mastery and martial skills. I occasionally heard Reality Fighters disparage sport fighters for simultaneously being too constrained (because they utilize rules and a referee) and dishonorably violent (because they rely on referees to decide when an opponent has had enough), emphasizing a strong moral distinction between their event and the more popular sport of cage-fighting.

Why to fight with(out) rules: A case of resource partitioning?

Why do these two forms of regulation emerge in the two settings? In each case participants make reference to the overly regulated nature of professional fighting, and claim that their practices resist this. To understand these reactions to the Unified Rules of Mixed-Martial Arts, it is useful to examine the history of sport regulation. This rule set was developed in reaction to widespread public condemnation of the Ultimate Fighting Championship (UFC), the earliest high profile limited-rules event. With Senator John McCain famously deriding the sport as ‘human cockfighting,’ and a groundswell of political pressure to ban the UFC, cable companies began refusing to televise events in the mid-1990s. Promoters came to believe that the key to financial survival lay in convincing sport regulators that the violent spectacle had become a rule-bound sport (Maher, 2009). The Unified Rules, which specified close to 30 fouls and instituted weight classes and gloves, was key to the first cases of sanctioning.

On the defensive from political critics who had impacted their ability to profit, promoters wrote these rules and demanded regulatory oversight so as to re-open the market. In theoretical terms, a political demand for a ban led to a shaky market environment, which the industry responded to by lobbying for reform. In California, the state sanctioning succeeded. In New York, the ban was still in effect while I conducted fieldwork. What effect do these contrasting state environments have on local markets for fight products, and in turn on the practices?

As Peterson and Anand (2004) summarize in an annual review on the ‘production of culture’ perspective, the state, via laws and regulation, provides a backdrop against which cultural producers develop and maneuver in the market. In a classic example, Griswold (1981) described how 19th-century American publishers preferred selling British novels about domestic manners, for they were not required to pay copyright fees as they did with American authors. Those American writers turned to specialty novels about such themes as men fighting the forces of nature to gain a market share, until the copyright law of 1909 put them on equal footing with British authors, and many abandoned the forces of nature theme. Where literary critics attributed the differing forms of novel to cultural differences between the British and Americans, Griswold argued that this apparent cultural difference was in fact an effect of how laws and regulation created niche markets.

A further explanation can be found in Carroll’s (1985) notion of ‘resource partitioning,’ whereby competition among large generalist organizations opens niches to specialist organizations. Since contemporary consumer industries are able to support both distinctive and general market niches, a minority of consumers will seek to align themselves with specialist groups against the oligarchic mass-industry. The American beer market has come to support a large number of microbreweries, for instance, alongside the dominant major brands. What can the production of culture perspective and resource partitioning tell us about fighting? Consider the following top-down explanation of fight practices relative to the state and linked market niches.

In the case of New York, the ban on professional fighting prevented generalist groups from tapping the market for watching live fights. The lack of generalist organizations means a non-innovative, generalist product like sportive cage-fighting can become an ‘underground’ specialist phenomenon. Consumers are of course aware of generalist firms like the Ultimate Fighting Championship, and may consume pay-per-views or other products, but also turn to events like the ECT in order to see live fighting. Although Monroe offered the fighters the option of using ‘old-school’ rules instead of the Unified Rules, the practice largely resembled contemporary regulated MMA because that is what fighters prepare for and what fans want to see. The ‘rationalized rule enforcement’ of the ECT, then, appears to be explained by basic institutional isomorphism: organizers and participants have mimicked the existing logics and are simply awaiting legalization, or will eventually participate in sanctioned events. Thus, it is the state and regulation that influence the specific dynamic of generalist and specialist firms and their products. This direct emulation of professional MMA is hardly an accident, for the event is designed to provide fights for competitors and fans that are simply unavailable in the area.

In the case of California, the presence of generalist organizations in the form of state sanctioned cage-fighting creates a niche for an ‘extreme’ weapons sport. If they moved closer to the rule sets of MMA, there would be nothing to distinguish the group. In a state glutted with cage-fighting, the RFA distinguishes itself with ‘no rules,’ which emphasizes both extreme violence and danger. ‘Honorable self-regulation’ appears to be defined explicitly in contrast to sport fighting. If the ban in New York State leads the ECT to sell itself as MMA and therefore underground, the presence of generalist organizations in California leads the RFA to distinguish itself by being more extreme than MMA.

This explanation of regulatory impacts on culture and resource partitioning shows that ‘underground-ness’ only exists relative to a particular legal institutional environment. The combination of these two theoretical perspectives is illuminating, but as I will show, potentially obscures other relations of the micro-worlds to the state and market. For where these theories point to direct pressures on organizations – for instance, paying copy-write fees or fearing an audit – few such direct pressures exist on the fighters I studied. To actually trace out the impact I turned my attention to how fighters and organizers address practical problems related to state intervention. I found that each organization addressed the question of the state most directly by deciding where to hold fights, as changing locations could offer respite from intervention. Yet this soon alerted me to a missing piece for my account of macro influence, for when I looked for the state I could not actually find it.

Where to fight without rules: The (non) action of regulatory agencies

In a perceptive discussion of micro-macro linkage, Dorothy Smith (2005) argues that ethnographers must pay attention to the moments when actors themselves come to experience larger forces of social structure or power, rather than impute it when it is not visible. Contrasted with the extended case method’s attempt to see social forces through theory, she notes that in her approach, ‘Those dimensions of “the system” that for Burawoy emerge at the level of “wider structures”… are explored ethnographically rather than theoretically’ (p. 35). Thus, one has not seen ‘the state’ unless actual interventions occur. She explains:

The everyday world is hooked up to complexities of relations and organization of which such totally strange and apparently arbitrary interventions emerge, such as those experienced by gay men enjoying sex in a steambath when the police raid it… Such interventions are not happening all the time; they make visible what is ordinarily taken for granted, that the very organization of the everyday is permeated with connections that extend beyond it. (Smith, 2005: 40)

The ‘underground’ discourse and ostensibly illicit nature of each of the fight events belies the fact that state actors have little interest in shutting either down. A combat sport journalist familiar with the New York underground scene emphasized that the ECT is actually not against the law. The fights do not have sanctioning from the State Athletic Commission, but they are not officially banned either. He explained that despite the continued invocation of the term ‘underground’:

They’re ‘underground’ only in that they’re not advertised, not public, but they exist legally due to a loophole in state law… By enumerating in that statute, professional bouts or exhibitions, they are saying professional bouts are banned. Amateur bouts are not.

Yet like the organizers of the New York event, Canyon did not especially fear state intervention into any of his events. Despite the aim of staying ‘off the authorities’ radar’ the RFA is not in any substantive sense illegal, and the events have operated in gyms and public parks for years. Some participants recounted that when police noticed them carrying weapons in the park, they simply explained that the blades were dull and that it was for self-defense training purposes. Furthermore, they stated that if there was trouble from law enforcement, the group had many police officers that could negotiate the situation. Given that state regulation is not directly impacting the practices, then being ‘beyond the reach of the bureaucratic state’ takes on a different meaning. When I asked a senior Reality Fighter about this, he told me that the ‘underground’ talk is ‘pure marketing.’

In neither case, then, does ‘underground-ness’ imply the direct presence of the state. It is true that both the ECT and RFA are aware of the relevant statutes and history of sanctioning, and necessarily keep their practices within certain boundaries, but they go beyond simple compliance. They build ‘underground-ness’ from the ground up, in part through invoking ‘the state.’ In the ECT the practice of only announcing fights via text and word of mouth, moving through neighborhoods, and asking people to keep it secret, produces a feeling of excitement and distinction for the event. In the RFA, talk of the ‘bureaucratic state’ and staying away from the authorities belies the fact that there is little to no attention from officials, other than the occasional interest from police officers when fights are in public spaces. While I first thought each organization moved locations as an ‘art of not being governed’ (Scott, 2014), I found that they were in fact building imagined state governance through talk and the practice of moving fight location.

Several divergent perspectives help us understand this construction of the state in daily life. Following an ethnomethodological approach, ‘the state’ here is a situational achievement that is only activated in particular interactions, or via people who act as agents of a larger state, such as police or other street level bureaucrats (Garfinkel, 1996). In the Foucaultian tradition, governing practices create the illusion of a cohesive entity, elaborated in other parts of cultural sociology that focus on the production of boundaries between ‘the state’ versus ‘society.’ More recently, Maryl and Quinn (2016) have used examples of struggles over education policy to show how the ‘the state’ entity emerges in boundary work, and then becomes a resource and political weapon. Yet what my cases offer is organizations invoking ‘the state’ not to contest or expand actual state regulation, but to structure local projects tied to a discourse of state power and subsequent resistance.

Beyond simply noting that people are using ‘state talk’ for local purposes, or dismissing ‘structural relations,’ I take these as a new set of empirical puzzles regarding the state and fight commodity market. Why do event organizers move fights throughout the city, or invoke images of the 'bureaucratic state' and celebrate native land, if State Athletic Commission regulators are not actually that concerned? And if this is ‘pure marketing’ for the RFA but little money is being made, and there are few consumers, what is the ‘extreme’ and ‘underground’ discourse marketing for? In answering these questions, I connect the insights from the comparative setup and ‘structural relations’ with the local pragmatic interactions and direct relations of everyday life.

‘The state’ as organizational resource

How should we think of the ‘macro’ or ‘structural relations’ if agents of the state are not directly intervening in the two fight practices? The empirical absence might lead to a facile dismissal of ‘macro’ forces in the ECT and RFA and rejection of the original comparative analysis, yet this surprise instead presents an analytic opportunity to forge a more nuanced account of micro-macro relations. Here we can trace out how actual state policy and market conditions in each case act as conditions of possibility for practices that also have their own logic. Given the concrete power of the actual state and market, the discursive ‘state’ can act as a meaningful organizational resource. What does the ‘state talk’ tell us in each case, especially as it relates to the original comparative question of divergent structural locations and rule sets? Here two projects, marketing and self-construction, bridge the gap between structural relations and on-the-ground practice or rules.

Discussion and conclusion

This article has aimed to further debates over micro and macro relations in ethnography. Although ethnographers across a range of traditions are interested in this linkage, there is a methodological question: how can we see the macro connecting to the actually observed micro? Michael Burawoy’s comparative extended case method has offered one controversial way to do this. My intervention has been to explore ECM multi-case ethnography through an application of the method, critiquing and extending it.

I have shown how the multi-case approach disclosed certain puzzles in my data that I would not have looked to without the comparison. This is the promise of a mixed comparative case design, where analysts select ostensibly similar cases but vary one major condition to identify causality. This typically means a convergence on meso-organizational form (e.g. factory, casino, sport event) but divergence in macro-location (e.g. state regime, economic system) to explain divergence in micro-everyday practice (e.g. management strategy, game of tips, fight rule set). It directs the ethnographer to see macro forces in micro action. This led me to an intriguing empirical puzzle: why does the New York-based ECT organize its ‘underground’ fights with a codified rule set and referee, while the California-based RFA organizes ‘underground’ fights with ‘no-rules’ and self-regulation? My initial solution combined theoretical accounts of resource partitioning and the production of culture to theorize how differential regulation can lead to very different definitions of ‘underground’ fighting.

Yet I also found a ‘surprise’ (Tavory and Timmermans, 2014) that was not just an ‘anomaly’ to the favored theory: each group moves their fight events around to avoid regulation while acknowledging that they are not actually being monitored. What explains their continued invocation of the state and authorities? Here I ‘pivoted’ and ‘refracted’ through a new framework. I argued that the celebration of ‘undergroundness’ related to two different kinds of projects. In New York, the fact that the ECT fights served as feeders into professional circuits meant that they had to balance the need to produce a generalist product (mainstream cage-fighting rules) while distinguishing themselves with sexy marketing rhetoric. Rotating fights through the boroughs offers distinction and builds excitement among fans. The RFA in California, on the other hand, has a relative autonomy from the market. It invokes the need to escape the ‘bureaucratic state’ because of a general cultural orientation toward American individualism, and the ‘no-rules’ fights similarly link to a desire to embody libertarian values. This is a story about participants’ identity construction, more than economic activity.

How can we bridge these puzzles and literatures? Consider a false dichotomy linked to straw-manned positions: are ritual combat interactions a product we explain in reference to ‘external’ factors like regulatory environments and resulting markets, or themselves an explanatory crucible where habitus and morally valued selves are formed through embodied rituals of collective effervescence (see also Abramson and Modzelewski, 2011)? In turn, is ‘the state’ a powerful structuring force through long chains of relations, or simply an analytic abstraction that is mobilized at a micro-level?

The answer in each case is both. The larger institutional environments enable and constrain forms of practice and organizational identity, even if indirectly, setting the boundaries in which various projects of careers or identity construction can become pragmatically possible. The false invocations of ‘state intervention’ only make sense within the structurally delimited possibilities that have rendered each fight organization ‘underground’ relative to their local ecology. Quasi regulation is indeed a precondition for the social psychological experience, but people then draw upon it in diverse projects not reducible to the larger environment.

To return to the debates that inspired this article, my findings have methodological consequences for how ethnographers should approach comparison. I have argued that Burawoy’s (2009) multi-case ethnography does offer a genuine innovation for helping analysts see macro influences. Yet the criticism that the approach can lead to theoretical rigidity with macro privileging (Tavory and Timmermans, 2014), imputing ‘structure’ without empirical observation (Sanchez-Jankowski, 2002; Smith, 2005), and ‘substantialism’ by reifying groups (Desmond, 2014) has some merit. I showed where Sallaz’s (2009) multi-case ethnography, for instance, runs the risk of a literally top-down model. Rather than reengaging Goffman’s social psychology of risk to get at the ‘upward’ structuration of card player demand on dealer and house, Sallaz looks to the state and economic relations to theorize ‘downward’ pressures. Had I focused largely on organizational position and failed to consider the needs of participants, I might have missed the ECT fighters’ goal of accessing professional leagues, and the RFA fighters’ celebration of embodied libertarian self-mastery.

On the other hand, a radically micro and relational perspective that eschews comparative thinking may miss the larger forces that are conditions for local interaction. Burawoy (2017) is correct that this may end up ignoring both structure and theory, such as Desmond’s invocation of eviction and exploitation that misses the Marxist history of that term, state analysis, and the wider housing market. For my study at hand, this would be to say that there is no ‘state’ aside from ‘the state’ that people mobilize in talk. This would be the position of some ethnomethodologists – the opposite of reifying actors and groups. Yet this would be untenable, for as outlined above, the conditions of possibility for local action are indirectly informed by the actual state via the construction of regulatory categories that shape market competition.

The methodological point is to recognize how different frames disclose and obscure different puzzles. What it means for an organization to be ‘underground’ is necessarily defined in relation to a broader political economic field of regulation – the kinds of factors that the multi-case ECM points us to. Yet both ‘underground-ness’ and ‘the state’ also become resources for local life projects that pull them in directions not easily explained by macro forces. When it comes to the disparate uses of these categories in everyday practice, we may need to reach for multiple tools that are not easily integrated a priori, but become truly productive in the gestalt of research practice.