How Migrations Affect Private Orders: Norms and Practices in the Fishery of Marseille

The major aim of this article is to examine how migrations affect private gover- nance, taking as a case study the Prud’homie de peˆche , a private order that has governed the ﬁshery of Marseille for the past six centuries. Scholarship gener- ally argues that social norms guarantee the efﬁciency of private orders and their ability to resist the arrival of newcomers. My data suggest that the Prud’homie has failed to accommodate social changes prompted by migratory ﬂows, not despite but because of its social norms. This paper suggests that social norms are not only powerful tools of governance for private orders, but also forces of inertia that can prevent these orders from accommodating social changes. research focuses on the emergence and evolution of judicial governance. This article is part of a book project on the limits of private governance to be published by Hart Publishing.

norms. The importance of private orders has been noted in various communities, such as diamond traders in New York (Bernstein 1992;Richman 2006), lobster fishers in Maine (Acheson 2003), ranchers in Shasta County (Ellickson 1991), whalers in New England (Ellickson 1989), traders in Mexican California (Clay 1997), and even organized crime (Skarbek 2014). Much of the scholarship on private orders holds a positive view of their functioning and social effects. Many scholars have expressed the view that the social norms of private orders are conducive to greater social welfare. For instance, on the basis of studies of Shasta County ranchers, New England whalers and "households," Ellickson famously suggests that private orders produce "welfaremaximizing" norms (Ellickson 1989;1991;2007). Bernstein argues in an important article on diamond traders that "extralegal rules" persist when they are "Pareto superior" (Bernstein 1992: 115). More recently, Stringham has emphasized the "superb track record" of private orders (Stringham 2015: 7) and Richman has offered a "positive theory of private ordering" (Richman 2017: 75). Others have taken a more nuanced view of the efficiency of social norms as compared to legal rules. For instance, Feldman argues that the tuna court in Tokyo provides a system of dispute resolution that is both fast and inexpensive, although it is based on formal legal rules and procedures (Feldman 2006). Yet others have challenged the empirical basis of studies that tout the merits of systems of private governance based on social norms. For instance, Kaden has exposed the "myth" of a "law merchant" or "lex mercatoria" based on commercial customs (Kadens 2012;. Others have challenged the existence of a private order of "Maghribi traders" operating in the eleventh-century Mediterranean (Edwards and Ogilvie 2011;Goldberg 2016).
What is conspicuous, however, is the relatively limited amount of research concerning the occurrence of social changes within private orders. This situation is in part due to the fact that this scholarship usually adopts a "static" view of these institutions, providing snapshots rather an evolutionary account of their existence (Aviram 2011). In this article, I examine how private orders evolve when confronted with migratory flows, a key feature of our global world and the object of thriving scholarship in sociology (see, e.g., Paul 2011;Portes 2010;Portes and Manning 1986;Portes and Shafer 2007;Tarrius 2008). Sociologists and other scholars of migration have investigated not only how and why individuals migrate-including the ways in which migration flows transform both origin and host communities and how these flows generate diversity-but also exclusion and discrimination in these host communities. Despite widespread interest in these subjects, scholars of private orders tend to disregard the effects of migratory flows on these institutions. In fact, much of the scholarship has traditionally assumed that private orders are well equipped to resist migratory flows. Scholars usually posit that private orders use a wide range of countermeasures to control outsiders, whether they come individually or in clusters (Axelrod 1990: 67;Ellickson 1991: 57-59). For instance, Ellickson mentions the case of Frank Ellis, a recent entrant into the private order of Shasta County who was subject to a smear campaign and violent self-help when he failed to abide by local norms (1991: 57-59). Axelrod argues that social groups with strong cooperative ties can withstand invasion not only from single individuals, but also from groups of individuals (1990: 67-68). On this account, private orders thus have the capacity either to reject outsiders completely or, in a more nuanced version of the same theory, to assimilate them into the social fabric (Clay 1997: 224). In other words, the robust social norms underpinning private orders can insulate them from the social changes generated by population inflows.
However, the literature also abounds with examples suggesting that private orders might not be so resilient when faced with the arrival of outsiders. For instance, Richman mentions the successful competition presented by Indian merchants to the close-knit communities of Jewish diamond traders in New York and Antwerp, as well as the impact of such competition on these communities (Richman 2006: 387, 411, 412). Slade suggests that strong social norms among the "thieves-in-law" (an elite group within the Georgian mafia) did not prevent the infiltration of other criminal groups into their terrain in the 1990s (Slade 2012). Clay argues that the waves of migration caused by the Gold Rush in California in the late 1840s led to the "eventual collapse" of the system of private governance established among local merchants (Clay 1997: 224-25). Varese shows how a mafia group was able to settle in a small Italian town characterized by a high level of interpersonal trust (Varese 2006). Engel notes the "pervasive sense of a breakdown in the traditional relationships and reciprocities" after the arrival of outsiders in a "small and close-knit community" in Illinois (Engel 1987: 555-57). The rich material presented in these studies still lacks a common theoretical frame capable of explaining how newcomers managed to bring about social changes in these communities.
This article seeks to contribute to the understanding of social changes in private orders based on a case study drawn from a centuries-old private order in the fishery of Marseille (France). This private order, called the Prud'homie de Pêche (hereafter "Prud'homie"), offers rich empirical evidence to test the idea that private orders are impervious to social changes brought about by migrations and to highlight the mechanisms through which these changes occur. In fact, migratory flows have been an integral part of the history of Marseille (Echinard and Témime 1989, Lopez and Témime 1990, Jordi, Sayad and Témime 1991, a city with an important port, situated at the crossroads of maritime routes connecting Europe with North Africa and the Middle East. I have focused on two historical episodes of labor migration into the fishery of Marseille, namely, inflows of Catalan fishers in the eighteenth century and Italian fishers in the twentieth century. The data that I have compiled for the purposes of this study has enabled me to explore the ways in which a longstanding system of private governance has evolved when facing migratory flows.
These data suggest that the strong social norms of the Prud'homie have ensured its preservation for several centuries, but have also been a factor of social inertia when it was confronted with evolving social practices. On this basis, the paper argues that social norms are the great strength of private orders, but that they can also be a powerful force of inertia leading to their demise when they are called to accommodate new social practices.

Background on the Prud'homie
This article explores the case study of the Prud'homie, a private order that has regulated the fishery of Marseille since the fifteenth century. I will begin by presenting my methods and data sources below, before providing a broad overview of the Prud'homie and my research question.

Methods and Data Sources
This research project is based on extensive empirical evidence that I have gathered over the past five years within the framework of a longitudinal study of the Prud'homie. The aim of this study is to examine the institutional changes that the Prud'homie has undergone over time, with a particular focus on the effects that episodes of labor migration may have had on this institution. For this purpose, I have collected data from a combination of three main sources: archival documents, interviews, and ethnographic research.
The Prud'homie collected its archives over the centuries and donated them to the local administration in 1933 (a decision that proved to be prescient before the destruction of the Prud'homie's building during the Second World War). In order to mitigate issues of selective deposit and survival (Lawless et al. 2016: 107), I reviewed over 2,500 archival documents, ranging from 1175 up until the present day, taken from several sources: the archives of the local administration to which the Prud'homie donated its documents (Département des Bouches-du-Rhône), the French national archives, the city archives of Marseille, the archives of the French navy, the archives of the Chamber of Commerce of Marseille, and private archives. I have listed and summarized each of the documents that I reviewed by chronological order in an original data set.
In addition to archival documentation, I interviewed 28 stakeholders (some of them several times) who have been active in different capacities in the fishery of Marseille over the past few decades: fishers, members of the broader community (spouses, fishmongers, the priest of their local parish, etc.), members of the Prud'homie, and governmental officers. 1 I supplemented these data with a series of field trips to Marseille, during which I built ties with local fishers, attended social events in their community (particularly religious ones), and participated in several fishing trips.
These data have allowed me to retrace the origins of the Prud'homie, assess its evolution from different standpoints (within and outside of the Prud'homie) and examine the effects of labor migration on its system of governance. Before presenting my research question, I will explain how the case of the Prud'homie offers a relevant example of private ordering.

The Prud'homie: A Case Study in Private Ordering
My starting point is Ellickson's observation that private orders usually arise in "close-knit groups," defined as "social network[s] whose members have credible and reciprocal prospects for the application of power against one another and a good supply of information on past and present internal events" (Ellickson 1991: 181;McMillan andWoodruff 2000: 2422). This analysis of private orders is also consistent with the idea that members of close-knit groups usually display significant levels of social capital that guarantee a high rate of compliance with social norms (Coleman 1990: 300-21;Piskorski and Gorbatâi 2017). In these "close-knit groups," individuals develop "strong ties," defined as a "combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie" (Granovetter 1973(Granovetter : 1361.
Before focusing on the social norms implemented by the Prud'homie, it is worth providing further details about its origins.
The birth of the Prud'homie occurred in 1431, when the fishers of Marseille agreed to elect every year four of their "best possible" peers to create rules and settle disputes arising in their fishery (DA 250E6). 2 In creating the Prud'homie, the fishers of Marseille were particularly committed to solving collective action problems that regularly arose among fishing crews. For instance, when different crews joined together to fish for tuna, there was considerable temptation for each crew to contribute fewer men or lower-quality nets to the common venture. More generally, fishing crews were tempted to concentrate their efforts on the best fishing locations, thus resulting in the depletion of resources at these sites. These fishing crews therefore sought to maximize their individual gains at the expense of the other fishers.
The Prud'homie emerged to solve these collective action problems by fostering social norms of long-term reciprocity. Specifically, it encouraged fishing crews to coordinate their efforts in order to avoid freeriding problems and preserve the resources of the fishery in the long term. To achieve this purpose, the Prud'homie developed a wide array of legislative, judicial, and police powers. For instance, the Prud'homie implemented "sequential rules" that allowed each crew to fish in turn at the best locations without jeopardizing the fishery's resources (DA 250E2). The goal was therefore to guarantee long-term payoffs by fostering cooperation among fishing crews.
The Prud'homie is not only a formal institution with elected officials, but also the connecting node of various overlapping social networks: a professional network, a neighborhood network, a religious network, and a familial network. The position of private orders at the intersection of various social circles or networks has already been noted in the literature (Ellickson 1991: 55;Richman 2017: 48-49). In sociological terms, private orders are embedded in the "multiplex" relationships of their communities (Gluckman 1955: 19). In the case of the Prud'homie, these relations are professional, spatial, religious, and familial. Every member of the Prud'homie owns a fishing boat and operates within the bounds of the fishery of Marseille (roughly delimited by a three-mile line off the coast extending from Carry-le-Rouet to Cape Morgiou 3 ). Hence, every member of the Prud'homie is a fisher, and membership in the Prud'homie has been a powerful identity marker for the fishers of Marseille. Even today, some fishers introduce themselves as being "from the Prud'homie" rather than being "from Marseille." In addition, the Prud'homie was physically embedded at the heart of a small neighborhood of Marseille named St. Jean, where the vast majority of fishers used to live (at least, until the neighborhood was bombed in 1943). St. Jean was composed of small and narrow streets where inhabitants developed a strong sense of belonging. A few streets in St. Jean bore the name of the Prud'homie and of its most prominent members (Bouyala d'Arnaud 1959: 92, 160, 164). In a letter dating from 1830, the Prud'homie described St. Jean as a place where "everyone knows each other" and where "one can easily call a meeting in order to discuss an issue that undermines the general interest" (DA 250E126 1830). Fishers in Marseille are also deeply religious-even "superstitious" according to the current priest of the fisher parish (St. Laurent church). The Prud'homie has actively exploited these beliefs by nurturing strong ties with the Catholic Church. For instance, the Prud'homie celebrates St. Pierre-the patron saint of fishers-in St. Laurent church at the end of June. It has also made numerous donations to the Catholic Church throughout its history. Finally, fishers in Marseille traditionally originate from a limited number of families that constitute true fisher "dynasties." All of the fishers whom I interviewed in Marseille (with one exception) had succeeded their fathers, who, in turn, had succeeded their own fathers before them. These dynasties of fishers can be quite ancient: one of my interviewees was the descendant of seven generations of fishers that I was able to trace all the way back to the eighteenth century.
Belonging to one or several of these networks does not guarantee membership in the Prud'homie, but members of the Prud'homie usually belong to all four networks. Because of its position at the intersection of these "multiplex" ties, the Prud'homie does not need to do much in order to police recalcitrant members. In fact, its position guarantees a high level of compliance with social norms. Just recently, a former member of the Prud'homie reported to me the importance of these ties in the community: "in Marseille, if you are an outsider, you are dead-you are estranged." For a long time, the Prud'homie could simply rely on these various ties by "naming and shaming" noncompliant members. For instance, in 1653, the Prud'homie summoned some of its members to attend the weekly assembly on Sundays, emphasizing the "great prejudice" that their absence caused to the community (DA 250E4 1653). The Prud'homie also had the power to exclude noncompliant members from the community. For instance, the Prud'homie expelled a fisher named Jacques Clappier for having insulted one of its members in 1677 (DA 250E4 1677). When social pressure did not suffice, the Prud'homie could also fine its members (although this power does not appear to have been used very frequently).
Remarkably, the Prud'homie has remained relatively well insulated from the reach of state powers, despite the high degree of centralization in the French legal system (Merryman and Pérez-Perdomo 2007: 20-23). Because it predated the emergence of a centralized state in France, the Prud'homie first developed its governance functions on its own and then lobbied hard in order to preserve them in the face of expanding state power. These lobbying efforts have been quite successful in securing its formal powers and, even today, state courts are not allowed to hear appeals against its judgments (a complete anomaly in the French legal system). Nonetheless, the Prud'homie no longer exercises the formidable powers it has on paper: it adjudicated its last dispute in the late 1960s and no longer creates or enforces rules. This paper will explore the reasons for its demise.

Private Orders and the Entry of Outsiders
Private orders seem well equipped to regulate the behavior of insiders, but less capable of preventing outsiders from penetrating into one or several of their networks. For instance, ranchers in Shasta County could not prevent Frank Ellis from settling on their territory and practicing his trade (Ellickson 1991). The Jewish diamond merchants of New York could not prevent the Palanpuri Jains from effectively competing with them and becoming a growing player in the Diamond Dealers Club (Richman 2006). Similarly, while the Prud'homie could easily ensure the compliance of insiders with its social norms, it could not prevent outsiders from living in St. Jean, attending mass, or even fishing in Marseille; it could simply make their life more challenging by denying them the benefits deriving from membership (for instance, the possibility to dry fishing nets on docks owned by the Prud'homie, or access to its tribunal). In short, entry barriers might, in fact, be more porous than is usually assumed. Even in a trading community, such as a community of diamond dealers (in which entry barriers are arguably easier to maintain than in a producing community, such as the Prud'homie, due to the bilateral character of trading), Indian merchants were able to operate in the shadow of the private order, before gaining a secure foothold deep inside of it (Richman 2006: 410). Thus, private orders cannot systematically deny "entry," defined as the successful operation of outsiders in one or several of the social networks in which the host community is embedded. As individuals successfully gain "entry" into one or several of these networks, scholarship predicts that the private orders concerned should be able to socialize or exclude these newcomers (Ellickson 1991: 57-58;Axelrod 1990: 67-68). This observation is consistent with the game theoretic account of "evolutionary stable strategies" that are resistant to "mutant strategies" (Maynard Smith and Price 1973). However, as we have seen, many studies suggest that close-knit communities might not be so resilient when facing the arrival of outsiders (Clay 1997;Engel 1987;Richman 2006;Slade 2012;Varese 2006). The goal of this paper is to further examine the processes through which private orders resist, or fail to resist, when outsiders gain entry into one or several of their networks.
Of the likely factors accounting for the evolution of social groups confronted with newcomers, the literature mentions the relative number of newcomers as a key explanatory variable. For instance, Moss Kanter argues that "proportions," which she defines as the "relative numbers of socially and culturally different people in a group," are critical in shaping the evolution of social norms (1977: 965). In her study of women in the workplace, she analyzes the incorporation of outsiders within a group in terms of "tipping points" (1977: 986), observing that women (or any underrepresented group) "need to be added to total group or organization membership in sufficient proportion to counteract the effects of tokenism" (1977: 988). She argues that minorities can evolve social norms-no matter how tight-knit the social groups are-when the numbers of these minorities reach a certain critical size. In a recent study based on experimental evidence, Centola et al. have complemented Moss Kanter's study by identifying "tipping points" for the evolution of social norms (2018). They argue that a critical mass of approximately 30% of individuals is sufficient to overturn an "established social convention" in a social group (2018: 1118). They even suggest that the "memory length" of individuals, a proxy for the robustness of social norms based on the number of past interactions, does not significantly alter these "tipping points" (2018: 1117).
However, the "quantitative approach" to social change does not account for cases where a tiny fraction of a community successfully initiates social change. For instance, Varese reports the case of a small criminal group that successfully infiltrated a community characterized by high levels of trust in Italy (2006: 424-33). Varese concludes from this case study that "long-term virtuous trends can be reversed relatively quickly," even when prompted by a tiny minority of outsiders (2006: 433). In addition, the quantitative approach does not draw a distinction between social norms and practices. For instance, the small criminal group identified by Varese affected the processes through which procurement contracts were awarded in the host community (2006: 424-33), but does not appear to have modified the deeper normative beliefs in this community. Similarly, a growing proportion of women in the workplace might not affect significantly-or with sufficient speed-its deeply ingrained social norms (Moss Kanter 1977: 988). Most authors agree that social norms are more than a recurring pattern of practices in a given society and that they also set expectations as to how social practices should be regulated (Parsons 1949: 75;Coleman 1990: 242). In other words, social norms are not a mere reflection of what is, they are also an expression of what ought to be in the eyes of group members. Although behavioral patterns usually coincide with what society regards as being proper or correct, these two elements do not always overlap. Accordingly, the extent to which the arrival of newcomers impacts social practices and/or norms in systems of private governance remains to be analyzed. The case study of the Prud'homie provides fine-grained material for exploring these questions.

Migration Inflows and the Prud'homie
My data focus on specific periods in the history of the Prud'homie during which it faced inflows of foreign populations. These inflows crystallized in two episodes of migration that unfolded over several decades: the migration of Catalan fishers in the eighteenth century and that of Italian fishers in the twentieth century.

The Arrival of Catalan Fishers
Catalan fishers arrived in Marseille in successive streams of migration from the 1720s until the early nineteenth century (Faget 2011;Echinard and Temime 1989: 93;Grisel 2019). On the basis of archival documents, I assessed the size of this community of foreign fishers. In 1826, the police department of Marseille conducted a survey of the population of Catalan fishers in 1826 (MA 18F6). It established a list of 117 individuals, the majority of whom (83 individuals) were born in Marseille. An exhaustive review of the fishing boats based in the port of Marseille between 1816 and 1818 provided evidence for the presence of 22 Catalan boats with 115 crew members (112 of whom were described as "Spanish") (AA 13/P10/3 1819). These empirical data show that the Catalan fishers represented 14% or 16% of the entire community (depending on whether this proportion is calculated on the basis of the number of crew members or of boats) approximately hundred years after their first arrival (Grisel 2019). Figure 1 provides a snapshot of the community and of its size immediately after the relevant time period (eighteenth century).
It shows that the Catalan fishers represented a nontrivial proportion of the total population of fishers in Marseille, but one that was inferior to the tipping point identified by Centola et al. (2018).
The settlement of Catalan fishers in Marseille was particularly difficult and generated many conflicts with local fishers united around the Prud'homie (Echinard and Temime 1989: 93). As they progressively operated in the fishery of Marseille, the Catalans played a classic freeriding game: they started fishing without paying the costs associated with the maintenance of the fishery. For instance, the Catalans refused to pay the tax imposed by the Prud'homie on the sale of fish and did not comply with its social norms. The Catalan fishers refused to abide by certain fundamental rules of the Prud'homie, such as the prohibition on fishing on Sundays and the obligation to submit themselves to the jurisdiction of its court on that day. In addition, the Catalans brought a fishing technique from Spain that was already practiced in Marseille, although less intensively. This technique, called palangre, consisted of floating a long line to which several smaller lines and hooks were attached. Figure 2 below offers an illustration of this technique based on the detail of an original map.
The Catalan fishers used the palangre in more intensive ways than local fishers and ignored the Prud'homie's regulations relating to its use. For instance, they used smaller hooks than what was allowed by the Prud'homie, resulting in the capture of smaller species and younger fish (thus impacting the fishing stocks more aggressively). They also used their palangres further offshore, which was made possible by their more sophisticated boats  (Faget 2011: 45-48;Grisel 2019). The Prud'homie was outraged by these fishing practices, particularly since the Catalan fishers were successfully competing with the local fishers by landing more sizable catches.
In response to their arrival, the Prud'homie combined strategies of exclusion and assimilation, without a clear vision for the way forward. All of these strategies failed at solving the conflicts between the French and the Catalan fishers, which persisted for more than a century after the first arrival of the Catalans (DA 250E126). The Catalans consistently refused to abide by the social norms of the Prud'homie and posed a credible threat to the local fishers by successfully competing with them.

The Arrival of Italian Fishers
In the late nineteenth and early twentieth centuries, successive waves of Italian migrants arrived in Marseille. These migratory flows were so important that about one-fourth of the population of Marseille was Italian by the eve of the First World War (Temime 1986). As part of these migration flows, Italian fishers settled in the neighborhood of St. Jean, which quickly gained the nickname "Little Naples" (Temime 1986: 557; Sportiello 1981: 77-81). The majority of Italian fishers came from three cities located in the vicinity of Naples: Sperlonga, Procida, and Cetara. Migratory flows are usually "bumpy" (Gans 1992), but the arrival of Italian fishers in Marseille occurred relatively smoothly. When I asked an old fisher, who had migrated from Italy in the early 1950s, whether the arrival of Italian fishers had created any tensions with local fishers, I saw incomprehension etched on his face. I did not push my line of questioning any further at that point, but repeated the same question several months later. The old fisher's wife volunteered an answer: conflict was impossible because all other fishers were Italian. The fisher's wife added: "Why do you think my husband's French is so bad?" (implying that he mostly spoke Italian in Marseille). This statement finds further support in the writings of Sportiello who reported in the early 1980s that "all the fishers of St. Jean that are currently working are of Italian descent, no local fisher is left" (Sportiello 1981: 113). These statements are exaggerated, but they capture the fact that Italian fishers were numerous enough to create their own close-knit group that operated in relative social autonomy from local communities. A memorandum drafted in 1963 by the French Maritime Affairs office provides further support for this observation (DA 2331W291 1963). The memorandum describes how fishers from Sperlonga constituted a tight-knit group referred to in Marseille as the "Sperlonga clan." It is not entirely clear how many fishers came from Sperlonga. My interviewees mentioned a "very high number of fishers" who controlled or worked on "approximately 50 boats." In order to better assess the size of the "Sperlonga clan," I reviewed all of the French naturalization decrees issued between 1925 and 1948. These decrees provide a good proxy for assessing the size of the "Sperlonga clan" because French law required fishing crews to include a minimum number of French nationals (and foreign fishers therefore had an incentive to acquire French citizenship as soon as possible after their arrival). The results of my survey are staggering. I counted no less than fifty-two fishers from Sperlonga who settled in Marseille and acquired French citizenship during this period. Counting their families, this population included 182 individuals, all of whom came from Sperlonga. These fishers usually followed one or several family members who had previously migrated to Marseille. For instance, the Di Lelio family counted no less than six fishers who successively arrived from Sperlonga in the 1920s and 1930s. One of my interviewees came from Sperlonga with his parents after the Second World War to join other family members (i.e., his uncles) who had migrated to Marseille in the 1930s.
These figures are conservative estimates of the population of Italian fishers in Marseille as they do not include all of the fishers who kept their Italian citizenship. We should also add to this number the numerous fishers who came from other parts of Italy. For instance, I counted eleven fishers from Procida who acquired French citizenship between 1924 and 1947. Another related group was composed of fishers who migrated from Italy to Algeria in the late nineteenth century, before moving to Marseille after the independence of Algeria in 1962 (Vermeren 2015). The Prud'homie reported the arrival of 120 trawlers from North Africa in 1962, with 10 settling in Marseille (PA 1962). One of my interviewees described how his family of Neapolitan descent fled Algeria on a fishing boat in 1962, wandering from port to port along the Spanish and French coast before settling in Marseille.
I estimate that, all in all, around 15% of the entire population of fishers originated from Italy and a significant proportion of them originated from Sperlonga, Procida, and Cetara in the mid-twentieth century (see Figure 3). This figure, albeit significant, is below the tipping point of 30% identified by Centola et al. (2018).
The memorandum of 1963 cited above further describes how the "Sperlonga clan" brought its own techniques and practices in contravention with the rules of the Prud'homie. Italian fishers first worked on trawlers before converting en masse to a specific fishing technique called lamparo in the early 1960s. The lamparo technique is based on the use of lights that attract pelagic fish (typically sardines or sea breams), which are then encircled with a large net. A fishing treatise from 1909 gives an account of the ways in which fishers of Italian origin practiced lamparo on the Algerian coasts, by using iron grills on which they fired wood soaked with kerosene. Figure 4 portrays one of these lamparos, as used in Algeria at the beginning of the twentieth century.
The fishers of Marseille had experimented with the lamparo during the Second World War in order to improve war-time food supply, but the Prud'homie insisted-and for a while succeeded-in banning this aggressive form of fishing after 1946. For instance, in 1948, the Prud'homie opposed the reintroduction of lamparo in Marseille because of "the impossibility of controlling it" and the "unfortunate abuses" resulting from it (PA 1948). In short, the Prud'homie blamed the lamparo for destroying the marine wildlife and contravening its social norm of long-term sustainability.
The arrival of Italian fishers undermined the prohibition of lamparo fishing by the Prud'homie. This is no coincidence: recent migrants were less likely to observe the traditions of the Prud'homie and favored short-term gains over the preservation of the fishery. In fact, these fishers largely ignored the Prud'homie's regulations banning lamparo. Because the lamparo fishers were not allowed to appoint members to the Prud'homie, they had no qualms rising up against the Prud'homie in defense of their own interests. Several interviewees told me that these fishers were extremely "hard working" and "tough," but also "unrooted," "individualistic," and "here to make money." The memorandum of 1963 corroborates this testimony: according to this memorandum, members of the "Sperlonga clan" were "greedy," "unable to think about the future," and did not engage in "self-enforcement" like members of the Prud'homie (DA 2331W291 1963). The arrival of fishers from North Africa after 1962 did not affect their practices: a former lamparo fisher told me how these newcomersoften of the same Italian origins (a significant proportion of them originated from Procida before settling in Algeria)-showed their counterparts how to use dynamite (a practice that was prohibited by the Prud'homie) in order to maximize their catch. The new migrants from North Africa showed no more concern for the long-term preservation of the fishery than the other Italian fishers: they had lost everything and one fisher explained to me how his family had to "make up for their losses" after resettling from Algeria in 1962. Source: Garau (1909: 38).

Discussion
The case studies set out above suggest that private orders exhibit weaknesses that make them vulnerable to social changes, when confronted with the arrival of newcomers. These case studies allow us to contrast two situations: (1) when entry barriers are strong enough to prevent the settlement of outsiders; (2) when these outsiders are able to infiltrate the social networks of the private order.

Outsiders and Trespassers in the Fishery of Marseille
The respective ways in which Catalan and Italian fishers settled in Marseille influenced their interactions with the Prud'homie. While Catalan fishers settled on the fringes of the Prud'homie's networks, Italian fishers were able to gain entry into some of them. A prime example of "outsiders," the Catalan fishers created their own colony at the outskirts of the Prud'homie's neighborhood. They chose to occupy an old site called the Vieilles Infirmeries (literally the "Old Infirmeries") that had been abandoned a few decades earlier. The Vieilles Infirmeries is located on the south side of the port of Marseille, while the neighborhood of St. Jean is located on the north side. This area lies beyond the reach of the Prud'homie, which was unable to mobilize the means of social control that were pervasive in St. Jean. The Catalan fishers were able to store prohibited fishing gear (DA C4027, 1767) and could easily resist any attempts at control exercised by the Prud'homie over their territory (DA 250E41, 1774). From the shores of the Vieilles Infirmeries (which is now called the "Catalans' beach"), Catalan fishers could access the fishery of Marseille, while preserving their own norms of social conduct. The Catalan fishers showed little concern for the long-term exploitation of the fishery. They invested significant resources in increasing their yield in the short term, and some of them regularly returned to Catalonia after each fishing season (NA C4/181 1785). The Catalan fishers are prime examples of "outsiders" who operate in the background of a private order. These "outsiders" compete with the private order's "insiders" without fear of retribution, which cannot be exacted on individuals who live along rather than within the private order.
Italian fishers settled in a different way. Although some of them resided on the fringes of the fishery, most Italian fishers moved to the heart of St. Jean, carving out a true "enclave" within the neighborhood of the Prud'homie. In addition, the Italians also became active figures in the local Catholic community. For instance, the "Sperlonga clan" replicated the religious celebration of their hometown in Marseille. The "St. Leon," named after Sperlonga's patron saint, was so well attended that it soon surpassed the "St. Pierre," the traditional celebration of the local fishers in Marseille (Sportiello 1981). Fishers from Procida also brought to Marseille their own festivities in honor of their saint patron (St. Michel), which were celebrated at the end of September (Sportiello 1981).
How did Italian fishers manage to move into networks in which the Prud'homie was embedded? Italian fishers migrated in small clusters, progressively creating "chains of related individuals or households" (Mormino 1982: 402) that effectively grounded their community within these networks. In addition, several intermediaries facilitated the integration of the Italian fishers in Prud'homie's networks. For instance, Dominique Guaraccino, a fisher born in Procida in 1862, migrated to Marseille in the late nineteenth century. His father Joseph (also a fisher) was the organizer of the celebration of St. Michel in Procida, and Dominique decided to perpetuate his father's legacy and to continue the celebration of St. Michel in Marseille (which his own son Raphaël continued to organize until his death in 1980). Dominique Guarracino not only became an active member of the religious community, he even managed to be elected to the Prud'homie in the 1920s and was awarded the Légion d'Honneur, the highest French civil and military decoration, in 1926 (not a small feat for a first-generation immigrant). Guarracino certainly managed to leverage his involvement in the religious circles of the Prud'homie to gain acceptance, and even recognition, from his French peers. Another example of an intermediary is Henri Tasso, a French politician who was born in Marseille from Italian parents. Tasso was elected member of the lower house of Parliament , Mayor of Marseille (1935-39), and Undersecretary for the Merchant Navy in the French government from 1936 until 1937. Throughout his political career, Tasso strongly supported the Italian fishers who voted for him on a massive scale (Sportiello 1981: 125), and was also a fervent advocate of the Prud'homie. During the celebration of St. Leon (the patron saint of Sperlonga), the procession stops in front of a statue of Tasso and the band plays the French national anthem-a strong symbol of the complex allegiances that still exist in this community. Another notorious intermediary was Paul Ciaramaglia, a fisher born in Marseille whose extended family originated in the Sperlonga region and Procida. Ciaramaglia was the main spokesman for the lamparo fishers from the 1960s until the 1990s. Several interviewees spoke to me of Ciaramaglia's formidable political acumen and his ability to use all negotiating tools-horse-trading, gamesmanship, and even threats-in order to prevail on behalf of his constituency. His hybrid identity-both French and Italian-made him an ideal middleman in the debates opposing both communities. Despite the strong resistance of the Prud'homie to lamparo, the political acumen of individuals like Ciaramaglia led to the acceptance and diffusion of lamparo in the fishery of Marseille. For instance, Ciaramaglia managed to negotiate a political compromise with the Prud'homie in 1962, leading to the acceptance of lamparo in Marseille (DA 2331W291). Figure 5 is a map portraying the agreement reached between the Italian fishers, represented by Ciaramaglia, and the Prud'homie in 1962. According to this agreement, fishers were allowed to practice lamparo more than one mile out from the eastern coasts of Marseille and more than two miles out from the western coasts. This compromise therefore allowed the lamparo fishers to practice their technique in the fishery of Marseille (which extended three miles from the coast), while limiting their fishing grounds (a limitation that was consistent with the Prud'homie's social norm of the long-term preservation of the fishery). Even though the Prud'homie authorized the lamparo fishers to operate on its fishing territory, it did not allow these fishers to elect or be elected as members of the Prud'homie. In addition, it soon became clear that lamparo fishers were not abiding by the terms of this agreement, which they used to gain a foothold within the fishery of Marseille. The Prud'homie actively complained, but nothing could be done about these breaches: social practices had changed and there was no turning back.

The Complex Interplay of Norms and Practices
The first case study captures a situation in which outsiders (i.e., Catalan fishers) settle at the fringes of the Prud'homie's private order. In this situation, entry barriers are sufficiently strong to keep outsiders at bay. This case study seems in line with the assumption of robustness posited by the theory of private governance. The mode of interaction between the Prud'homie and these outsiders was confrontational and aggressive. The private order violently rejected the outsiders, who openly competed with its members, particularly since they could mobilize techniques and production processes that were prohibited within the private order. Both features-the possibility for outsiders to compete with insiders and the lack of a credible social threat against them-can affect the community of insiders and its practices. For instance, Catalan fishers were able to provide more fish of higher quality than local fishers because of their superior, more aggressive, fishing techniques. Although entry barriers remained sufficiently high to preserve social norms, the nearby presence of and competition from outsiders sufficed to polarize the local community of fishers and to affect their practices, creating a rift between those who were committed to competing with the outsiders (even at the expense of the long-term preservation of the fishery) and those who refused to do so. The social norms established and backed by the Prud'homie remained strong but could not mend the increasing rift between "traditionalists" (who were attached to the norms of the Prud'homie) and "modernists" (who embraced the need for practices to evolve). The Prud'homie naturally stood with the former, but soon developed its own aggressive techniques (namely trawling) in order to keep up with the Catalan fishers. For instance, the fishers of Marseille engaged on a massive scale in a new practice involving two sailboats pulling a single net, thus considerably expanding their dragging power in the first half of the nineteenth century. Ill at ease with this clear violation of its mandate, the Prud'homie entered into a period of institutional schizophrenia in the first half of the nineteenth century, with its members making use of aggressive fishing techniques on a massive scale (in order to keep up with the outsiders), while remaining committed to the social norms of the Prud'homie.
This institutional schizophrenia emerges even more clearly when outsiders gain access to some of the networks in which the private order is embedded, as was the case with the Italian fishers in the twentieth century. Italian fishers managed to settle in the heart of St. Jean, the district occupied by the Prud'homie itself, and were active members of the Catholic community. Because of these connections, Italian fishers were able to introduce their own fishing technique (the lamparo) into the fishery of Marseille. In a repetition of what had already happened with the Catalan fishers, the Prud'homie fiercely resisted the Italian fishers, but without much success. Even some figures from the Prud'homie had difficulties resisting the prospect of greater financial rewards when Italian fishers introduced the lamparo into the fishery of Marseille. The schizophrenic attitude of the Prud'homie concerning the need to keep up with outsiders, while remaining faithful to its social norms, reached new heights. At a meeting held in 1945, the most senior member of the Prud'homie (also called the "first Prud'homme"), Scotto, tendered his resignation from the Prud'homie complaining that the state no longer supported the Prud'homie's regulatory powers, specifically concerning the prohibition of the lamparo (PA 1945). Other attendees promptly called the first Prud'homme's bluff, pointing out that Scotto was in the process of equipping his own boat for lamparo fishing, while denouncing its disastrous effects on the fishery. But the response of the first Prud'homme is even more telling: before leaving the meeting room and tendering his resignation, the first Prud'homme candidly recognized that he "was still an opponent of lamparo but subsequent to the exhaustion of fish due to the use of this technique, material circumstances had forced him to practice this type of fishing" (PA 1945).

The Two Faces of Social Norms: Stability and Social Inertia
This paper has explored one of the key explanations for the evolution of social groups, which focuses on the relative proportion of newcomers compared to incumbents. The case studies explored in this paper contribute to refining this explanation by suggesting that social changes might occur even when the number of newcomers falls below the threshold identified in the literature, as was the case with the Catalan and Italian fishers. Both case studies also suggest that the ways in which outsiders interact with insiders matter as much, and perhaps even more, than the proportion of individuals involved in the interaction. In the first case study, the interaction between incumbents and challengers was aggressive and polarizing, and the challengers were shut out from the social circles of the incumbents. In the second case study, these interactions were tense, but more cooperative, insofar as the challengers managed to enter some of the social networks of the Prud'homie (notably its spatial and religious networks) and establish social platforms with incumbents. These observations enable us to revisit the existing scholarship in light of the nature of the interactions between incumbents and challengers. In Table 1 below, I have coded for the mode of interaction between incumbents and challengers (collaborative or adversarial) and the outcome of their interactions for each of the cases reported in the literature. A collaborative interaction is one where actors build a political coalition and combine resources within a field or across fields (Fligstein and McAdam 2012: 15;Grisel 2017). Competition unfolds when social actors vie for power and legitimacy in a winlose perspective (Fligstein and McAdam 2012: 14;Grisel 2017). The various outcomes reported in the literature can be categorized as follows: assimilation, collaboration, visibility, polarization, and rejection. Each of these five categories can be placed on a continuum from successful to failed integration. A "successful" outcome is one where incumbents and challengers are less prone to conflict and establish interactive platforms. Table 1 suggests that outcomes tend to be more successful when challengers find collaborative platforms with incumbents.
The evidence presented in this paper, together with the review of the existing scholarship, therefore suggests that cooperation, rather than numerical submergence, is key to a successful interaction between challengers and incumbents. These observations are also consistent with indications that social norms might grow weaker when a community creates positive interactions with another community (Portes andSensenbrenner 1993: 1336). The focus should therefore shift from the size of the outsider community to the type of interaction (collaborative or adversarial) between insiders and outsiders. These dynamics give rise to my first proposition: the intensity of social change is proportional to the level of cooperation between challengers and incumbents, irrespective of the relative number of individuals involved. However, this observation requires some further qualifications and nuance. As seen above, the evolution of social practices does not systematically lead to the evolution of social norms. In the cases presented above, outsiders were able to force the social practices in the private order to evolve, but were not necessarily able to change its social norms. The dichotomy that I observed between the discourse of the incumbents (who remained strongly attached to their social norms) and the reality of their practices (which evolved towards the use of aggressive fishing techniques) illustrates this difficulty. In the first case study, the social norms of the community remained relatively stable, although the competition from the Catalans affected social practices. In the second case study, the social norms also remained stable, but the growing gap between the evolving fishing practices (individualistic and short-term oriented) and the social norms of the Prud'homie (communitarian and long-term oriented) led to the alienation of the Prud'homie. The Prud'homie still exists, its members are elected every three years, and its formidable powers remain untouched. But a word keeps coming back in the mouth of my informants: according to them, the Prud'homie is "nothing." I pondered for a long time the meaning of this "nothingness." My informants did not mean to disparage the Prud'homie. Most of them are proud of the Prud'homie and of its heritage. What they mean is that the new practices in the fishery of Marseille have hollowed out the social role of the Prud'homie and that its norms have become an empty shell of long-term cooperation in an ocean of selfishness and competition.
The case study of the Prud'homie therefore suggests that the imaginative force and capacity for change displayed by private orders are impeded by their own social norms. The argument that private governance based on social norms is well equipped to resist social changes therefore seems incomplete. Social norms evolve, but they evolve more slowly and less intensively than social practices. The gap between the speed and intensity of changes in social practices, on the one hand, and the speed and intensity of changes in social norms, on the other, is key to analyzing the evolution of private orders. When this gap is too broad, social norms become a mantra to which social actors pay lip service, while failing to comply with them. These observations can be summarized in a second proposition: the greater the gap between social norms and practices, the less effective a system of private governance will be.
In conclusion, this article presents a case study drawn from a medieval private order that has operated without interruption for about 600 years in the fishery of Marseille (South of France). This case study allows us to track the evolution of a private order over a long period of time and to gather data concerning periods of social changes provoked by migratory flows. These data are used to test the assumption, which underpins most of the literature on private ordering, that private orders are resistant to these types of social changes because of their strong social norms. The evidence presented in this paper suggests that private orders are susceptible to collapse not despite, but because of the strong social norms that support their activities. The robustness of social norms is key to the efficiency of private orders, but also a major obstacle when they are called to accommodate new practices. More generally, this contribution aims to complement the optimistic analyses of private governance that have attracted considerable attention in the past few decades and open avenues for future research.

Florian Grisel is Associate Professor of Socio-Legal Studies at Oxford
University. His research focuses on the emergence and evolution of judicial governance. This article is part of a book project on the limits of private governance to be published by Hart Publishing.