The social organization of difference

ABSTRACT Rapidly diversifying societies, rising inequalities and the increasing significance of social differences are concurrent processes calling for a reexamination and reworking of certain conceptual and theoretical tools within the social sciences. Here, bringing together a range of theories and findings from various disciplines, a conceptual model is offered to facilitate analyses of such intertwined social processes. The model highlights mutually conditioning relationships between the fundamental conceptual domains of: social structures (here described as configurations), social categories (or representations) and social interactions (or encounters). The connections between these domains produce and reproduce, differentially in distinct times, scales and contexts, what can be called “the social organization of difference”.

breadth of literature concerning pertinent approaches, concepts and findings. The literature in question concerns three fundamental fields of social scientific theory or abstraction. These are: group categorizations, social interactions, and social stratification. The combined, mutually conditioning dynamics of these three abstract domains produce what I call "the social organization of difference". Greater attention to the three-way working of these, I argue, will lead to better understanding of how social changes related to difference take place and generate various outcomes.
Prior to introducing a model of the social organization of difference, I must highlight the ideas of certain prominent scholars who have made substantial contributions toward these three fields and shaped my understanding of the inherent relationships between them. There are of course many, but in particular are: Fredrik Barth (1969Barth ( , 1972, Charles Tilly (1998Tilly ( , 2001Tilly ( , 2005, Douglas Massey (2007;Massey and Brodman 2014), Michèle Lamont (2014;Lamont and Molnár 2002;Lamont, Beljean, and Clair 2014), Andreas Wimmer (2008 and Rogers Brubaker (2004, 2016b. Some of the often overlapping lessons that I take from their works include these: . There are endemic, mutual connectionsactually, processes of co-productionbetween the domains of human meanings, social interaction and social structure; . Social interactions, themselves structured by status and material inequalities, continuously shape social categories; shifting category meanings, in turn, can influence the nature and modes of social interaction and affect various aspects of social structure; . These mutually conditioning, cross-domain connections are situated both within and across micro-, mesoand macro-levels; . Meanings or categories are not just within individual minds, but are public; although public meanings are unevenly distributed in a population, they nevertheless both shape and are shaped by social interactions, status positions and hierarchies; . Social actions can be predicated on categories in ways that are conscious and deliberate, non-conscious and automatic, or in somewhat in-between way through sets of dispositions; . There are both general and discrete processes pertaining to different kinds of difference.
These lessons form the backdrop or starting point for considering the model outlined in this article. Specific insights from these scholars are cited throughout the piece.
The social organization of difference: a heuristic model Especially in our current period characterized by heightened dynamics of diversification, a core task of social science should be to join disciplinary forces in order to arrive at better understandings of the ways that multiple aspects of difference relate to social relations and processes of social and economic stratification. "Diversity" is not a very satisfactory concept for framing this task within social science. It is a largely unwieldy concept for research and theory due to its ambiguity, normativity and polysemy (Vertovec 2012(Vertovec , 2015. Instead we should try to comprehend "the social organization of difference" (Vertovec 2009(Vertovec , 2015(Vertovec , 2019. Intentionally echoing Barth's (1969) concern with the "social organization of culture difference", the notion at once suggests a system. Moreover, it represents a tripartite scheme capturing the relationship between the three conceptual domains in question. That is, we have "social" (concerning interpersonal interactions and behaviours), "organization" (relating to patterns, forms, institutions and structures of society) and "difference" (referring to socially constructed categories).
The model offered here is not specific to race and ethnicity, but is applicable to numerous modes of difference: race, ethnicity, culture, gender, sexuality, age, disability, legal status and citizenship (see Vertovec 2015). Within the model, each category of difference can be examined in terms of its own meanings, constraints and processes, thereby avoiding a tendency to treat all differences as the same in effect. This is in keeping with the fact that every criterion or marker of difference (and their intersections) has a unique history of discrimination, with differential self-and other-ascribed meanings and discrete social, economic and political outcomes (cf. Delgado and Stefancic 2017).
The "social organization of difference" model draws directly on J. Clyde Mitchell's analytical strategy for situational analysis (see Mitchell 1987;Rogers and Vertovec 1995;Vertovec 2019). This entails a similar tripartite model of different domains or levels of abstraction of structural, categorical and interpersonal actions in reflexive relation to one another. The method sees an analytical juxtaposition of (1) a situation or set of social actions, (2) the structural setting in which the situation is constituted, and (3) the meanings that the social actors themselves attribute to their actions. "The general perspective," wrote Mitchell (1987:, 9) is that "the behaviour of social actors may be interpreted as the resultant of the actors' shared understandings of the situation in which they find themselves and the constraints imposed upon these actors by the wider social order in which they are enmeshed." The following model and its constituent domains configurationsrepresentationsencounters is not intended for explicating mechanisms or causal explanations, but rather for interpreting how social structures, meanings and social actions work generatively to shape each other. The approach perhaps represents a form of what Tilly called "systematic constructivism" (2008,196), addressing how social constructionof categories, social relations, status systemsactually works by examining the dynamics of mutually conditioning systems. In using the model, broadly following Mitchell's approach, the task for the social researcher is firstly to isolate a phenomenon or case to be described within one of the three abstract domains, and secondly to account for the conditioning influences of the other two domains upon it. The key feature, and indeed point, of the model is a kind of thought exercise. The model highlights the approach that from whatever "entry point" (by way of a phenomenon or case to be interpreted in any domainsuch as an image in the domain of "representations", an organizational arrangement as a "configuration", or a specific interaction as "encounter"), a full analysis of the phenomenon or caseits content, development and implicationsshould entail a look at how it has been shaped by, and itself impacts, the two other domains.
Before engaging with the question how phenomena across these domains are linked, let us briefly look at what can be said to comprise each of the three domains themselves.

Configurations
By configurations, I refer to a variety of phenomena embodying stratified social structures (cf. Grusky 2014). The configurations concept is centered on the idea of social structure as an array of social positions (cf. Porpora 1989). It is based on the understanding that stratification has multiple facets, comprising an interplay of economic, cultural, social, power-based, honorific, civil and physical phenomena (Grusky and Weisshaar 2014). Class has conventionally been a significant notion for describing stratified social positions that includes many of these elements, especially economic ones. I suggest "configurations" rather than simply class-centred hierarchies or stratified social structures in order to convey a patterned assemblage or formation of social positions of both vertical and horizontal inequalities (cf. Stewart 2008). Configurations entail historically produced arrangements of social hierarchy, differential power, cultural distinction, economic wealth, poverty and other material outcomes.
I consider the rubric of configuration as somewhat analogous to that of "matrix of domination," put forward by Patricia Collins (2000).
A matrix of domination sees social structure as having multiple, interlocking levels of domination that stem from the societal configuration of race, class, and gender relations. This structural pattern affects individual consciousness, group interaction, and group access to institutional power and privileges. (Andersen and Collins 2018, 400). These latter outcomes described by Andersen and Hill Collins are facets of social position. A key contribution of this "matrix" concept is that it points to the ways that stratification works through the combined effects of structural influences. No one factor is in play. This approach is also akin to the social ecology perspective of Massey and Brodman (2014), which describes stratification and differential social structural outcomes as being conditioned by the interaction of multiple spheres.
Configurations are comprised of social positions, yet they are both expressed in and shaped by formations such as political structures and institutions, laws and criminal justice systems, legal statuses and legal frameworks, corporations and businesses, public organizational structures including health systems, schools and civil institutions. Configurations are most manifest in indicators such as wealth and income distribution, the allocation of jobs, health profiles and life expectancy, incarceration rates and various sites and modes of power relations. Despite undeniable progress, there is considerable evidence for the persistence if not exacerbation of difference-based disparities in these kinds of indicators (Manduca 2018). Configurations, as systems or structures of stratified social positions, also fundamentally manifest as space, particularly through systems of segregation (see, among others, Massey, Rothwell, and Domina 2009). This relates to what Lamont, Beljean, and Clair (2014) consider to be "place-based inequality", and what Brubaker (2015:, 34) describes as "social separation," that is "concentration in residential, occupational, institutional, social-relational, marital, consumption, media, and recreational space."

Representations
Categorization and classification are of particular interest in Social Psychology. In this discipline, categorization is regarded as an intrinsic ability of the human mind that allows it to simplify a complex world or render it intelligible (Augoustinos 2001). Some key questions that have shaped research into social categorization include: what is a category's nature and attributes, how is it bounded, how is it related to other categories or how is it ranked, and importantly, how are social categories produced, reproduced and changed?
One prominent set of responses to these questions is to be found in social representations theory (albeit a fragmented and diffuse body of literature; cf. Howarth 2006). Initiated by Serge Moscovici (1984), this body of theory was inspired by Durkheim's notion of collective representations. Unlike the Durkeimian idea which posits a broad and consistent set of meanings maintained over time across a population, Moscovici emphasized ongoing processes of dynamic and interactive co-construction of representations (hence the "social" qualifier). Social representations can be seen as systems of everyday meaning that allow people to know and interpret their social world, from situation to situation. Social representations research and theory focuses on "meaning-making processes" among and between people (Sammut et al. 2015, 5-6). Through such micro-level processes, social structures are created and reproduced.
Social categories or representations tend to be "groupist" in nature (Brubaker 2004). This means that people often tend to regard identities as based on groups that: have clear borders, are homogeneous in values and practices, act as one, and pit themselves against other, similarly conceived groups. Nevertheless, many representations of groups are not evenly distributed or shared, but evolve and circulate in specific milieus, and they are often contingent, temporary or fluctuating. In this way, it is crucial to recognize the plurality and positionality of representations in the public sphere. This realization also significantly informs theoretical work not just on how representations are dispersed, but also on how social representations change, how and to whom they are communicated, objectified, diffused, projected through images and propositions, made intelligible vis-à-vis other existing concepts and classifications, legitimized and reified.
Representations are often intersectional. Caroline Howarth (2002) provides an apt illustration through an analysis of pervasive representations of race, place, and gender as well as their social effects in Brixton, London. These kinds of representations are also invested with moral meanings as well. Examples here are widespread notions of "soccer moms" (conveying notions associated with white, middle class, virtuous women) and "welfare queens" (conveying notions associated with black, poor, deceitful women) (Winter 2008, 160).
The notion of representations here also underscores the role of depictions frames, schemas, images and discourses about categories, particularly in the public space and everyday life. Representations are the stuff not only of social interactions, but of public communication. In this wayagain, going beyond some of the work to date on social categorization, the notion of representations takes into strong consideration the place of mass media and their effects.
The news media, for one, "play an influential role in shaping what and how people think about an issue" (Haynes, Merolla, and Karthik Ramakrishnan 2016, 19, emphasis in original). This is particularly evident in the increasing amount of research on the media framing of immigrants and immigration (such as Helbling 2014, Thorbjørnsrud 2015, Caviedes 2015. "From the perspective of scholars of migrants and minorities," Erik Bleich, Irene Bloomraad and Els de Graauw (2015: 862) suggest, "the relevance of framing and representation is clear: it is vital to understand how different groups are portrayed and the extent to which media representations affect public opinion, political mobilization and policy outcomes." It is extremely common practice, across mass media and within political rhetoric, to represent immigrants as alien threats to the nation (Vertovec 2011). Migrant representations are regularly framed by notions of invasion, criminality, or economic threat (Bleich, Bloemraad, and Graauw 2015) or of system abuse through frames such as "anchor babies" and "chain migration" (Alamillo, Haynes, and Madrid 2019).
Increasingly we are witnessing how selective representations, via framing and strong imagery, are keys to understanding the power of social media, too. This is especially interesting to consider givenand because ofthe brevity of format in many applications. In social media, categorical framings are especially brusque. It is also hugely important to study due to the sheer scale of the phenomenon. For example, no less than 68% of American adults use Facebook, the great majority of them daily (Pew Research Center 2018). "When people use Facebook to see exactly what they want to see," Cass Sunstein (2017, 2) says, "their understanding of the world will be greatly affected" (Ibid.: 2). Of course this includesperhaps foremostrepresentations of social groups.
The dynamics and effects of public representations of difference are not monolithic, but vary across social groups. Moreover, mass media or social media audiences are "not just passive vessels ready to be filled with biased frames and subframes" (Ortega and Feagin 2017, 20). People are often quite aware of competing frames in the public sphere, knowing which politicians, movements or parts of the political spectrum are associated with which frames, thereby being able to follow or resist them (Haynes, Merolla, and Karthik Ramakrishnan 2016, 177). Indeed, as observed by Brubaker (2004:, 68-9), A common thread in studies of everyday classification is the recognition that ordinary actors usually have considerable room for maneuver in the ways in which they use even highly institutionalized and powerfully sanctioned categories. They are often able to deploy such categories strategically, bending them to their own purposes; or they may adhere nominally to official classificatory schemes while infusing official categories with alternative, unofficial meanings.
This point was evident in Baumann's (1996) Contesting Cultures, a detailed ethnography of Southall, west London, demonstrating how locals are quite able to distinguish, interweave and strategically use "dominant" (official) and "demotic" (everyday) representations of cultures and ethnicities (cf. Brubaker 2015).
Further, public representations of difference are not fixed. The case of Muslims in the United Kingdom show this. Prior to Rushdie Affair at the end of the 1980s, British Muslims were often represented in the public sphere as a rather uncontentious sub-population, just one of several in multicultural Britain. Since the Rushdie Affair and alongside the rise of international Islamicist terrorism with which British Muslims have been increasingly impugned, representations of "Muslim" in the UK have been transformed into an ever-more stigmatized category, with direct effects on interactions and status (Poole 2002). Representations of Asians in America have travelled in the other direction, as it were, from "yellow peril" to "model minority" in the space of about a hundred years (Hsu 2015). Following the Covid19 pandemic, Asians in America and many other countries have been, once again, portrayed by many as a threatthis time, to public health.
Categorization is natural, social psychologists tell us, but group categories and their contents are social. Received and co-constructed representations of categories filter the social world and our activities in it. And they are directly related to processes surrounding inequality. As Tilly (2005:, 111) put it, "Categories matters. To the extent that routine social life endows them with readily available names, markers, intergroup practices, and internal connections, categories facilitate unequal treatment by both members and outsiders." We are surrounded by and immersed by categories and their representations through public communication. They are a large part of our daily information processing and are fundamental to the social organization of difference.
These socially, culturally, and historically constituted ideas and beliefs, or cultural models, get inscribed in institutions and practices (e.g. language, law, organizational policies), and daily experiences (e.g. reading the newspaper, watching television, taking a test) such that they organize and coordinate individual understanding and psychological processes (e.g. categorization, attitudes, anxiety, motivation) and behavior (e.g. voting, interpersonal discrimination, disengagement) … (Plaut 2010, 82) Our actions and interactions are conditioned, channeled, inhibited and interpreted by representations. Conversely, representations are shaped, reified and legitimized in our social actions.

Encounters
Representations and unequal social structures are formed, manifested and remade through interactions across social categories and boundaries of difference. Such interactions are increasingly the subject of a cross-disciplinary literature on "encounters".
In a review of recent geographical works, Helen Wilson (2017:, 451) contends that ""encounter" is a conceptually charged construct that is worthy of sustained and critical attention." Rather than a mere synonym for meeting, she describes encounters as a specific "genre of contact", often considered to describe interactions that are by chance, casual or fleeting. Such brief interactions are considered to be emblematic of public spaces such as markets, transport links, shops and cafés. In any context, "encounters are fundamentally about difference and are thus central to understanding the embodied nature of social distinctions and the contingency of identity and belonging" (Ibid.: 452). Much work around encounters, Wilson notes, assumes a lack of commonality, an "us versus them". Most often this is about inter-ethnic encounters, but also sometimes about crossing class, legal status, religion and sexuality. Encounter is a notion most employed in instances in which there are "clear distinctions of social identity and categorization, with an attention to how difference is negotiated, constructed and legitimated within contingent moments of encounter" (Ibid.: 454). Beyond these common uses of the notion, Wilson stresses that "Encounters make difference" (Ibid.: 455, emphasis in original). Interactionsas encounters across differencemay constitute, construct and reproduce representations of difference.
Wilson's overview is paralleled by a review of anthropological ethnographies of encounter by Faier and Rofel (2014). They, too, distill a general definition of encounters as engagements across difference. In this literature, there is often concern about "how culture making occurs through everyday encounters among members of two or more groups with different cultural backgrounds and unequally positioned stakes in their relationships" (Ibid.: 364). What is shown to arise from encounters across difference are "new cultural meanings and worlds" (Ibid.: 365). This is not always the result, however: among significantly differently situated groups, encounters may also reproduce or reify boundaries and identities (cf. Baumann 1996; Lamont and Molnár 2002;Wimmer 2013).
Here we can immediately see the relevance of what I am calling representations, or how the actors within an encounter categorize and narrate each other (especially by way of publically communicated categories involving boundaries, attributes and statuses). Tilly, too, observed this process and stated that, "When previously unconnected clusters of persons encounter each other, members of each cluster react to the encounter by creating names, practices, and understandings that mark the points of contact between them" (2005, 112, emphasis in original). Furthermore, encounters do not take place in a vacuum, but within historically produced configurations of power and status. Hence, Matejskova and Leitner (2011, 721) state that "Real-life contact between members of different social groups is always structurally mediated and embedded in particular historical and geographical contexts of power relations between and within social groups." Just as Wilson (2017:, 455) puts forward that encounters "make difference", she also asserts that encounters should be seen as "meetings that also make (a) difference" (Ibid.). That is, encounters can change representations or even status and structural positions. "[W]hilst fleeting encounters have been dismissed as having little meaning or little ability to transform values and belief," Wilson (Ibid.: 463) writes, "it is possible that encounters accumulate, to gradually shift relations and behaviour over timeto both positive and negative effect." One of the mechanisms by which this is accomplished is "to produce moments of cultural destabilization that allow participants to establish new intercultural understandings" Sadgrove 2014, 1980). This approach to encounters across differenceas leading to shifts in perception, representation and attitudesrelates directly to the long-established and well-researched area of contact theory. Stemming from the work of Gordon Allport (1954), contact theory posits that personal contact with out-group members can reduce prejudice. It is important to note that the theory posits certain conditions, including equivalent status of groups within the situation, common goals, cooperation/lack of competition and support of wider authorities. Allport also acknowledged the continued role of inequality in preventing positive outcomes of contact. Miles Hewstone (1996) adds that for contact to work positively, group affiliation must be clear and salient. Persons with whom contact takes place must also be considered as representative of an out-group. Considerable evidence within social psychology clearly demonstrates that, given such suitable conditions, contact indeed "works" (see especially Pettigrew and Tropp's 2006 metaanalysis of 515 contact studies).
Gill Valentine (2008) suggests that perhaps many positive encounters may be merely enactments of codes of everyday urban etiquette or civility. This "ephemeral civility of the minor Goffmanian interaction rituals, casual conversations, shared greetings, little jokes, bits of gossip, small talk about the weather or how long a wait it is for a bus" (Collins 2000, 250) may mean very little in terms of changing attitudes, behaviours, everyday representations and micro-status positions. Or, as Wilson suggests above, they might actually accumulate to influence or change modes of interactions and the representations that accompany them.
The answer as to whether or not encounters produce meaningful or longterm change in categorization is not straight forward. For a start, several studies within the geography of encounter literature place emphasis on the role of specific contact spaces themselves (e.g. Mayblin, Valentine, and Andersson 2016). The approach is grounded in Ash Amin's (2002:, 959) noted call for observing "micropublics of everyday social contact and encounter" such as music clubs, theatre groups, and communal gardens. In this way, Piekut and Valentine (2017, 176) set out to measure and understand the effects of different kinds of encounter spaces, reasoning that, "because the nature of encounter is socially produced differently in different types of space, depending on whether the encounter setting is more public or private inter-ethnic contact in different spaces will have a different effect on attitudes towards minorities." While Wilson suggests that fleeting encounters across difference might accumulate to change attitudes and representations, Matejskova and Leitner (2011) note how such attitudes toward individual immigrants often do not "scale-up" to transform group stereotypes. This resonates with Valentine's (2013:, 6) view that, In the context of both personal and community insecurity it is possible to see why some people find it hard to have mutual regard for groups they perceive as an economic or cultural threat. … This means that prejudiced individuals can have a vested interest in remaining intolerant despite positive individual social encounters with communities/individuals different from themselves.
Beyond interactions that serve to maintain an adverse status quo, Amin (2013:, 5) calls our attention to a "darker aspect of everyday encounters of difference". This is when encounters are marked by not just unpleasantness, but sheer bigotry, hostility and conflict. These might occur alongside otherwise positive experiences. For instance, even within an otherwise convivial "cosmopolitan canopy," Elijah Anderson (2011:, 291n.) witnesses the "the nigger moment," "a situation in which a black person is made to feel his provisional status most acutely." Further, we must recognize the possibility that there might occur neither clearly positive nor negative outcomes of encounters. Through a study of Polish migrants in Britain, Anna Gawlewicz (2015:, 268) understands the ""meaningful-ness" of encounters more broadly, as a capacity to form, alter or complicate people's feelings about difference." By way of encounters and their effects, her informants reported neither positive nor negative attitudes, but "in-between" or "complicated" feelings. Also within this field, new work attempts to assess intentional or engineered encounters (meetings, clubs, festivals and such) meant to break down prejudices and increase respect for others (e.g. Paulsen Galal and Hvenegård-Lassen 2020), or to observe non-conscious and habitual encounters (e.g. Wilson 2011). With a wide variety of interesting sites, methods and outcomes, everyday encounters and their influence on representations of difference and systems of social stratification currently presents a rich, indeed boom field of research and theory.
How or under what conditions does mutual conditioning between the three domains take place? Below, I take a closer look at what key scholars have said about each of the axes of mutual conditioning depicted in Figure  1. The available literature suggests no specific set of conditions or singular process of mutual conditioning, but rather a variety of processes, conditions and mechanism potentially at play.

Representations -configurations
Across various disciplines, there are a number of studies that I would place on the axis referring to the mutual influence of representations and configurations. Many years ago, Omi and Winant's (1986) racial formation theory advocated a kind of mutual influence of these domains, as they put forward the premise that race must be seen through the linkage of social structure and cultural representation. For them, the category of race itself is framed in terms of social structure. More recently, on broader terms, James Jones and his co-authors (Jones, Dovidio, and Vietze 2014, 12) describe how "Societies construct significance for any concept or thing by imbuing it with beliefs and assumptions and by applying actions and organizing structures to it." Categories such as Black, White, Asian, they (Ibid.) suggest, "are socially meaningful when they result in differences in treatment and different social outcomes within a diverse society." Plaut (2010:, 84) similarly describes societal processes "perpetuating representations of and behaviour toward social groups and ultimately reproducing cultural and structural realities." As evident in the earlier-referred to works of, for instance, Tilly and Wimmer (and to a certain degree, Massey), sometimes structuring by way of representations occurs through conscious acts of discrimination, exclusion or various acts of "boundary work". Yet we must also recognize, following Brubaker and Lamont (and again Massey), that taken-for-granted or non-conscious processes utilizing representations and effectively reproducing stratifications are also part of everyday social dynamics. It is often hard to determine the degree of conscious/deliberate and non-conscious/inadvertent structuring-by-representation. Lamont, Beljean, and Clair (2014) describe what they call "cultural processes" connecting social structures and individual cognition. These entail mechanisms of identification (including racialization and stigmatization), which seem to be more non-conscious, and mechanisms of rationalization (including standardization and evaluation) which seem to be more conscious. Together, such cultural processes and mechanisms serve to bridge micro and macro levels (Massey 2014). Individuals categorize themselves and others, usually following broad contextual frames, subsequently serving to reconfigure social relations and social structures over time (Tomaskovic-Devey and Avent-Holt 2014; also see below).
There is often a question as to which categories are most salient in particular places or times as well as to how people get arrayedor array themselves into those categories. This is evident in the research of Saperstein, Penner, and Light (2013), who examine the consequences of consciously and unconsciously organizing our workplaces, families, neighborhoods, and governments around such categorical distinctions of difference. Similarly, Domina, Penner, and Penner (2017, 311) describe how schools function as "social sorting machines, creating categories that serve as the foundation of later life inequalities." Elsewhere, Saperstein and Penner (2014) examine the social positions individuals occupy, how they identify themselves with social positions and the implications social positions have on how they are seen by others. Their findings suggest that, some portion of Americans who experience an increase in their social position are "whitened" as a result of this mobility, and similarly, some portion of those who experience a decrease in their social position are "darkened". Perversely, this implies that in the contemporary United States, the more fluid race is at the individual level, the more entrenched racial inequality will be at the societal level, as changes in the classifications and identifications of individuals serve to reinforce the existing racial order. (Ibid.: 678).
Although these findings are contestedfor instance, that they might mainly relate at best to mixed race and Latinx respondents (Alba, Insolera, and Lindeman 2016)they point to the possibilities and dynamics of fluidity of people's self-representations of category and social position. Parallel research and experiments have led to the development of system justification theory, which contends that many peoplesurprisingly including members of disadvantaged minorities as well as members of privileged majoritiesare motivated to maintain and legitimize, usually at an unconscious level, an existing social categorical order or status quo (e.g. Jost, Banaji, and Nosek 2004). Whether ranking is "upwards" or "downwards", this shows a clear relationship between the meanings embedded in certain representations and the social status accorded to them.

Configurations -encounters
The mutual conditioning of social structure and direct social interactions concerns the basic stuff of Sociology, significantly including debates around structure, agency and structuration (notably Giddens 1984). Social positions, status and power condition social interactions, and social interactions produce or reproduce social positions, status and power. With reference to such a dynamic, Randall Collins (2000Collins ( , 2004 underlines to what he describes as situational stratification. In this approach, stratification is something negotiated during actual interactions, which themselves are articulations of power and status. Macro-conditions of power and status impinge upon small everyday acts and interactions. In turn, macro conditions only form through microactions. Correspondingly, Brubaker (2015:, 25-6) suggests that, "It is in and through these everyday encounters that respect, recognition, and status are distributed in iterative and cumulatively consequential ways." Dirksmeier and Helbrecht (2015, 487), too, submit that, "Encounters between strangers become the foundation for social stratification in general and, hence, any encounter is a subliminal negotiation of dominance, of distinction, and an expression of Foucault's microphysics of power." For similar reasons, Tomaskovic-Devey and Avent-Holt (2014) stress that researchers should examine the relatively durable kinds of relationships that are situated in local fields of action or socially organized spaces such as workplaces, schools, and neighborhoods. They see this as a crucial task of "specifying the social context in which the social relations that engender meaning-making develop" (Ibid.: 622).
Such dynamics are perhaps best portrayed ethnographically. For instance, in a study of small town Minnesota, Helga Leitner (2012:, 830) shows how residents "draw on a broad scale and longer term conceptions of nation, community, and place to racialize immigrants of color as out of place and to set conditions of belonging to the local and national community." In this way, macro-structures and their associated positionalities serve to shape encounters and relations with newcomers. However, the encounters do not always reproduce social structures. Rather, Leitner shows how encounters with difference are also able to disrupt preconceived categories, boundaries and structures. She (Ibid.: 833) writes, that "positionalities and encounters are co-constitutive. Encounters frequently reflect and reproduce the positionalities of those involved but also hold open the possibility of positionalities being called into question through the encounter."

Encounters -representations
Again, a fundamental theme across the social sciences concerns the ways that social categories are constructed, formulated, shaped, blurred, sharpened and reproduced through social interactions (especially across modes of difference such as race, ethnicity, gender). Converselyas engaged by key theoretical works on social and symbolic boundariessocial scientists have been equally concerned with the ways that social categories themselves differentially shape, constrain, and enable social interactions. In this way, as Massey (2007:, 242) describes, neither categories nor social relations remain fixed: The definition of categorical boundaries and the content of conceptual categories are not, however, automatic. They are learned through instruction and modified by experience. As social beings, people constantly test, extend, and refine the social schemas they carry in their heads, typically though interactions and discussions with other people.
In terms of studying social categories, such an approach as advocated hereemphasizing the mutually conditioning roles of configurations, representations and encounterscan help social scientists better to develop a kind of conditional constructivism. Like Wimmer (2008Wimmer ( , 2013, I also seek an approach to representations that avoids category essentialization (that deems groups bounded, closed and homogenous) on the one hand, and radical constructivism (that considers groups to be effectively created, racialized or ethnicized almost entirely by institutions, policies, and powerful actors). By acknowledging the co-constitutive play of encounters, we can appreciate how representations are not just consumed, reified or utilized freely and openly by actors, but done so through social practices and interactions within a field of inter-related or conflicting, multi-scalar constraints, structures and opportunities.

Domain lag
Conditions of change in any single domain may, through variable processes, trigger and shape changes in one or both of the other domains. Yet while the three domains of the configurations-representations-encounters model are mutually conditioning, reasons for change are highly contingent and, especially, paces of change are not predictable.
The co-evolution of phenomena within the three domains is not necessarily an even process: changes in one domain might develop long before transformations are felt in either of the other domains. For instance, policy changes regularly take a considerable period to modify social relations, and shifting social relations might not be reflected in public or political discourse for some time. In such cases, we might consider the nature and implications of what can be called "domain lag". (Vertovec 2015, 16) There is often lag, or temporal delay, affecting processes of change between domains, with developments in one often moving faster than in others. This might be due to either exogenous or endogenous shifts.
For example with regard to the domain of configurations, Germany experienced a massive influx of over one million asylum-seekers in 2015-16 (Bock and Macdonald 2019): this exogenous or externally originating process entailed a rapid and largescale shift in social and political structures. New modes of encounter subsequently ensued across German society, from migrants' engagements with state bureaucracies through exchanges between migrants and volunteer groups to everyday interactions involving the recent migrants in towns and urban neighborhoods. New public representations of the newcomers in Germany, however, have taken much longer to emerge. This is because the influx included migrants of many kinds that were not familiar in the German context, such that standard or longstanding categories, frames and schemas were not fit for purpose (cf. Vollmer and Karakayali 2018). As examples in the domain of representations, we have seen endogenous patterns of change surrounding the adoption of new language to talk of disabled persons (Thomas 2015), progress in the ways LGBTQ persons and relationships are represented on television (Albertson 2018), and attempts at fostering new modes of signification through the strategic adoption of nomenclature such as "African American" (Philogène 2001) and "People of Color" (Pérez 2020). It remains to be demonstrated whether such changes of representation have led to changing social encounters or shifting status configurations. And in the domain of encounters, as exemplified in neighborhoods of London, we can observe the emergence of commonplace patterns of everyday interaction across difference remain unremarked upon (as changes to representations) or that do not appear yet to impact patterns of social stratification (as changes to configurations) (Wessendorf 2014).
Developments in any single domain needn't immediately "cause" change in another domain. Cross-domain effects may be incremental over time (as in each example above, in which a substantial change in a domain only gradually, or some time later, shifts phenomena or patterns in another domain). In so many sociological studies and theories addressing links between categorization, social interactions and social structuring, the temporal dimension remains largely under-examined. It is hoped that the social organization of difference model offered here might provide a useful framework for elaborating and elucidating such temporal links.

Conclusion
There exists a considerable body of scholarly work concerning aspects of social stratification, categorization and social interactionor in my terms, configurations, representations and encounters. It is not only the inherent nature of the processes within these topic areas or abstract domains (each a substantial field in its own right) that matters for understanding difference and social structure, but appreciating the linkage of these domains. Such linkage creates, shapes, reproduces or shifts the social organization of differenceincluding its distribution of inequalityin any society, within and across macro-, meso-and micro-scales (cf. Saperstein, Penner, and Light 2013).
Many key authors highlighted in this article have pursued the linkage of these domains in various ways themselves. Massey (2007:, 16) has stated that, No matter what their position in the system, people seek to define for themselves the content and meaning of social categories, embracing some elements ascribed to them by the dominant society and rejecting others, simultaneously accepting and resisting the constraints and opportunities associated with their particular social status. Through daily interactions with individuals and institutions, people construct an understanding of the lines between specific social groups … Lamont and her colleagues (2014, 586) have sought to conceptualize cultural processes as ongoing classifying representations/practices that unfold in the context of structures (organizations, institutions) to produce various types of outcomes. These processes shape everyday interactions and result in an array of consequences that may feed into the distribution of resources and recognition. And Brubaker (2004:, 13) has drawn attention to such linkage across different scales of analysis, suggesting that, From above, we can focus on the ways in which categories are proposed, propagated, embedded in multifarious forms of "governmentality". From below, we can study the "micropolitics" of categories, the ways in which the categorized appropriate, internalize, subvert, evade or transform the categories that are imposed on them … In this article, I have reviewed and grouped a range of literature and approaches by way of a conceptual model. It offers a condensed framework, language and technique for relating the kinds of phenomena and processes described above. The model of configurations-representations-encounters analytically abstracts and distinguishes domains of phenomena incorporating social structure, status and stratification (configurations), categories, images and discourses (representations) and social interactions, from fleeting to sustained (encounters). It can be used to account for the social organization of difference in particular contexts or with an eye to a particular issue. This is evident in Marco Martiniello's (2015) use of the model to examine artistic production among ethnicized minorities, or Jay Marlowe's (2020) analysis of refugee incorporation.
The social organization of difference model can importantly be used comparatively, too. A close examination of specific processes and factors within and mutually influencing the three domains can shed light on specifics in the social organization of difference between countries such as the USA, Brazil and South Africa; within a country between cities like Houston, Los Angeles and Miami; or within a single city, for instance between neighborhoods such as Englewood, Lincoln Park and Little Village in Chicago. Each context or scalenation, city or neighborhoodwill have its own historically produced social organization of difference, its own set of configurations, representations and encounters. In this way, the model can be comparatively utilized brings to elucidate the kinds of conditions that make contexts similar to, or different from, each other.
Further fruitful areas for developing the model might entail the burgeoning field of emotion and affect studies and how it relates to the links between social categorizations, interactions and modes of stratification (see, among others, Bonilla-Silva 2019). Another might be to look at how processes or phenomena of categorical fusionhighlighted by work on themes such as mixed race identities (e.g. Aspinall and Song 2013), multiple categorical distinctions (Tomaskovic-Devey and Avent-Holt 2014) and intersectional complexity (McCall 2005), transnationalism (e.g. Bauböck and Faist 2014) and creolization (e.g. Cohen and Sheringham 2016)serve to disrupt or destabilize the social organization of difference in any particular setting (see Brubaker 2016a). Indeed, as Brubaker points out in his illuminating book Trans, we are in an "age of unsettled identities" in which challenges to established categories have been spectacular, as indicated by the stunningly rapid shift toward social and legal recognition of gay marriage, the mainstreaming of transgender options and identities, and the gathering challenges to the binary regime of sex itself. But racial and ethnic categories have also been profoundly unsettled: by demands for the recognition of multiracial identities, by the increasing fluidity and fragmentation of the ethnoracial landscape, and by the proliferation of crossover forms of racial identification. (Brubaker 2016b, 5) Certainly there is important work to be done on how disruptions to established categories and the emergence of new ones relates to shifting social positions, new inequalities and modes of social interaction, as suggested by the configurations-representations-encounters model. These topics will remain or even intensify in importance within an era characterized by politics of difference and unprecedented processes of diversification. also obliged to students and colleagues for their engagement in seminars on the topic at the Max Planck Institute, Autonomous University Barcelona, University of Tübingen, University of Göttingen, Monash University as well as at Erasmus University Rotterdam, where I conducted much research under the auspices of a Visiting Professorship.

Disclosure statement
No potential conflict of interest was reported by the author(s).