Introduction

In the history of sociology, which spans over a hundred years since the emergence of this field in the nineteenth century, sociologists' conceptualization of space has continued to evolve. The sociological significance of the term "space" has gradually expanded from its initial physical and geographical meanings to the more abstract concept of "social space." This evolution is particularly evident in the works of several French social thinkers. In Bourdieu's "field" (Bourdieu and Wacquant 1992), Latour's "actor-network" (Latour 2005) or Lefebvre's concepts of "spatial practice," "representations of space," and "representational spaces" (Lefebvre 1991,1974), the understanding of space has transcended geographical and physical limitations. Various social meanings, including power, production, construction, and interactions, are incorporated into these concepts. In addition to these influential French theories, another theoretical tradition of social space was established by German sociologist Georg Simmel and the Chicago School of Sociology in America, with a particular focus on "social forms" and "human ecology" (Simmel 19712009; Park and Burgess 1969,1921). However, Simmel and the Chicago School share a common trait in their sociological approach, which puts less emphasis on theoretical systematization. They were less inclined to apply a comprehensive set of theoretical vocabulary to various aspects of social life, as Bourdieu did, for example. Consequently, until the early twenty-first century, within this theoretical tradition spanning from Georg Simmel to Robert E. Park, Erving Goffman, and contemporary sociologists such as Andrew Abbott, their discussions on social space were not comprehensively examined or analyzed. This gap exists in both the Chinese-speaking and English-speaking academic communities.

According to Simmel and the Chicago School, social space is not based on power relations or class struggles, nor does it consider social structures the fundamental elements for understanding space. Instead, it is rooted in the processes of social interaction within specific temporal and spatial contexts. It emphasizes the interdependence of physical space and social space and has two characteristics: endogeneity and temporality. This notably contrasts with the Marxist tradition of class analysis and Bourdieu's field theory. This article does not attempt to comprehensively review the theoretical literature from this sociological tradition spanning over a century. Instead, it focuses on the three most representative authors within this tradition: Simmel, Park, and Goffman. This article examines social spaces in the writings of these three sociologists to uncover the significance of space and time in understanding various concrete and abstract social forms. These three theorists were selected for analysis not only because they represent the same sociological tradition in different periods but also because, although their social theories may seem distinct, they share fundamental ontological and epistemological assumptions about society. This theoretical lineage is particularly evident in their analyses of spaces. The discussion in the following sections primarily employs several classic texts as analytical materials; these include Chapter 9, "Space and the Spatial Ordering of Society," from Simmel's Sociology: Inquiries into the Construction of Social Forms (Simmel 2009), the coauthored books Introduction to the Science of Sociology and The City: Suggestions for Investigation of Human Behavior in the Urban Environment by Park, Burgess, McKenzie, and others (Park and Burgess 1969,1921; Park et al. 1967, 1925), as well as chapters from Goffman's Asylums: Essays on the Social Situation of Mental Patients and Other Inmates and Relations in Public, specifically its Chapter 2, "Territories of the Self" (Goffman 19611971).

Before discussing the spatial theories of these three sociologists separately, the definition of social space needs to be clarified. Although different terms are used in the realm of social theory, such as Bourdieu's "field" or the Chicago School's "ecology," in its most abstract sense, all social spaces consist of three basic elements: (1) actors, (2) spatial locations, and (3) relations between actors and spatial locations (Abbott 2005; Liu and Emirbayer 2016). These three elements illustrate the distinctions between social space and the three commonly used concepts of "social structure," "social system," and "social network." Social structure focuses on the organizational structure of social life, not the autonomy of the actors within the structure. Social system emphasizes the roles and functions of actors within a functional system, not their spatial locations. Social network emphasizes the ties between actors, not the connections between them and spatial locations. In contrast, the key concern of social space is precisely how actors and spatial locations are connected, and these modes of connection and their changes determine the social structure and historical dynamics of the entire space.

Simmel's social forms

According to Simmel, sociology ultimately revolves around the study of social forms. In his works, the term "form" has three distinct yet interconnected meanings: (1) geometric meaning, such as shape, configuration, structure, location, position, and figure; (2) transcendental meaning, such as presupposition, representation, consciousness, and knowability; and (3) vitalistic meaning, such as growth, energy, renewal, vitality, and flow (Silver and Brocic 2019). Among these, the geometric meaning is most evident in Simmel's theory of social space, but the other two meanings have also significantly impacted the sociological imagination and approach of the Chicago School. Simmel believes that sociology should be positioned as the geometry of social science, with its fundamental task being the study of various forms of society founded upon "an abstraction from concrete reality" (Simmel 1950: 11). He suggests that the study of the content carried by these forms should be left to other related disciplines, such as political science and history. The contemporary American sociologist Emily Erikson provides an apt metaphor for this concept: Simmel's sociology studies the shape of a vase, not what liquid is in the vase, and more importantly, "the shape of a vase is not determined by the liquid it contains" (Erikson 2013: 225). The same social form may correspond to different content, "just as the sun, a snow globe, and a basketball are all spherical" (Silver and Brocic 2019: 118), and different social forms can influence the content they carry as well. This unique approach differs significantly from that of other founding figures in sociology, such as Marx, Durkheim, and Weber. It neither emphasizes power struggles nor social cohesion nor prescribes a specific path for the development of society. Instead, this approach focuses on describing the various basic forms of social life and their processes of change.

Although social forms are abstract concepts, they are not static; they emerge and evolve through social interactions. Simmel believes that social interactions among individuals can only occur within space. He argues that "if a number of persons inside certain spatial boundaries live isolated from one another, then each of them simply fills their own immediate space with their substance and their activity, and between this space and the space right next to them is unfilled space; practically stated: nothing." (Simmel 2009: 545). Moreover, space gains sociological significance only through processes of interaction. Kant defines space as "the possibility of being together" (Simmel 2009: 545), which, in itself, is a sociological definition. The various connections and interactions among people make it possible for them to come together. Therefore, social space and social interaction are interdependent and inseparable.

As long as processes of social interaction exist, the issue of boundaries between actors inevitably arises. Simmel believes that "it is not the lands, not the properties, not the city district and the rural district that set one another's boundaries, but the residents or owners performing the reciprocal action." (Simmel 2009: 551) That is, only when social interaction between individuals or groups begins will boundaries in the sociological sense gradually form. Therefore, "the boundary is not a spatial fact with sociological effects, but a sociological reality that is formed spatially" (Simmel 2009: 551). Simmel's sociological interpretation of the concept of "boundary" shares similarities with the late 20th-century concept of "boundary-work" in the sociology of science proposed by Gieryn (1983), as well as Abbott's (1988) notion of "jurisdiction" in the field of sociology of professions. All these perspectives emphasize the spatial meanings of boundaries. They differ from another interpretation in contemporary sociology, which is represented by Lamont (1992, 2000) and Wimmer (2013), who see boundaries as markers of identity differences between races or classes. Abbott's (1995) discussion of "things of boundaries" more explicitly inherits Simmel's sociological perspective on social forms. It posits that boundaries are not just lines separating things; they shape the contents of individuals, groups, organizations, and other social entities. When boundaries are connected into a certain form, social entities emerge.

Simmel provides a vivid metaphor for explaining the significance of boundaries in social space, comparing it to the role of a frame in a work of art. He states that "the frame announces that inside of it there is a world subject to its own norms, a world that is not drawn into the determinants and dynamics of the surrounding world; while it symbolizes the self-sufficiency of the artwork, at the same time by its very nature it highlights the reality and imprint of the surroundings" (Simmel 2009:548). Similarly, boundaries, on the one hand, separate a social space from the surrounding world, ensuring that the social space remains "internally cohesive," and, on the other hand, serve as "the interacting unity, the functional relationship of each element to each acquires its spatial expression in the framing boundary" (Simmel 2009: 549). This metaphor of a picture frame effectively illustrates Simmel's assumption of the "endogeneity" of social space, suggesting that social space, to a significant extent, is autonomous in relation to the surrounding world. Because boundaries exist, although social space may bear the "marks" of the external environment, its social structure is fundamentally determined by interactions among internal actors. The subsequent analysis will demonstrate that this assumption of endogeneity persists in the discussions of social space by Chicago School sociologists such as Park and Goffman.

In addition to boundaries, social distance is another core concept in Simmel's theory of social space. Contemporary Western sociology often understands social distance from the perspective of social network analysis, particularly based on Simmel's discussions of "dyad" and "triad" (Simmel 1950). Some sociologists, such as Black (2000), even try to measure social distance empirically. However, in Simmel's writings, social distance is closely related to human emotions and mental states and is not an objective concept that is detached from actors. This tendency is particularly evident in his seminal work, "The Metropolis and Mental Life" (Simmel 1971: 324–339). Simmel's fundamental assumption about human nature is notably more pessimistic than Durkheim's discourse on social cohesion. While he emphasizes social interaction, Simmel does not believe that everyone desires to engage actively with others. The blasé attitude that is commonly observed in urban life expresses people's tendency to maintain social distance and avoid interaction with others.

The link between social distance and human emotions is spatiotemporal in nature and is subject to the constraints of both space and time. According to Simmel, when the spatial distance between people decreases, it can either lead to "the most exuberant joy" or "the most unbearable coercion" (Simmel 2009: 569). Spatial distance often works in conjunction with temporal distance to influence people's emotions. For instance, a brief separation between lovers might trigger strong emotions akin to the saying, "Absence makes the heart grow fonder." However, if the separation extends over time, their emotions may gradually cool down or even lead to indifference (Simmel 2009: 568). Simmel provides another example that involves friends and neighbors: "It is good to have one's neighbors as friends, but it is dangerous to have one's friends as neighbors. There are probably only very few friendship relationships that do not involve some kind of distance in their closeness; spatial remoteness takes the place of the often embarrassing and irritating rules by which it is necessary to maintain that inner distance with continuous contact" (Simmel 2009: 569). Among those who cannot maintain physical spatial distance, such as family members, the strictest ethical and moral rules are often established to prevent behaviors that would disrupt customs, such as incest.

Simmel aims to present a clear image of social space through his discussions of boundaries and social distance. This social space is characterized by its endogeneity, as it is based on social interactions and closely related to human emotions. The constraints actors face within this social space are spatial and temporal. Only under specific spatiotemporal constraints can we fully understand the social interaction processes among actors. In this sense, although social forms are abstract concepts that transcend concrete content, the study of forms of social space is not a logical deduction detached from the social context. Instead, this study involves observing and understanding the interactions among individuals within specific physical spaces and their evolution over time. While physical space and social space are closely related, they should not be conflated. The abovementioned difference between the physical distance and social distance of neighbors and friends is a good example. Furthermore, the vitalistic significance of social forms discussed earlier implies that these forms have a life process. Indeed, "It emerges creatively from out of the flow of interaction, gives direction and energy, but also can become exhausted to the point of provoking a search for renewal" (Silver and Brocic 2019: 118). As physical space changes with societal transformations such as wars and urbanization, the forms of social space also continuously grow and change. They can come into existence, and they can also cease to exist.

Simmel's perspective on space differs significantly from Pierre Bourdieu's field theory. Fields are imagined virtual spaces that are almost unaffected by physical space constraints, and they represent a relationship between actors and spatial positions. Positions within a field only acquire meaning when they are occupied by a particular individual or group (Liu and Emirbayer 2016). In contrast, Simmel's concept of social space is grounded in physical space and time, so the presence of actors at every spatial location is not necessary. This is what he refers to as "empty space." Simmel believes that empty space, in the context of social groups, serves both as a boundary and a connection, stating, "While the empty, unoccupied border area between two tribes functions as a neutral zone for commercial or other traffic, it is the simplest such structure in its purely and most clearly negative character, which serves as a means for this unique differentiated form of relationship among antagonistic elements and in which it is embodied, so that, in the end, empty space itself is revealed as a vehicle and expression of sociological interaction" (Simmel 2009: 620). In other words, the existence of empty space provides the possibility for social interaction. Actors' mobility within social space relies on the premise that spatial positions may not be occupied by anyone. If all positions within a space are fully occupied, mobility becomes impossible. This interpretation of empty space highlights a fundamental difference between Simmel's and Bourdieu's theories of social space. The essence of the field in Bourdieu's theory is a relatively stable structural space for which social interaction and mobility are not required as prerequisites. However, social interaction and mobility are at the core of Simmel's theory of space, which was inherited in later discussions of social space in the Chicago School.

Park's human ecology

As the oldest academic tradition in the history of American sociology, the Chicago School's sociological theory and research methods have been significantly influenced by Georg Simmel for two primary reasons. First, the founder of the Chicago School, Albion Small, and one of its leading figures, Robert E. Park, had visited or studied in Germany, which had the highest academic standards in Europe at the time. During their time there, both Small and Park became acquainted with Simmel, and they brought his sociological ideas back to the United States. Second, Chicago was one of the most important American cities for immigrants in the early twentieth century. It attracted numerous ethnic groups from Europe and around the world, so its population was highly mobile and heterogeneous. This diversity led to a wide range of forms of social interaction, creating a vibrant and colorful social environment. It provided an empirical basis for scholars such as William I. Thomas, Robert E. Park, and other leading figures of the Chicago School to apply Simmel's theoretical perspective on social forms and social interactions. In contrast to the long-standing and relatively stable European societies, the reconstruction of ethnic relations and the nation-building process in early twentieth century America were closely related. The spatial differentiation and social forms that resulted from interactions among various ethnic groups were different from the metropolitan life that Simmel described with Berlin as a prototype. This difference is particularly evident in the research on cities by Park and the Chicago School (Park et al. 1967, 1925).

The Chicago School's significant contribution to the theory of social space was the development of the research approach known as "human ecology," which serves as the core for developing a series of concepts and analytical tools for understanding interaction processes within social space. While Park is often considered the founder of human ecology, it was Park and his colleagues and students at the University of Chicago, such as Ernest W. Burgess and Roderick D. McKenzie, who collectively developed this biology-inspired sociological approach. In an early article included in the book The City: Suggestions for Investigation of Human Behavior in the Urban Environment, McKenzie provides a classic definition of human ecology as "a study of the spatial and temporal relations of human beings as affected by the selective, distributive, and accommodative forces of the environment" (McKenzie 1967, 1925: 63–64). In another article titled "Human Ecology," Park emphasizes that "society, from the ecological point of view, and in so far as it is a territorial unit, is just the area within which biotic competition has declined and the struggle for existence has assumed higher and more sublimated forms" (Park 1936: 7). This highlights a fundamental assumption of human ecology: understanding human interactions within the context of space and time. While Park and his colleagues incorporated some biological elements that were popular in the early 20th-century American social sciences, their theoretical approach remains in line with Simmel's discussion on social interaction. This fundamental assumption is further elaborated in the following passage from McKenzie's (1967, 1925: 64) work.

Human ecology is fundamentally interested in the effect of position, in both time and space, upon human institutions and human behavior. "Society is made up of individuals spatially separated, territorially distributed, and capable of independent locomotion." These spatial relationships of human beings are the products of competition and selection and are continuously in process of change as new factors enter to disturb the competitive relations or to facilitate mobility. Human institutions and human nature itself become accommodated to certain spatial relationships of human beings. As these spatial relationships change, the physical basis of social relations is altered, thereby producing social and political problems.

The term "position" refers to the location and distribution of actors in space, and in this context, "space" encompasses both social and physical/geographical dimensions. Therefore, changes in spatial relationships lead to changes in the "physical basis of social relations," which have social and political implications. Importantly, positions are not static but rather determined by various social interactions among individuals, which is a focal point of human ecology. In contrast, Bourdieu's field theory, while having a similar concept of "position" in terms of spatial relationships, holds that positions within a field are not primarily derived from interactions between actors but are more determined by the field's own social structure (Liu and Emirbayer 2016).

Park and Burgess outline four basic forms of social interaction in their book Introduction to the Science of Sociology (Park and Burgess 1969, 1921): competition, conflict, accommodation, and assimilation. Among these forms, competition holds the most fundamental position. As Park states, "Competition operates in the human (as it does in the plant and animal) community to bring about and restore the communal equilibrium, when, either by the advent of some intrusive factor from without or in the normal course of its life-history, that equilibrium is disturbed" (Park 1936: 7). According to Park, conflict, accommodation, assimilation, dominance, succession, and other forms of interaction are ultimately rooted in competition between species or populations. However, at times, these interactions might manifest as a more sublimated form of, namely, "a struggle for power and prestige" (Park 1936: 10). This idea of tracing the origins of human power struggles back to biotic competition contrasts sharply with the power theories that emphasize domination and subordination, as advocated by sociologists such as Max Weber and Pierre Bourdieu. This perspective is also reflected in the theoretical discussions of contemporary Chicago School sociologists. For instance, Andrew Abbott explicitly states that his theory of profession systems on jurisdictional conflicts among professions (Abbott 1988) is not a power model but a competition and equilibrating model. Indeed, "The equilibrating forces prevail, assuming that no profession delivering bad services can stand in definitely against competent outsiders, however powerful it may be" (Abbott 1988: 135).

However, competition as understood by the Chicago School is not equivalent to the biotic survival of the fittest or market competition in the economic sense. Park and Burgess believe that social contacts lead to interactions and that "competition, strictly speaking, is interaction without social contact" (Park and Burgess 1969, 1921: 507). This logical paradox exists because competition in human society is always intertwined with other interaction processes, such as conflict, accommodation, and assimilation. Therefore, Park and Burgess introduce a new concept—competitive cooperation—to describe the social order created by competition, in which "each individual, being free to pursue his own profit, and, in a sense, compelled to do so, makes every other individual a means to that end" (Park and Burgess 1969, 1921: 508). Although this social order based on competition transcends individual interests, it is also constrained by cultural processes such as customs, laws, and traditions rather than being pure market competition in the economic sense. Furthermore, competition and conflict, as two forms of interaction, complement each other. Indeed, "Competition determines the position of the individual in the community; conflict fixes his place in society. Location, position, ecological interdependence—these are the characteristics of the community. Status, subordination and superordination, control—these are the distinctive marks of a society" (Park and Burgess 1969, 1921:574–575).

In addition to the four basic processes of social interaction (i.e., competition, conflict, accommodation, and assimilation), Park and Burgess also emphasize another concept—isolation—in their book, Introduction to the Science of Sociology. Isolation is the opposite of social contact and interaction, and it is unimaginable for an individual to be completely isolated from social groups. However, just as geographical isolation implies spatial separation, the essential feature of isolation in sociology is the exclusion of communication. Park and Burgess argue that "the life-history of any group when analyzed is found to incorporate within it elements of isolation as well as of social contact. Membership in a group increases contact within the circle of participants but decreases contact with persons without. Isolation is for this reason a factor in the preservation of individuality and unity" (Park and Burgess 1969, 1921: 229). Evidently, isolation is not only a static state but also the result of the dynamic process of segregation, which reduces contact. It is precisely the social process of segregation and isolation that leads to the creation of the unoccupied "empty space" described by Simmel. According to Park and Burgess, the formation of nations and ethnic groups throughout history was a consequence of isolation because geographical separation led to the isolation of different groups from each other. This, in turn, was the prerequisite for the formation of different races and cultures. Some contemporary American sociologists who study race-related issues interpret Park's human ecology as a racist linear progression in which various ethnic groups move from isolation to competition, conflict, and accommodation, ultimately culminating in assimilation by the white population (Morris 2015). However, this is a serious misinterpretation because Park and Burgess's concepts are not in a linear relationship with an evolutionary goal. Instead, they describe diverse interaction processes among actors within the same social space that are closely related to the spatial locations of these actors (i.e., "position").

In addition to these internal processes of interaction within a space, the evolution of social space is also influenced by external ecological changes, as discussed by McKenzie in his study of urban community changes, focusing on the concepts of "invasion" and "succession." According to McKenzie, "in the human community the formations, segregations, and associations that appear constitute the outcome of a series of invasions," and "[t]he invasion, of course, may be into an unoccupied territory or into territory with various degrees of occupancy" (McKenzie 1967, 1925:74–75). Evidently, this discussion about invasion reflects the social reality of a large influx of immigrants into Chicago and other major American cities in the early twentieth century. However, the more important theoretical point here is that the accommodation processes that are triggered by invasion within a social space drive differentiation and integration among various social groups. As McKenzie states, "The general effect of the continuous processes of invasions and accommodations is to give to the developed community well-defined areas, each having its own peculiar selective and cultural characteristics" (McKenzie 1967, 1925: 77). Additionally, "Each formation or ecological organization within a community serves as a selective or magnetic force attracting to itself appropriate population elements and repelling incongruous units, thus making for biological and cultural subdivisions of a city's population" (McKenzie 1967, 1925: 78). Burgess's "concentric circle model" of urban community distribution, which was based on Chicago as a prototype, does not depict a static map but rather a dynamic urban spatial structure that emerges from these processes of social interaction (Park et al. 1967, 1925).

Indeed, while the human ecological perspective of the Chicago School follows Simmel's theory of social space, significant modifications are made, particularly by drawing from emerging biological theories of the early twentieth century. The human ecological perspective draws an analogy between human societies and animal and plant ecosystems. Qualitative research methods such as ethnography are used to describe and analyze the diversity, growth, competition, and interaction processes of various groups within human communities. For social space theory, human ecology introduces a set of basic analytical concepts, including competition, conflict, accommodation, assimilation, isolation, dominance, and succession, to describe processes of social interaction. It also expands the unit of analysis from interactions between individuals or groups to larger social entities such as communities and cities. Through the use of Chicago, a rapidly developing city at the time, as its laboratory, human ecology could showcase the universality and diversity of interaction processes within social space from various aspects, such as communities, ethnic groups, and occupational groups, and it paints a picture of the "natural history" of human life across multiple dimensions. Notably, compared to the focus of symbolic interactionism (Blumer 1969) of the Second Chicago School after World War II, as well as that of social theorists such as Erving Goffman on microlevel interactions, the application of the first generation of the Chicago School's ecological theory is not limited to the micro level. From the perspective of social interaction, interactions among individuals or groups and interactions among organizations or nations are not fundamentally different in terms of basic social forms. The concept of "competitive cooperation" proposed by Park exists at the micro level among individuals and groups and the macro level among organizations and nations. Furthermore, the distinction between the so-called "micro" and "macro" analytical levels is not significant from the perspective of the Chicago School. The patterns of interaction between social entities and how these interactions unfold over time and space are most important.

Goffman's territories of the self

Although Erving Goffman is often grouped with figures such as George Herbert Mead, Herbert G. Blumer, and Anselm L. Strauss as a representative of symbolic interactionism, Goffman's social theory and the symbolic interactionism pioneered by Blumer differ notably. In a 1980 interview, Goffman referred to himself as "a structural functionalist in the traditional sense" and "a Hughesian urban ethnographer" and resisted the label of symbolic interactionism (Verhoeven 1993: 318). He believes that the sociologists of his generation, including Howard Becker, Eliot Freidson, and himself, who studied at the University of Chicago in the 1940s, were heavily influenced by Everett Hughes and were what he called "social ethnographers" (Verhoeven 1993: 331). However, Blumer, who introduced the concept of "symbolic interactionism," was not interested in ethnography. Goffman even maintains that the major weakness of Blumer's symbolic interactionism is that it is "abstract," "without substance," and "'anti' system," leaving no room for structure, organization, or patterns and being merely a "critical, primitive approach" (Verhoeven 1993: 331–334). In contrast, Goffman considers his own sociology to be "old-fashioned, conservative, and unsophisticated" in his epistemology (Verhoeven 1993: 327). He is concerned with both social order and social processes, and this theoretical approach that differs from symbolic interactionism is particularly evident in his discussions on social space.

In his book Asylums, Goffman introduces the famous concept of "total institution," which refers to "a place of residence and work where a large number of like-situated individuals, cut off from the wider society for an appreciable period of time, together lead an enclosed, formally administered round of life" (Goffman 1961: xxi). In addition to mental hospitals, examples of total institutions include prisons, the military, concentration camps, and boarding schools. Goffman argues that a fundamental spatial arrangement in modern society involves people having "to sleep, play, and work in different places, with different co-participants, under different authorities, and without an over-all rational plan" (Goffman 1961: 5–6). However, Goffman argues that total institutions "create and sustain a particular kind of tension between the home world and the institutional world and use this persistent tension as strategic leverage in the management of men" (1961: 13). Those who enter these institutions undergo a process of "mortification," in which their sense of self and personal territories are invaded and contaminated in various physical and interpersonal ways. At the same time, each institutional member engages in "secondary adjustments," participating in practices that do not directly challenge the authority of staff but allow them to obtain forbidden satisfactions, such as prisoners using and exchanging contraband in prison. The process of secondary adjustments is also a process of "fraternalization," in which individuals who were initially socially distant from each other start to support one another, gradually forming common "counter-mores" in opposition to the total institution's efforts to mold them into a "single, equalitarian community of fate" (Goffman 1961: 56). In this sense, Goffman likens a total institution to "a kind of dead sea in which little islands of vivid, encapturing activity appear. Such activity can help the individual withstand the psychological stress usually engendered by assaults upon the self." (Goffman 1961: 69).

Similar to Simmel's abovementioned frame analogy, Goffman's concept of total institution represents a highly endogenous social space. Once individuals enter these institutions, they become isolated from the world outside this space. In fact, this underlying assumption is also present in the human ecology research of the first generation of Chicago School sociologists, such as Park, who views the city of Chicago as a relatively closed social space, focusing attention only on the interactions of people and society within the space, without much consideration of the impact of environmental factors outside the space on these processes of interaction. The only possible exception might be Thomas and Znaniecki's classic work, The Polish Peasant in Europe and America (Thomas and Znaniecki 1995, 1918), which includes discussions about the lives of Polish peasants in their European homeland. However, Goffman's portrayal of total institutions in his book Asylums goes even further by completely demarcating the space from the environment, where even the identities that individuals acquire in the external environment must be gradually erased through the process of mortification, ultimately achieving the institution's "total control" over individuals.

However, as Goffman's analogy of the "dead sea" reveals, there is also an "underlife" within total institutions, "the underlife of the institution, being to a social establishment what an underworld is to a city" (Goffman 1961: 199). The world of each institutional member is divided into three different spaces: the unreachable space, the monitored space, and the free areas. The unreachable spaces for most people become the free areas for a few, and each institutional member gradually forms their "personal territories" within these free areas or "group territories." Goffman's concept of "territories" is built upon the Chicago School human ecology's focus on position. However, compared to the positions of various ethnic groups in the city, as described by Park and Burgess, the social nature of group territories and personal territories in total institutions is stronger, while their physical nature is weaker. More importantly, the concept of "territory" is highly subjective, and the various territories of each institutional member differ. The same physical space can be a free area for one person but a monitored space for another and can become a personal territory for different members at different times.

The sociological research on Goffman is prolific, but not much attention has been given to his unique concept of territory and its contribution to the theoretical advancement of social space. This section focuses on this key concept in Goffman's work, aiming to elucidate his distinctive contributions to the theory of social space. Ten years after the publication of Asylums, Goffman further developed his theory of personal territory in another essay titled "The Territories of the Self" (Goffman 1971). At the beginning of this essay, Goffman writes the following:

"At the center of social organization is the concept of claims, and around this center, properly, the student must consider the vicissitudes of maintaining them... one type of claim becomes crucial: it is a claim exerted in regard to "territory." This concept from ethology seems apt because the claim is not so much to a discrete and particular matter but rather to a field of things -to a preserve-and because the boundaries of the field are ordinarily patrolled and defended by the claimant" (Goffman 1971: 28–29).

A person's territory can be fixed, such as a house or yard, or situational, such as park benches or restaurant tables, which belong to someone during a certain period. Additionally, territories can also be "'egocentric' preserves which move around with the claimant, he being in the center" (Goffman 1971: 29), such as a wallet, backpack, or mobile phone that one carries with them. However, regardless of its type, a territory is a "field of things" that needs to be "patrolled and defended." Similar to McKenzie's discussion of invasion and accommodation in urban communities, Goffman believes that the encroachment of personal territory and the resulting interaction processes are crucial for forming boundaries in social space. Goffman argues that territorial offenses encompass two different "sins": one is "intrusion," in which someone intrudes into a territory where he or she has no right of access or otherwise contaminating a preserve; the other is "obtrusion," in which someone extends their territorial claims into a wider sphere, making others feel that they are encroaching his or her territory, even if that might not be the case (Goffman 1971: 50–51). For example, a person who occupies multiple seats with his or her belongings in a subway car or plays loud music on his or her phone is committing a form of territorial obtrusion against the personal territories of others around him or her, even though he or she may not have directly intruded someone else's personal territory.

Therefore, Goffman's discussion of territories adheres to the human ecological perspective of the Chicago School, as organizational changes in society are understood through the definition and competition of spatial positions. However, what sets Goffman apart is his reinterpretation of different positions in social space, centered on the self as an actor. In his essay "The Territories of the Self," he develops eight categories of personal territories: personal space, the stall, use space, the turn, the sheath, possessional territory, information preserve, and conversational preserve.

Personal space is "the space surrounding an individual, anywhere within which an entering other causes the individual to feel encroached upon, leading him to show displeasure and sometimes to withdraw" (Goffman 1971: 29–30). The boundaries of personal space are not fixed but are continually adjusted through social interactions between people. For example, when one enters an elevator with few people, personal space is quite spacious, but as more passengers enter, the boundaries of personal space for each person continuously readjust. Conversely, if many people suddenly exit a crowded train car, the remaining passengers may "acquire a measure of uneasiness, caught between two opposing inclinations-to obtain maximum distance from others and to inhibit avoidance behavior that might give offense" (Goffman 1971: 32).

The stall refers to spaces where people can make temporary occupancy claims, such as comfortable chairs, a desk with a good view, or a telephone booth. While personal space cannot be separated from an individual, a stall can be temporarily vacated. Additionally, while personal space is challenging to share with others, stalls are often open to multiple people's claims, as with a box at a stadium or theater. The boundaries of stalls are generally quite clear and not as subject to frequent changes as personal space boundaries. In contrast, use space refers to the space around an individual that others respect due to apparent instrumental needs. For instance, in an art gallery, when someone is standing in front of a painting, others passing by will try to avoid entering the space between that person and the artwork. Similarly, when an athlete is using a piece of equipment in a gym or a worker is performing a task on a construction site, others will make efforts to avoid these use spaces. Compared to stalls, the existence of use space is contextual, and its boundaries are poorly defined.

The turn refers to a specific situation in which some individuals gain relative advantages over others based on rules such as "women and children first," "priority boarding for first-class passengers," or "first-come, first-served." The turn requires a set of ordering rules and a claiming mechanism, such as a passenger list. When people in a queue use their bodies to establish an order, they must maintain their position in the order and protect their personal space, especially when the queue is tightly packed.

The sheath refers to the skin and clothing that cover the body. This may be the most peculiar category among Goffman's eight types of personal territory because the so-called "sheath" is actually a part of the individual and represents "the purest kind of egocentric territoriality" (Goffman 1971: 38). Goffman believes that in different cultural contexts, there are variations in which parts of the body must not be violated. For example, American middle-class society may not mind the violation of the elbow but rather care about violations of other parts of the body. Possessional territory refers to items that belong to a specific individual, regardless of where these items are located. Examples include jackets, hats, gloves, phones, backpacks, and parcels. Additionally, control over certain devices can also be a form of possessional territory, such as control over radios, televisions, temperature, windows, and light. Goffman even considers children as a form of possessional territory for parents because when parents are with their children, the children are treated as their personal possessions.

Informational preserve refers to spaces in which a person stores information that he or she does not want others to know about, such as sunglasses, pockets, purses, letters, and more. This information encompasses both physiological characteristics and other personal privacy details. Conversational preserve refers to a person's right to control his or her conversational circle, including who can get him or her to speak and the right to keep the content of the conversation unknown to those outside the circle. In the virtual online spaces of the twenty-first century, informational and conversational preserves can be found everywhere on various electronic devices and social media platforms, such as chat logs on phones or WeChat groups and moments.

Assessment of the theoretical rigor of Goffman's classification of personal territories would reveal some apparent issues. For example, the same piece of clothing could function as both a sheath and a possessional territory. The concept of a "turn" is not a physical space but describes the social form of relative positions among different actors within a social space. Equating children with personal possessions is quite perplexing. While there are similarities from the perspective of ownership and control, children's agency as individuals is completely disregarded in this analogy. However, these eight categories of personal territories are fascinating because they originate from the subjective feelings and experiences of the self. They portray how social spaces are perceived and constructed with the self at the center, transforming from externally imposed structural constraints to subjective, situational territories. In analyzing and discussing these diverse territories, Goffman vividly outlines various spatial forms surrounding each individual and the social interactions revolving around these spaces.

As discussed above, the interaction processes concerning personal territories are triggered by others' intrusions, and to protect their territories, individuals employ various "markers." There are many types of markers; some are "central markers," which transform the space around the marker into personal territory, such as a backpack on a seat or a drink glass on a bar counter. Goffman even suggests that one's body is essentially a central marker of personal territory. Some markers are "boundary markers" that are used to demarcate boundaries between two adjacent territories, such as armrests between seats in a theater or the bar used in supermarket checkout counters to separate one customer's batch of articles from the next. There are also "ear markers," which involve leaving a signature or imprint on an object to signify ownership, such as writing one's name or a pattern on clothing, equipment, livestock, or even enslaved people. An interesting phenomenon related to markers is that the things used as markers are often not secure. The abovementioned concept of "the stall" is an example of this. To claim a stall, an individual must place personal belongings as markers at risk of theft. In places where theft is common, stalls are often scarcer than in other areas.

Compared to Simmel's relatively abstract concepts of social forms, Goffman's understanding of social space starts from individual experiences and connects space with concrete entities such as bodies and objects, blurring the boundaries between physical space and social space. When the unit of analysis shifts from the groups and communities emphasized by the first generation of the Chicago School to individuals, the interactions regarding territories become more than simply general social processes such as competition and conflict; they also incorporate subjective factors such as human psychology and observation. From this perspective, Goffman's theory of personal territories actually addresses a significant shortcoming in Park and Burgess's human ecology, which tends to view spatial interactions as passive evolutionary processes similar to plant growth and reproduction, with little consideration of human cognition and perception. In Goffman's writings, one can discern not only the spatial interaction tradition of human ecology but also traces of the American pragmatist tradition of social psychology, which is rooted in the work of scholars such as Charles Sanders Peirce, William James, and George Herbert Mead. This tradition represents another cornerstone of the theoretical foundation of the Chicago School of Sociology.

In his later years, in the presidential address for the American Sociological Association Annual Meeting in 1982, Goffman introduced the concept of "interaction order." He argues that face-to-face interactions between individuals create "social situations," and social situations are the fundamental units of analysis for the interaction order (Goffman 1983). He believes that a key characteristic of face-to-face interaction processes is that they are "relatively circumscribed in space and most certainly in time" (Goffman 1983: 3). Emotion, mood, cognition, bodily orientation, and muscular effort are intrinsically involved in these processes. When individuals bring their bodies and associated possessions into a social situation and come into contact with the bodies and possessions of others, individuals will be faced with personal-territory "contingencies," thus becoming susceptible to invasion (Goffman 1983: 4). Indeed, interactions between bodies not only lead to mutual invasion but also generate social techniques for managing these personal-territory contingencies. Evidently, what Goffman refers to as "social situations" are microtemporal spaces, and once we move beyond the face-to-face interactions between individuals, concepts such as total institutions, personal territories, and interaction orders become less relevant. This highlights an important distinction between the spatial theories of the first and second generations of the Chicago School. In the works of scholars such as Park, Burgess, and McKenzie of the earlier generation, the basic unit of analysis for space was often larger social entities such as communities and groups than individuals, which involved interactions that extended beyond what Goffman describes as social situations. Goffman's emphasis on social situations allows various social spaces that he describes to exhibit a more diverse range of human bodily and emotional interactions. However, the cost of this emphasis is that it limits the theoretical application of social space to the micro level.

Conclusion: the fluidity of social spaces

Is social space an entity or a rhetoric? Is it based on social structures or processes of interaction? Over the past century, different schools of sociological thought have offered various answers to these two questions. For scholars such as Simmel and those of the Chicago School, the existence of social space is rooted in social interactions among actors, and social entities and structures emerge from these interaction processes. In contrast to Bourdieu's assertion that "[t]he truth of the interaction is not to be found in interaction itself" (Bourdieu 2005: 148), neither the social forms of Simmel nor the urban ecology of Park nor the personal territories of Goffman have as strong a preconceived assumption about social structures as field theory does. Instead, they focus on the processes of interaction that produce these structures. Social processes such as competition, conflict, invasion, and accommodation, which are characterized by their continual change, imply that social space is a fluid space rather than the highly stable space depicted in Bourdieu's framework of structural positions, capital, and habitus.

These two theoretical assumptions about social space arose from their specific social contexts and empirical foundations. Bourdieu's focus was on the relatively stable French society of the mid to late twentieth century, where social hierarchies among classes, groups, and occupations were well defined, and the reproduction of social status and habits was apparent. Thus, his field theory emerged within this societal context. Although Bourdieu's early work drew from his fieldwork experience in Algeria, the empirical foundation of his field theory primarily rested in French society. In contrast, both Georg Simmel, when writing "The Metropolis and Mental Life" in late nineteenth-century Berlin, and scholars such as Park, Burgess, and McKenzie, when developing human ecology in early twentieth-century Chicago, operated in dynamic, diverse, and rapidly changing social environments. These contexts naturally influenced their discussions and imaginations of social space. In contrast, Goffman stands as an outlier. While he inherited the ecological and interactionist Chicago School tradition, he lived in a relatively stable mid- to late twentieth-century American middle-class society, which was somewhat similar to Bourdieu's context. Consequently, his concepts, such as total institutions and personal territories, exhibit less fluidity than the competitive interactions emphasized by first-generation Chicago School scholars. Nevertheless, even in light of these variations, Goffman's basic assumption in his concept of "interaction order" still places interaction prior to structure (Goffman 1983) when compared to Bourdieu's field theory. In this sociological debate resembling the classic "chicken-and-egg" dilemma concerning structure and interaction, the basic stance of the Chicago School is indeed consistent.

Another feature of the Chicago School's theories of social space is the emphasis on the interdependence of physical space and social space. This contrasts with Bourdieu's field theory, which almost entirely detaches from physical space and becomes a purely sociological imagination. Simmel's discussions on social distance among neighbors and friends, Park and Burgess's discourse on urban community transitions, and Goffman's discussions on total institutions and personal territories all construct social space and describe processes of interaction on the basis of real, concrete physical spaces. In this sense, the social spaces described and understood by the Chicago School are not merely sociological rhetoric like fields, networks, or embeddedness but are vibrant and natural entities with a social history. Certainly, not every heir of the Chicago School adheres to this theoretical approach. For instance, Abbott's theory of the system of professions does not consider the constraints and shaping effects of physical space on the professional ecological system, turning the system of professions into an abstract social space (Abbott 1988). Given that this theory emerged in the 1980s, around the same time as Bourdieu's field theory, and theories such as Niklas Luhmann's social systems theory (Luhmann 1995, 1984) and Anthony Giddens's structuration theory (Giddens 1984) also arose during this period, it might reflect a common theoretical tendency among sociologists of that particular era toward the pursuit of abstraction.

Furthermore, in the works of Simmel and the Chicago School, social spaces have two important characteristics: endogeneity and temporality. As discussed earlier, this theoretical tradition has maintained a strong assumption regarding the external boundaries of space. Metaphors such as Simmel's picture frame, cities in human ecology, Goffman's total institutions, and Abbott's system of professions all depict spaces that are relatively closed off from the external environment. While, on the one hand, this endogeneity assumption is useful for focusing research attention on the internal processes of interaction within these spaces, on the other hand, it limits the analysis of the relationships between different social spaces and the cross-space mobility of actors (Liu 2021). In contrast, field theory is more flexible in this regard. Various fields are not only open but can also influence and embed within one another. However, the Chicago School's theory of social space is more sensitive to the temporality of social change than field theory, and the diversified processes of interaction, such as competition and conflict, are not as static as Bourdieu's power struggles but rather form different natural histories in the flow of time, which in turn shape the various social structures in space. In this sense, the fluidity of social space is fundamentally rooted in its temporality, and a space devoid of a temporal dimension lacks the possibility of interaction and evolution.

Finally, it is important to emphasize the presence of human emotions within social spaces. Starting with Simmel's classic discourse on the connection between social distance and emotions mentioned earlier, the sociologists within this academic tradition have never excluded emotions from their spatial theories. Abbott, in an article titled "Against Narrative: A Preface to Lyrical Sociology" (Abbott 2007), began by referencing the description of urban neighborhoods in the classic work The Gold Coast and the Slum (Zorbaugh 1929) as an example of using sociological writing to express personal emotions. Even Goffman's seemingly cold and heartless total institutions carry numerous experiences of humiliation, joy, sorrow, and longing rather than merely serving as an "iron cage" in the Weberian sense. Ultimately, the bonds between people are not only about power and capital, as emphasized by Bourdieu, or mechanical and organic solidarity, as described by Durkheim, but also about complex and natural emotions such as joy, excitement, anger, disappointment, and more that arise from social interactions. In the rapidly evolving landscape of twenty-first-century internet technology, artificial intelligence, and digital society, the basic forms of social interaction have extended from face-to-face social situations to virtual spaces in which everyone faces electronic devices and screens. Nevertheless, even in this context, the resonance of emotions, conflicts, and interconnections between individuals remains a significant aspect of understanding various interaction processes within social spaces. As sociological researchers, when we describe the natural history of social space, it is possible to transcend Goffman's concept of social situations, but we must not overlook the "beauty and sadness" (Abbott 2007: 96) that inhabit this time and space.