Conceptualising Space in Transnational Migration Studies. A Critical Perspective

This paper argues that space is a crucial dimension of study when studying migration and mobility. However, the conceptualisation of space, and how people “produce” transnational social space through their everyday cross-border actions, remain insufficiently explored in studies of transnational migration. After critically reviewing the literature, the paper raises the question of what could be a conceptually adequate and empirically practicable approach to examine space in studies of transnational migration and mobility. It proposes a conceptual model comprising three interrelated dimensions of analysis: materiality, social practice, and meaning.

temporary settlement followed by a return to the native country, the transnational perspective views migration as a simultaneous connection between two or more geographical localities. Clearly, space is at the core of this question. Everyday social interactions taking place between individuals living in different geographical locations do not happen in a "vacuum" but in space and through space. The significance of space for our everyday lives has long been observed by geographers (Thrift 2006). But how to understand space? This question has been of central preoccupation for geographers for quite some time (e.g. Soja 1989, Allen 2003, Harvey 2006, Thrift 2006. Besides, the 'spatial turn' across the spectrum of humanities and social sciences has sought to understand the concept of space beyond the play of spatial metaphors (e.g. Smith and Katz 1993). The question thus need to be raised of how migration scholars have conceptualised space and empirically addressed the spatial dimensions of transnational migrant's lives. The following section presents a critical review of the literature regarding this question.

Critical review of the literature: How has space been conceptualised in studies of transnational migration?
For a start, it is surprising to realise that there has been relatively little debate among migration scholars on how to conceptualise space and how to empirically examine it in the context of transnational migration (Pries 2008:78;Riaño & Richter 2012). Although space is clearly a core dimension of study, studies that specifically focus on the spatial dimension of transnational social practices, are rather rare. In recent years, there has been growing interest among researchers in studying the spatial dimensions of transnational social practices (Richter 2012, Jackson et al 2004. Studies have been carried out from two main perspectives: (a) locality and (b) transnational social space.
Studies that use the perspective of locality have pointed to the importance of embedding migrants' lives in specific localities (Levitt & Jaworsky 2007). Ley (2004) has argued that transnational migrants are "not always in the air" but must necessarily touch down somewhere. Other authors argue that migrants are "unable to escape their local context" despite being "transnational" (Featherstone 2007;Landolt 2001).
For example, locally bound state regulations, such as migration legislation, continue to play a role in shaping the possibilities of action and mobility of individuals (e.g. Bauböck 2007;Bommes 2005;Dahinden 2008;Levitt et al 2003). Glick Schiller et al (2006) have argued that studying the "specificities of locality" is important to understand the characteristics of Riaño, Yvonne 2017. "Conceptualising Space in Transnational Migration Studies. A Critical Perspective". In: Youkhana E. (Ed. 3 migrants' transnational lives. Moreover, Anthias (2008) has proposed the term translocational positionality to address issues of identity in terms of multiple social locations. She argues that "although we may move across national borders and remain middle class or women (for example) the movement will transform our social place and the way we experience this at all social levels and in different ways" (idem:15). Thus, we need to think of the multiplicity of locations that shape the identities of mobile individuals.
Locality studies seem an exciting field of research for understanding the spatialities of transnational social practices. The question needs to be raised, however, of how scholars understand the concept of "locality" and "location". The concept of "locality" and "location" have not always been sufficiently defined. A variety of uses has been given in the literature to the concepts: sometimes it is used to mean geographical location (geographic use) or social location (sociological use); some other times it is used to mean a concentration of population (demographic use), and yet some other times to mean an administrative division in rural areas or the size of human settlement (scalar use). Anthias' use of the term "location", for example, seems to mainly refer to an individual's position in society as a result of his/her gender, class and ethnicity. But individuals are also positioned in material space; i.e. in concrete geographical locations (i.e. countryside, city, specific neighbourhoods) where they live their everyday lives, and move daily across different geographical localities. What is the relationship between social position and geographical position? This relationship does not seem sufficiently discussed in the migration literature.
More clarity is necessary about the social and material dimensions of the concept "location", and how they intersect.
Furthermore, Glick- Schiller and Çağlar (2010) argue that locality matters in migration research. Glick Schiller et al (2006) use the concept of city size to examine "locality" and the "specificities of locality". They point out that paying attention to city size is important because "the scale of cities reflects their positioning within neoliberal processes of local, national, regional, and global rescaling" (Glick Schiller et al 2006).  project. Besides, locality is used in a metaphorical way to refer to the idea of a sociopolitical context, thus, for example when scholars use the term "locality" to refer to the state regulations that shape the possibilities of action and mobility of individuals (e.g. Bauböck 1994Bauböck , 2007Bommes 2005;Dahinden 2008;Hess 2008;Levitt et al 2003).
From a methodological point of view, the concept of "locality" also raises some important questions. The first one is the problem of how to delimit localities for the empirical study. Massey (1991) has argued that "localities are not simply spatial areas you can easily draw a line around" but should be "defined in terms of the sets of social relations or processes in question". This leads us to the second question of what should be the methodological point of departure for the empirical examination of transnational practices: localities as sociophysical entities or the individuals themselves? Some scholars have argued that our understanding of space cannot be merely reduced to physical structures and the geometry of location and size (e.g. Blaut 1961, Duncan & Savage 1989. Following an additional but related line of argumentation, Marston (2000; questions our use of a geographical scale perspective to understand human experience. In her view, scalar concepts such as the "global", the "local" and "locality" are social constructs, which do not necessarily correspond to people's experience. Thus, if we want to reconstruct individuals' notions of space, rather than starting with pre-existing notions of scale, it may make more sense to start the empirical inquiry with the actual experiences of individuals, track the geographical extent of their transnational practices and then draw conclusions on how space is constituted in a transnational context (Riaño & Richter 2012).
The perspective of transnational social space is the second set of migration studies with a specific spatial focus. Scholars have been specifically preoccupied by space and have used a variety of understandings. Faist (1998;, for example, defines it as a combination of sustained social and symbolic ties that are contained in social networks and organisations that can be found in multiple states. Interestingly, space is used in this definition as a merely abstract concept. The notion of space is used to denote an abstract social field constituted by several social ties. The material dimensions of space are thus not addressed by this concept. Perhaps the most comprehensive work on transnational social space has been carried out by Pries (2001;2008). His work is a response to the observation that social practices are "no longer embedded in uni-local geographic spaces" but "have a multipolar 5 society as ensuing from the transnational social spaces that have come to characterize our contemporary social world. In his view, transnational social spaces are "relatively dense and durable configurations of social practices, systems of symbols and artefacts" (Pries 2001).
His empirical analysis of transnational social spaces focuses on "pluri-locally spanned social realities and entities that grow up either from the grassroots by international migration or through a complex top-down and bottom-up process brought about by international business companies" (Pries 2001). In his later work, Pries (2008) uses the concept to empirically examine the "social realities" and "entities" that grow out of the transnational social lives of international migrants and the organisational structures of transnational organizations. After such an empirical examination, he addresses the question of how to conceptualise space thus reviewing the work of key authors in the field such as Bourdieu, Giddens and Simmel who emphasize aspects of power and social exclusion (Bourdieu 1985;Giddens 1984;Simmel 1903). Finally, he produces a categorisation that distinguishes between three types of transnational social spaces: (a) at the micro-level are the households of migrant families; (b) at the meso-level are non profit-and profit oriented international organizations such as multinational corporations; and (c) at the macro level are cross-border international regimes such as the United Nations.
The former work advances our understanding of space in studies of transnational migration as it specifically addresses some important dimensions that are constitutive of social space such as social practices and symbolic representations. At the same time, questions remain open regarding the conceptualization of transnational social space. Why? First, space is used as a merely abstract concept. The question of the material dimension of space is not addressed. Clearly, social space cannot exist without a material basis. Secondly, the concept of space is used in a metaphorical way. Concepts such as "realities", "entities", "households", "organizations", and "regimes" are used to refer to space. Such a metaphorical use has being criticized, among others, by Jackson et al (2004), for being insufficient. Thus, again, the concept of space remains vague and tailored to the specific research project. Pries' geographical typification of transnational social spaces also raises the question earlier posed if geographical scale is the most relevant criterion to categorise transnational social space.
Thus, in conclusion, although space is clearly a core dimension of study for the understanding of transnational social practices, its conceptualisation, and how people  6 "produce" transnational social space through their everyday cross-border actions, remain insufficiently explored. In that sense, this paper supports Featherstone's (2007) view that we need to carry out much more scientific work which explicitly focuses on the spatialities of transnational social practices, both in theoretical and empirical terms.

What can be a fruitful approach to conceptualise transnational social space in studies of migration and mobility?
It has become apparent that we need a precise, comprehensive and practical approach to conceptualise space in studies of transnational migration. In order to address this question we would like to call attention to the widely used understanding by human geographers of space as a "social production" (e.g. Buttimer 1969;1972;Chombart de Lauwe 1952, Löw, 2001Lefebvre 2005Lefebvre [1974 ;Massey 1994;. I understand this approach as been based on two premises: First, space is not merely an abstract notion but has a concrete material basis. Social space cannot exist without a material basis. Second, physical space is not merely a container where human action takes place. Rather, physical space is transformed by the actions of individuals. Interactions taking place between material space and human action invest the former with symbolic meaning thus transforming it into social space. Following the ideas of the above mentioned authors, and my own work (Riaño 1996(Riaño , 1998 in part with other colleagues (Riaño & Richter 2012), I propose three dimensions to conceptualize and empirically study socially produced space: (a) materiality, (b) social practice, (c) and meaning. The following sections present my on-going efforts to develop an integrated and practicable approach for conceptualising and empirically examining space in studies of transnational migration. Figure 1 below summarizes the proposed approach. For the sake of analytical purposes, these three dimensions are presented as separate, but they are necessarily interrelated. 8 Social practice, the second dimension, refers to the way groups and individuals use and appropriate material space for large scale activities such as agriculture, industry, commerce, recreation or housing, as well as for daily activities such as social interaction, economic exchange, religious practice, political protest, etc. For studies of transnational migration studying social practice means inquiring into how groups and individuals use and appropriate specific material spaces, for specific purposes, and how such use and appropriation is imbedded in structures of power. Also, studying social practice means recognising social networks as the basic social structure around which social practice is organized. Social networks can be understood as "the social alliances structuring everyday action" (Riaño 2000). Social networks connect individuals in Meaning, the third dimension, refers to the significance that human beings attribute to the material spaces in which / through which they carry out their human actions.
Understanding meaning is understanding social space. Indeed, for Buttimer (1972) social space is a "dynamic continuum upon which the experiencer lives and moves and searches for meaning". The process of investing physical space with meaning is contingent upon the changes that occur through everyday interactions and actions (Massey 1994). Although space is materially constituted, such material structures are reworked into a meaningful composite through the social actions and interactions of groups and individuals (Fredrich et al 2007). Long ago, Massey described such social actions and interactions as the "trajectories" of individuals (1994) that intersect at a specific place and give that place a specific meaning. For Johnston (2004: p. 68) "a space becomes a place through human interactions with it, both through physical manipulation, via such activities as agriculture, architecture, and landscape, and symbolically, via such activities as remembering, formulating, , depicting, and narrating". Thus, empirical analyses of transnational migration need to focus on the question of symbolic meaning, i.e. giving explicit focus in the empirical enquiry as to how material places are actually invested with symbolic meaning, thus producing social spaces. Further, studying meaning implies understanding symbolic representations and geographical imaginaries. Groups and individuals have specific imaginations of distant places where their friends, colleagues and relatives live and these imaginations play a role in the meaning they attribute to their places of origin. Represented spaces may be imbued with social meanings, that can be individual or collective, structured in layers, some deeper than others but always overlapping each other. People do not just give meaning to space but they also establish a relationship with it. Belonging has thus a spatial component. Levitt and Glick Schiller (2004) also elaborate on this issue and distinguish between "ways of being" and "ways of belonging" into a social field. Antonsich (2010 ) argues that belonging is a notion both vaguely defined and ill-theorized. It should be analysed as place-belongingness. In her analysis of belonging, and the related politics of belonging in migration studies, Youkhana (2015) incorporates space as an analytical category that cross-cuts established categorizations of race, class, gender, and stage in the life cycle.
The notion of "sense of place" is useful to understand the attachment and sense of belonging that people have to specific material spaces. For Yi-Fu Tuan (1974), sense of place refers to the affective bond between people and the physical settings in which / through which their actions take place. Sense of place can have a positive (e.g. belonging) or negative connotation (e.g. fear). The idea of cosmopolitanism (Appiah 2007;Hannerz 2006;Tarrius 2000) help us understand that it is possible to belong to multiple places.

MATERIAL STRUCTURES
Giving attention to the specific characteristics and implications of the concrete material sites in which/through which social practices take place.

SOCIAL PRACTICE
Examining how individuals use and appropriate different physical spaces, for different purposes, in different socioeconomic & political contexts, at different times, and with specific consequences regarding access to resources.

SYMBOLIC MEANING
Focusing on how material sites are (re)created, transformed and invested with meaning in the process of movement / migration

Tracing meanings
Two final points. First, conceptualising space in studies of transnational migration cannot be disentangled from time (Jaisson 1999;Massey 2005;Soja 1996). Space cannot exist outside time, as every place (a city, a square, a house, a room) is a specific place at a specific moment. It can, however, change (maybe slowly, maybe faster) as people cross it, use it, and appropriate it. To stress this fact, Massey uses the term time-space when referring to a socially produced space (Massey 2005). Also, transnational relations imply that there exist different places around the globe that are simultaneously important in one person's life (Levitt & Glick Schiller 2004). Although people are able to bridge spatial distance by using modern communication technologies they also experience the time lag that exists between different places. Time is thus constitutive of social space and can constrain or facilitate the transnational experience (Ley 2004). Thus, the empirical study of transnational social space needs to be much more clearly connected to the question of time.
Second, analysing transnational social space needs to be imbedded in geographically and temporally contingent power relations (gender, nationality, class, age). The latter necessarily affect the mobility and immobility of individuals, and thus the constitution of transnational social space. Mobility has often been treated in the literature as an unproblematic term. Authors such as Massey (1993:60) point out that "Time-space compression has not been happening for everyone in all spheres of activity". Mobility is thus not equal for all (Cresswell 2010). Empirical analyses of transnational social space need to take account of existing differences of mobility among individuals depending on their gender, class, and ethnicity.

Conclusions
This paper has argued that space is a crucial dimension of study when studying migration and mobility. Everyday social interactions taking place between individuals living in different geographical locations do not happen in a "vacuum" but in space and through space. A critical review of the transnational migration literature has shown that recent work has considerably advanced our understanding of transnational migrants' practices. At the same time, the conceptualisation of space, and how people "produce" transnational social space through their everyday cross-border actions, remain insufficiently explored.
Moreover, this paper argues that space is not merely an abstract or metaphorical concept, as it has often been used in the migration literature, but has a tangible material dimension.
In line with Featherstone (2007), the paper pleads for carrying out much more scientific work which explicitly focuses on the spatialities of transnational social practices, both in theoretical and empirical terms. Accordingly, this paper has raised the question of what can be a conceptually adequate and empirically practicable approach to examine space in studies of transnational migration and mobility.
The paper has proposed an initial conceptual model comprising three interrelated dimensions of analysis: materiality, social practice, and meaning. Addressing the material dimension of space means identifying, characterising and understanding the concrete material sites in which / through which the transnational lives of migrants take place.
Studying social practice means inquiring into how groups and individuals use and appropriate different material sites, for different purposes, in different socio-economic and political contexts, at different times, and with specific consequences regarding access to resources. Finally, the paper has argued that material sites are not merely a container where human action takes place. Rather, material sites are symbolically transformed into a meaningful composite through physical manipulation, social interaction, appropriation, experience, representation, and remembering. We need to explore how material sites are (re)created, transformed and invested with meaning in the process of transnational movement and migration. Ultimately, the proposed three-dimensional analytical model needs to be embedded in power relations (gender, nationality, class, age) and in a temporal perspective.
Finally, this paper has made some provocative methodological suggestions. It has argued that concepts commonly used by migration researchers such as "global", "micro", "macro" and "meso" are social constructs by researchers, which do not necessarily correspond to people's experience. Thus, I have suggested that if we want to reconstruct an individuals' experience of space, rather than starting with pre-existing notions of scale, it may make more sense to start the empirical inquiry with the actual experiences of individuals, track the geographical extent of their transnational practices, and then draw conclusions on how space is constituted in a transnational context.