Homeland as a multi-scalar community: (Dis)continuities in the US security/safety discourse and practice

This article takes steps from the birth and consolidation of “homeland” as the central discursive engine of the US national security enterprise; and takes issue with the dominant scholarly interpretation of the geographical and spatial implications of its emergence in terms of the dissolution of space and spatialization in security policy (Bialasiewicz et al., 2007: 416). We adopt a multi-scalar approach to exploring security discourse/practice, comparing the performativity of national and global security with the local practice/discourse of public safety—with empirical focus on the case of Memphis (TN). Our main arguments are that the homeland builds on the same performative elements of the emergence and consolidation of a certain conception of “community”, as it has become dominant in public safety policymaking at the local scale; and that the homeland/community performativity is the expression of a never-ending movement of production of multi-scalar geographies of the “good” and “evil”, made of the coexistence of centrifugal (pushing problems away) and centripetal (incorporating any given outside) dimensions.

Meanwhile, the term entered the popular discourse, as epitomized by the TV show Homeland, launched in 2011, which has been interpreted as both "an articulation of and a subversive critique of US foreign policy" (Negra and Lagerwey, 2015)-incidentally, Homeland was reported to be Obama's favorite show (Castonguay, 2015). The description of the nation as a "home" also features prominently in Donald Trump's discourse (Araujo, 2019). In the recent words of a journalistic commentator, the creation of the DHS "marked a shift in the way that Americans think and talk about the country, and about people": ultimately, the department was "destined to become a secret police force" (Gessen, 2020).
For De Genova, the emergent "Homeland Security State indubitably involves an ideological re-tooling operation " (2007: 422), that is, a vast discursive novelty-while he afterward argues that, in the field of border control, the new discourse served to reinforce and codify pre-existing cleavages (2007: 439). Kaplan (2008Kaplan ( /2009; see also Blunt and Dowling, 2006: 167-174;Araujo, 2019: 116) suggested that homeland was an "unfamiliar" term, a new idiom and vocabulary, which signaled a new paradigm of national identity (2003: 86). Kaplan questioned whether this discursive shift was also implying the emergence of a new form for the practice of national security: "Is the uncanny space of Guantanamo Base, as the repository of a repressed imperial history, a kind of ground zero, a new foundation on which the American homeland is being rebuilt?" (2003: 92). Indeed, many things changed in the practice of security after 9/11, especially the progressive "amalgamation" (Guillot, 2012) of foreign intervention and internal security 4 -e.g. in the management of border control (cf. De Genova, 2007). Loyd and Mountz qualified innovations in the field of border control and emphasized patterns of "continuation of historical cycles of political crises, expansion of confinement and policing" (Loyd and Mountz, 2018). Overall, in summary, critical literature interpreted the emergence of the homeland state through the lenses of "discontinuity", the component of Foucault's genealogical method (1994Foucault's genealogical method ( [1976) that emphasizes-without downplaying the existence of dimensions of persistence-ruptures and mutations that intervene between successive discursive formations and policy formulations.
As we anticipated, we are interested in the geographical and spatial implications of this discontinuity. Pease (2004: 193) suggested that the homeland was to be understood as a new driver for the US expansionary movement: "The 'Homeland' Bush invoked to legitimize the state's emergency actions did not designate either an enclosed territory or an imagined community; it redescribed the entire planet as the space that the US security apparatus was required to police in its war against global terrorism." For Kaplan, "the American homeland is the planet" (Kaplan, 2008(Kaplan, /2009, it is "about breaking out the boundaries between inside and outside" (2003: 90)-an argument De Genova (2007: 423) deemed persuasive. In short, according to these interpretations, at the same time as it is constructing consensus for global sovereignty by means of a concept discursively anchored on the nation, the homeland is materialized as the "dissolution of the inside/outside spatialization of security policy" (Bialasiewicz et al., 2007: 416).
It is with the idea of dissolution of space that we will engage: By focusing on the multiscalar nature of the homeland, we will offer a different interpretation of its novelty and spatiality. We will do so by exploring the global performativity (see the next section) of US security embedded in the discursive construction of homeland-remembering that the connection between "discourse" and "practice" is not linear. When examined through a multiscalar lens, the recent use of the homeland in national security resonates with the use of the concept of community at the local scale of security: public safety-the dominant position of the concept of community in the US (and Anglophone) understanding of local safety is evident in the omnipresence of terms and practices such as "community safety", "community policing" or "community partnerships" (see, among others, Appier, 2005;Sampson, 2002;Tulumello, 2018a). This connection has already been made, albeit marginally. The concept of community has been mentioned in critical accounts of homeland, but from static and mono-scalar perspectives. At the national security level, Pease (2004; see the passage quoted above) expressed skepticism about the interpretation of homeland as an imagined community, while Bialasiewicz et al. argued that the Bush administration wished to render "the 'homeland' zone of the continental US a homogeneous and virtuous 'domestic' community" (2007: 417). At the local level, Friedmann and Cannon (2007) suggested that homeland security concerns entered into competition with community policing practices. Our goal is to explore the homeland/community nexus at the intersection between discourse and practice, and among different scales and levels of government, ranging from the local to the global. In particular, we compare the dimensions of homeland, emphasized by existing literature, with parallel dimensions of community, which emerge from the experience of researching public safety policymaking in the USA, and particularly in Memphis (TN).
The article is structured as follows. The next section sets out our conceptual methodology, with focus on our use of performativity. Section "Defining community, US style" defines community as an instrument of inquiry. Section "Questioning homeland, or the multi-scalar community" presents the connections between the performativity of homeland and community, focusing on the discourses about, practices of, and blurring of scales between, national security and public safety. In section "Interpreting homeland/ community", we set out an interpretative framework to explore the duality homeland/community as the double movement to de-politicize the inside and pre-occupy the outside. In conclusion, we offer two takeaways: First, the new global security practices build, in fact, on old ways in which (crime) prevention has been conceptualized and acted upon on the local scale; and, second, the homeland/community performativity is the expression of a neverending movement of production of multi-scalar geographies of the "good" and "evil", made of the coexistence of centrifugal (pushing problems away) and centripetal (incorporating any given outside) dimensions.

Inquiring homeland: Conceptual methodology
During 2016, Simone (Author 1), a planner/geographer with an interest in public safety and urban geopolitics, was conducting field research in Memphis on crime prevention policies. 5 Born in Italy and based in Portugal, one of the aspects that surprised him the most was the omnipresence of the term "community" in academic, political and public debates-especially in instances in which he was rather expecting references to "society", a term he had heard once or twice during those months. 6 In time, he developed some ideas on the role of "community" for his field of research-e.g. in relation to the discursive construction of "community policing" or "community partnerships" in the growth of the security apparatus amid retrenching welfare provision (see Tulumello, 2018aTulumello, , 2018b. Back to Europe, Simone was invited to present a paper at a conference on the discourse of the homeland and, interested in linking it with his studies on public safety, invited Roberto (Author 2) to help him connect the dots by taking stock of his experience in using interpretive analysis in the field of urban policy (see Falanga, 2013). In the process, we found the approach to performativity employed by Bialasiewicz et al. to be particularly useful when building multiscalar connections between the "global" concept of homeland and the "local" concept of community: Understanding discourse as involving both the ideal and the material, the linguistic and the nonlinguistic, means that discourses are performative. Performative means that discourses constitute the objects of which they speak. For example, states are made possible by a wide range of discursive practices that include immigration policies, military deployments and strategies, cultural debates about normal social behaviour, political speeches and economic investments. The meanings, identities, social relations and political assemblages that are enacted in these performances combine the ideal and the material (2007: 406).
We operationalize our exploration of the performativity of the homeland through two epistemological tactics. First, we build on Robinson's reflections on global urban studies (2016) to set out a comparative approach of discourses/practices not among different places, but rather different scales: We adopt a wider, if looser, conceptualization of the "case" as "many instances, or distinctive outcomes, singularities, multiply resonant with one another, and interconnected though a shared genesis" (2016: 18)-our poles for comparison are discourses and practices, and their shared genesis a "goal", so to speak, that is, the intention to "secure" (at different scales). We operationalize this epistemological approach through the multi-scalar exploration of a case study, making use of a general approach of critical policy and discourse analysis (see section "Questioning homeland, or the multi-scalar community" for methodological details).
Second, with the goal of linking scales and levels, we adopt an interpretative framework typical of psychosociological research. Psychosociology draws from the Freudian perspective of the struggle between human instincts aimed to either aggregate and preserve (i.e. instinct of life), or to divide and combat the other (i.e. instinct of death) (Freud, 1921). In the social realm, research focusses on the ways in which people co-produce symbolical meanings about the world (Enriquez, 2003). The symbolical stance of meanings relies upon the construction of diverse emotional significances of the environment, which are shaped through a tense relation between "we" and the "others" (Matte Blanco, 1988). Accordingly, the definition of the identity (from the Latin idem, "identical to oneself") is necessarily paired by its opposite, what is not identity, that is, the otherness-cf. Young (1990) and Sibley (1995) on (in)justice and exclusion in political theory and social geography.
Understanding emotions as social connectors among human beings, the distinction between good and evil has effects over shared ideas about what can be considered as friendly, and what is perceived as threatening our life. In this vein, emotions further help distinguish between what can be considered as inside and outside, which constructs our feeling of belonging or non-belonging to a specific place. Likewise, emotions guide our understanding about hierarchical stances, whether we are in a position where we can exert power over something or someone, or if we are at the bottom-line (Carli and Paniccia, 2003;Carli et al., 2016). Such dichotomous distinctions characterize the emotional baseline of human life, and reach more elaborate forms of symbolization that help us organize personal and social experiences in the world. Discourses and practices can thus be interpreted in light of how we construct individually, and co-produce collectively, multiple emotionally embedded meanings of the world. In conclusion, this approach allows us to explore how actually existing security practice and discourse build upon-indeed, often exploit-important psychological dimensions of individual and social life; therefore allowing to transcend dimensions that are historically "local" (above all, the construction and reconstruction of social groups) toward national and transnational levels.

Defining community, US style
In line with our general approach, we will use community as a performative concept, that is, a discursively constructed object of practice, and focus on the dominant conception of community in the US public and political discourse, as well as its use in the field of public safety. Since, as we will see below, the idea of community is mobilized with many different meanings-which vary in time, in space and even through contradictory coexisting uses-we will focus our understanding on the conceptual core of this use, which we find in two dimensions. First, the long-discussed dichotomy between society and community, 7 and in particular the compatibility of the idea of community as a bottom-up construction of free individuals with the Anglo approach to liberalism, as opposed to the top-down construction of society and its institutions-see Norton and Bieri (2014) for the role this dichotomy plays in explaining USA-Europe differences in spatial planning. Many North American thinkers advocate the importance of community spirit to counter what they interpret as the dehumanizing dimensions of modern societies: see Dewey (1927: 146-151) on the "Great Community" and Hirschman's call (1994) for strong community spirit in society.
Second, the spatialized conceptualization of the local community as the proper unit for collective association. This is particularly evident in the work of the Chicago school of urban sociology-from Park's idealization of urban community as a "spatial pattern and a moral order" (Park, 1967: ch. 4) to theories about social disorganization and neighborhood effects (Sampson, 2002(Sampson, , 2019-and in the central role of "community development" in urban policy (see, e.g., Imbroscio, 2006;Stoecker, 1997). 8 The concept of community is increasingly used to foster integrated local development with the direct participation of citizens, and often justified by an understanding of the inherent or self-constructed capacity of local communities to contrast exclusion and segregation of minority groups (Michels and de Graaf, 2010). Communities are approached as the most adequate social and spatial locale to bond, allegedly internally "alike", members, and bridge them with the wider society, likewise with the others, the "other" (e.g. Putnam, 2000). Sternghtening "social capital" is hence expected to reduce socio-spatial inequalities and improve trust among citizens and towards decision-makers.
Against this backdrop, Hall and Massey (2010: 64;cf. Rose, 1996) observe the increasing dominance of the (local) community over (national) society, which they interpreted as a discursive shift coherent with dominance of market ideology, localism, austerity and neoliberalization. In this sense, community is less interesting for its ontological value and more for the shift in its discursive role. Esposito (1998) traces the progress by which community has been losing its original emphasis on shared responsibility, moving towards an emphasis on what a local group has in "common", that is, on homogeneity. 9 During the process, the local community becomes the only possible spatial mediation between the individual and the nation. By overlooking the uncertainty, instability and unpredictability of social groupings, the emergence of the community has been a component of the territorialization of the social into a governable object (Rose, 1996: 331;Forrest and Kearns, 2001;Wallace, 2010). Critique has explored how the increased centrality of community has impacted urban development: The vagueness of the concept itself (Lawless and Pearson, 2012) is often associated with state and market forms of co-optation of local organizations, and a trend for repackaging the society against phenomena of socioeconomic segregation (Amin, 2005).

Questioning homeland, or the multi-scalar community
We are now ready to discuss connections and divergences between the performative role played by the emergence of homeland at the center of national security and that played in local public safety by the concept of community. We therefore move to compare the dimensions emphasized by critiques of the homeland with reflections stemming from our research on public safety in Memphis, focusing on three dimensions for analysis: the discursive construction of homeland/community as inclusionary/exclusionary spaces; the enactment of security/safety; and the blurring of previously distinct dimensions (scales, practices).
With regard to data collection, we adopt a case study approach, using data collected from mixed sources including policy document analysis, qualitative interviews with policymakers and experts, 10 and participant observation. 11 Why Memphis: Introduction to the case study With the goal of theorizing multi-scalarly, we have selected Memphis for being an "ordinary" mid-size US metropolis (Memphis hosts 650 thousand and is the central city of a metropolitan area with roughly 1,350 thousand residents). Global urban studies (e.g. Robinson, 2016) have used the concept of ordinary cities as a call for overcoming generalizations based on "partial interpretations concentrating on paradigmatic examples" (Amin and Graham, 1997: 417)-that is, most often, large, global cities like New York or Chicago. In this light, Memphis-a city that has long remained marginal to urban theorization in North America (Tulumello, 2018a(Tulumello, , 2018b)-has a twofold value for our investigation. On the one hand, it is characterized by socio-spatial patterns of high (racial and class) residential segregation, deep inequality and spatial concentration of crime in minority-majority neighborhoods that have long been presented as characteristic of mid-and large-sized US cities-meaning that our reflections in this regard have value that can be adapted to most cities of similar size. On the other hand, however, national long-term trends of neoliberalization of urban policy (see, e.g. Simon, 2007), with their effects in terms of shifting away from social policy and toward aggressive policing, are particularly intense in Memphis (Tulumello, 2018a(Tulumello, , 2018b, thus giving us the opportunity to reflect on a relatively advanced version of the phenomena under study.
Since this latter dimension is crucial to our discussion, some characterizations of the policy/politics dimensions of public safety and crime control in Memphis will help the reader better place our following discussion (for a detailed discussion see Tulumello, 2018aTulumello, , 2018b. Though, as aforementioned, spatial patterns of crime in Memphis are those typical of US cities, this is not the case for general volumes thereof: Police data put it among the cities with highest crime rates in the USA-for instance, murder rates have consistently been four, five times national ones. While it is always important to take comparisons made with crime data with a pinch of salt, 12 three dimensions matter for our discussion. First, the fact that Memphis is widely considered, in the public and political discourse, a "dangerous" and "violent" city. Second, the fact that this discourse creates the backdrop against which law enforcement is prioritized and carried out. And, third, the deeply racialized nature of both the discourse and practice. The following excerpt from an interview with a lawyer, member of the Memphis Crime Commission, is exemplary of these dimensions: When you have. . . when your populations is two thirds African American, and you have the highest poverty rate of any city in the nation, somebody can link race to everything that you're talking about. I mean, it's not just worth arguing about that. We have too much crime by any measure in the world. We have too much poverty by any measure in the world. And they are linked, no question. You know we are not going to be the safest city in the country with that much poverty. They're too directly correlated. But you can't just go: "oh, we're just gonna let all this continue, because if we [. . .] talk about it, that is racist." That's nuts. I mean, our city will just fall apart if we don't talk about what is really happening, which is . . . Let's turn it around. All this conversation about Black Lives Matter. Black lives do matter. They don't matter the way they're talking about. That's peanuts. The number of black lives that are affected by police making a mistake or even doing something egregiously wrong-they should be penalized, punished, treated accordingly. But the reason that black lives matter is not because of those occasional screw-ups, they matter because a disproportionate number of the victims in this country and in this city are African American. They are the ones that are the victims, nine times out of then-no matter whether that's the right stat. You know what I know. And that's because a disproportionate number of perpetrators are African Americans. And if we are not willing to just say it and go deal with these issues . . . [. . .] If we are not willing to admit it, then how can we do anything about it? Tulumello (2018aTulumello ( , 2018b has explored the political economy of these dimensions, that is, the vicious circle through which racialized discourses on crime are used to justify investing on law enforcement at the expenses of social and urban policy, therefore contributing to those very (racialized) inequalities that are at the core of patterns of crime and violence in Memphis (and elsewhere in the USA). Importantly, as we will discuss below, the discourse and practice of public safety is always built in a contradictory fashion through coexisting claims about "community safety" and "zero tolerance" (cf. Garland, 2001), as exemplified by the first two goals of Memphis Crime Commission's Safety Plan: A) to strengthen community engagement in crime prevention efforts; and B) to strengthen law enforcement ability to reduce violent street crime. 13 Discursive security/safety James Bartlett centers his critique of Bush's security policy on the discursive construction of homeland as a (national) space imaginatively inhabited by a homogeneous people: "Homeland is a place, specifically a state, inhabited by people who were born there and are compatible with it, perhaps in terms of sharing the same race, ethnicity, or cultural background" (2001). Bartlett is doubtful that a "shared" ancestry is "the glue that holds the United States together" (2001), an idea that would eventually erase the actual complexity of what constitutes a nation with a long history of racial and ethnic tensions, injustice and conflict-see also Kaplan's critique of the "melting pot" (2003: 86).
These points resonate with critiques of the discursive construction of community at the local level. According to Young (1990: 12), the ideal of community is about "shared public life, [. . .] mutual recognition and identification." With the aim of building an (allegedly) homogeneous social space, the community is constructed in opposition to the outside at the same time as it excludes those internal components that may threaten the sense of identity. The result is the (ideal) suppression of differences, and hence the blurring and pacification of racial and class struggles and conflict, as discussed by critical takes on community and communitarianism (see Bader, 1995;Franzoni, 1999;Harvey, 2000: 164-170;Mohan and Stokke, 2000;Sennet, 2002Sennet, [1977Wimmer, 2008Wimmer, : 1041and, for counterarguments, Crary, 2013: 114-118;van Leeuwen, 2014). Though these processes have been discussed throughout the Western world, the centrality of "community" in the US discourse can also be explained by the fact that, in US cities, the "imagined community" is more easily visualized than in other places. US cities are among the most spatially segregated in the West, among racial/ethnic and class lines, and Memphis is no exception, with some studies putting it among the most segregated cities/metros (Figure 1; Florida and Mellander, 2015; Comen, 2019). The relative "homogeneity" of US neighborhoods constructs inclusionary/exclusionary patterns in an especially visible way: In a (say) majority White, affluent suburb, where commuting is exclusively performed by car, a (say) Black youth walking on the curb is visibly identified as "external" to the "community"-hence easily labelled as a potential threat. In other, more spatially mixed cities, differences are more visible, hence less evident, in the sense that they are ordinary rather than extraordinary. 14 What we have discussed so far resonates squarely with discussions of the way modern and neoliberal governmentalities need to frame the "us" in relation and opposition to the "them"-as discussed, e.g., in ethnic studies (Wimmer, 2008), political theory (Young, 1990), cultural geography (Sibley, 1995), literature and psychoanalysis (Kristeva, 1988), and human geography (Hutta, 2009). According to Kaplan (2003), the mainstream discourse of the homeland builds the "domestic", symbolized as a friendly inside, in opposition to the "foreign", more often perceived as a threatening place outside (imaginary) borders. Indeed, in the State of the Union Address of 2002 (see Introduction), Bush framed his "homeland" in opposition to the "axis of evil". 15 In more strategic, but not less rhetorical, fashion, this oppositional construction is evident in Thomas Barnett's "Pentagon's New Map" (Figure 2; see Barnett, 2004). Barnett is a "military geostrategist" (from his Wikipedia webpage), who served as the Assistant for Strategic Futures in the Office of the Secretary of Defense after 9/11; one of those "non-state scribes"  central to the imagination of Bush administration's defense strategy. Barnett's map divides the world into two macro-regions: the "functioning core", made up of the developed, long-term industrialized countries plus emerging economies such as the BRICS; and the "gap", made up of those regions not fully integrated in the global economy.
Barnett's strategic vision is that the USA should lead the process to progressively force the "gap" being integrated (cf. Bialasiewicz et al., 2007: 413). This strategy, together with the oppositional construction envisioned by Bush, builds a global understanding of the homeland as an expansionary process, where the integration of the "gap" is the precondition to guarantee the security of the homeland, and a new form of global hegemony.
In our research on Memphis, we witnessed a similar employment of the community, where the latter is constructed in opposition to external threats, from which it needs to be defended. Among the key informants we interviewed, two policymakers (a lawyer, member of Memphis Crime Commission, and a high-ranking police official) kept using terms such as "the bad guys" or the "knuckleheads", and the need to prevent them from offending the "community". 16 At the same time, the violence/community nexus is strongly spatialized in the public and political discourse, as exemplified by the practice of producing crime maps. We have collected four such examples (Figure 3), two produced by the Memphis Police Department (MPD), and two by real estate consultancy agencies.
The maps are quite different in style: The police murder map stands out for the crudeness of the red bodies representing the murders; the other maps use different methods (heat maps or color scales) to emphasize respectively crime hot spots, "safer" neighborhoods and highcrime cities within the metropolitan area. The four maps have in common the construction of a direct association of certain places with higher/lower levels of violence, reproducing the idea that (violent) crime is concentrated in inner-city, poor, majority-minority neighborhoods, and thence the racialization of discourses about community and crime (Camp and Heatherton, 2016;Hinton, 2016). The result is a process of "territorial stigmatization" (Wacquant et al., 2014) whereby the structural issues at the root of crime concentration, such as the intersection of poverty, exclusion and mental health issues, but also the priorities Figure 3. Examples of crime maps from Memphis from various sources: a) Murder map produced by MPD (source: www.wmcactionnews5.com/story/33061487/mpd-murder-map-reveals-most-deadly-areas-in-thecity/); b) heat map (part one, drugs and weapons offenses) produced by MPD (source: www.bizjournals.com/ memphis/news/2016/11/02/local-crime-hot-spots-remain-consistent.html); c) Safest Memphis Neighborhood map, produced by Location, Inc. (www.locationinc.com/) (source: NeighbourhoodScout, www.neighborhoodscout.com; reproduced by kind permission); d) crime map, Memphis metropolitan area, produced by Sperling's BestPlaces OpenStreetMap (source: www.bestplaces.net/crime/city/tennessee/ memphis; available on OpenStretMap, CC BY-SA 2.0). of policing (cf. Friedson and Sharkey, 2015;Tulumello, 2018b), are overlooked and the community is at the same time mobilized and blamed: I envision a community where the community is policing itself and you don't need law enforcement to come in because they are not allowing individuals to sell dope by their homes or harbor fugitives. . . or harbor any type of ill activity that spills out in the community (high-ranking police official, interview).
In a similar fashion to the construction of the global "axis of evil", the local community is built on a dual process of oppositionality with the external and internal cleansing of "bad apples", those inner components that threaten a common identity definition. This is a longterm trend, at least since the New Deal consensus was being replaced by the hegemony of repressive approaches to security and the use of criminalization as a system of governing (Hinton, 2016;Simon, 2007). More recently, Donald Trump's presidential bid made rich use of anti-city discourses, for instance the condemnation of "the crime and the gangs and the drugs that have stolen too many lives" (quoted in Swanstrom, 2017).
In summary, much in the same way as the rhetorical construction of "failed states" and dangerous countries works to overlook the (post-)colonial relations shaping the global security environment, territorial stigmatization over local communities works to conceal the role of uneven development and capitalist urbanization in producing the roots for urban violence (cf. Pavoni and Tulumello, 2020)-impacting in turn the practice of security, to which we now turn.

Enacted security/safety
The discourses highlighted above are reflected in the practice of security underpinned by both the homeland and the community, which we now discuss around the three concepts of co-production, exception and preemption.
To begin with, Kaplan (2003) comments that the imagined homeland of the Bush administration must be, at the same time, protected from external threats and mobilized against it-for instance, by way of involving citizens and community policing in the prevention of terrorism (Friedmann and Cannon, 2007;Lyons, 2002;Williams et al., 2016). This resonates with the different interpretations of the etymology of community (Fornari, 1979;but cf. Esposito, 1998; and section "Defining community, US style section"), which may either come from cum moenia ("with walls", to be protected) or cum munus ("with duty", to be activated). Policymakers in Memphis adopt precisely this double approach to the community, which needs be defended, and to become capable of self-defense (cf. Tulumello, 2018a), in line with global (or, at the very least, Western), recent transformations of public safety policies. Battistelli (2013; see also Garland, 2001) emphasizes the central position assumed by concepts of participation and co-production, which lie in between attempts at including the civil society in the definition of priorities for public safety and the call for self-defense to compensate for retrenching state provision.
In Memphis we found many examples of safety co-production, from Neighborhood Watch 17 to online crime-tip instruments. 18 Some of our interviewees, from MPD and the municipal department of urban planning, have been discussing the possibility of implementing practices of Crime Prevention Through Environmental Design, a technique of situational prevention of crime by means of urban design and planning, whose discourse is centered on the role of the "community" for self-defense (cf. Parnaby, 2006). Two excerpts from an interview with a high-ranking police official emphasize the role of community activation: Crime reduction is an environmental decision, is a political decision and a law enforcement decision. So crime reduction is responsibility not only of law enforcement but the entire community. So safety is, you know, to have a thriving community.
[Asked about the possibility of implementing CPTED practices:] Where you are looking into something as simple as asking people to cut back trees, take care of blight and litter, that makes the community more attractive, which has shown to be less attractive to criminals, because there more is a sense of people care about their community and they are more likely to cooperate with law enforcement. Kaplan (2003) highlights another dimension of the performance of homeland, namely the production of spaces and zones of exception (see Agamben, 2003) useful for the enactment of "security" practices that would be illegal otherwise, the most obvious example being Guantanamo Bay-but think also of "rendition flights" (i.e. kidnapping of suspect terrorists) by CIA in Europe (Hakimi, 2007) or the way (racial) profiling is silently embedded in airport security (Adey, 2009). One could interpret the pushbacks faced by Barack Obama in his attempts to shut down Guantanamo as the difficulty of reintegrating the "problem" of suspect terrorists into the national space and its legal order. It should not come as a surprise that the rhetoric of community is located within the very name of the most paradigmatic version of spaces of urban exception, the "gated community", a US phenomenon that has spread worldwide (Glasze et al., 2006). Scholarship has discussed how gated communities work as the construction of a homogeneous social space, expected to provide safety and identity to residents (Low, 2003;Roitman et al., 2010). Indeed, gated communities in Memphis are almost exclusively located in the eastern, wealthy, mostly White, suburbs; while, in line with recent trends of gentrification, some gated communities have appeared in the downtown area. 19 Importantly, statuses of legal exception are central to the (re)production of gated communities (Petti, 2007): from the illegal annexation of public goods to the suspension of certain rights of residents (including those to choose on the decoration of their own houses or to receive visits at any hour of the day).
Finally, after its early development as a response to 9/11 terrorist attacks, the banner of "preemptive strikes" has been attached to most war and international policing operationsfor instance drone strikes in countries like Pakistan and Afghanistan (Rogers and Hill, 2014). More generally, anticipating, and therefore preempting, future threats has become the central concern of the DHS (Anderson, 2010) and of the security machine more generally (Amoore, 2013). The local dimension of these trends is the long-term shift of safety policy toward prevention (Garland, 2001), which has found its ultimate expression in the advent of predictive policing, a technology-driven version of hot-spot policing. By analyzing in quasi real time crime reports and other inputs like CCTV with automatic plate readers, predictive policing determines patrol strategies. 20 In Memphis, the Police Department developed, in cooperation with the University of Memphis and IBM, the Blue CRUSH (Crime Prevention Utilizing Statistical History) program, which was considered a success during the years of its full implementation, 2007-2011, though on the basis of rather disputable data (Tulumello and Iapaolo, 2021). In response to the fact that many in Black and Brown neighborhoods were seeing predictive policing to be a process of "occupation" (Tulumello and Iapaolo, 2021), MPD launched the Community Outreach program-a discursive strategy that underpins how a number of practices, including aggressive order maintenance and nuisance abatement, have been historically labelled community policing in the USA (Goetz and Mitchell, 2003;Tulumello, 2018b).

Blurring security/safety
The discursive construction of the homeland deepened a set of transformations of security/ safety practice, namely the progressive blurring of previously distinct layers (Bigo, 2000). We refer, in particular, to the double, if not completely linear, movement of constabularization of the army and militarization of local police. The constabularization of the army is a recent process, as "many war-fighting interventions have increasingly involved a significant degree of peacekeeping, civil-conflict stabilization, and nation-building" (Campbell and Campbell, 2010: 331). Such process is in line with, and one of the components of, shifting strategies associated with the emergence of the DHS, for instance in the way the construction of what was "once" an empire (created through war and conquest) becomes the production of a global homeland (through nation-building and safety-making). The recent emergence of drones, with their role in banalizing micro-war interventions and blurring international policing and the military (Rogers and Hill, 2014), is paradigmatic of the constabularization of the army, at the same time as it opens to the complementary movement of militarization of public safety-see Bigo (2014) and Davis (2019) on the advent of drones in domestic policing.
The militarization of local police, 21 part and parcel of the shift away from the New Deal consensus in the management of crime, is however a much longer process, started roughly with federal funding for the "war on drugs" in the 1980s (Hinton, 2016;Simon, 2007; see also section "Discursive security/safety"), and then accelerated after 2001 under the efforts for the "war on terror" (Campbell and Campbell, 2010;Harwood, 2014). The DHS played a central role in this initiative, through funds, grants and equipment (see Walby and Lippert, 2015)-to the extent that popular media noticed the deployment of absurdly inappropriate resources in small cities with no security threats in sight. 22 A case from Memphis exemplifies the crucial connection of the two ideals of, and discourses about, homeland and community in this field. In July 2016, after two shootings of police officers in Baton Rouge and Dallas, Memphians witnessed the deployment of heavily armed police squads in tactical gear at highly trafficked locations: a showoff of force with employment of equipment made available within the framework of homeland security activities, justified by police leaders as a "precautionary measure" for "community safety" purposes (see WMCActionNews5, 2016).

Interpreting homeland/community: Depoliticization and pre-occupation
Having analyzed the performativity of community and homeland to show their common strains, we can now proceed to interpretation, setting out a framework inspired by ideas of psychosociological research. Though these dimensions have been long known to critical geography and political theory-see, e.g., the classicals by Young (1990) and Sibley (1995)-this perspective can help us further understand how security politics builds its political and social justification by mobilizing, in its multi-scalar performativity, important dimensions of psychological life.
In state-regulated contexts, the identification of who/what belongs or does not belong to a local community or a nation brings about a complex set of political issues that go along with the symbolical dimensions of identity definition, and operationalization of duties and rights. Regulative aspects compel political decisions taken within and over social systems, which are, from a Foucauldian perspective, operationalized through governmental technologies and procedures (see, e.g., Fuggle et al., 2015). From a psychosociological point of view, decision-making relies on the agreement among different interpretations-i.e. symbolical meanings-of the world. The sharing of common social representations helps explain the ambiguous, ambivalent and often contradictory construction of the relationships between the self and the others (Moscovici, 1989;Sibley, 1995). These are necessarily emotionally embedded processes that consolidate our take on emotions as social connectors, at the root of human and social life.
The Freudian tension between the instinct of life, which moves people to identify themselves with their place of belonging, and the instinct of death, which rather drives people to actively separate themselves from the others and, at occasion, defend their safety, is coupled by basic emotional categorizations, such as friend/enemy, inside/outside, top/down (Carli and Paniccia, 2003;Carli et al., 2016). It is on this ground that society and community can be imagined-and are often represented-as cohesive and homogenous in opposition to the outside, as shown by the performativity enacted through the concepts of "community" and "homeland". The invention of internal unity is a formidable, and politically powerful, answer to the struggle between the instincts of life and death, while the attempt to reconcile antagonist forces within the society-as evident in the discursive production of the USA as a "nation of immigrants" or the use of the local community as the scale where to find (allegedly) homogeneous intents-is functional for the preservation of life at the expense of the imagined enemy (Moscovici, 1989).
Both "community" and "homeland" are, thus, contended symbolical spaces of politicization/depoliticization, which can be used to produce delimitations and characterizations regarding who belongs to what, and which rights and duties are associated with each status. Considering the different scale of action, homeland defines a symbolical space where "who we are" is embedded in the political conception of the role of the USA in the world, constructed from within. 23 Likewise, community refers to the perception that some citizens can be considered members because they have (or acquire) the right to act in a specific context. We interpret this first mechanism in the definition of both concepts as the depoliticization of the "inside". In other words, the construction of the outside is the performative device through which the concerns about the structural roots of security issues are put aside in favor of an emphasis over the role of the external enemy.
The dichotomization of the social system through the characterization of insiders versus outsiders works at the same time as the construction of a common perception of fear (see, e.g., Hutta, 2009;Massumi, 2005;Pain and Smith, 2008). The perceived need to defend communities and nations from external attacks is narrated as an ontological truth and legitimizes the search for practical arrangements of protection. When a crisis occurs, and external forces are labelled as responsible, the tension between the instincts of life and death leads to disruptive impacts at individual and collective levels. 9/11 was exploited to give the perceived need for defense a new symbolical prominence, constructing homeland security policies as an apparently straightforward response. Crucially, attacks do not necessarily need to be real for similar decisions to be taken. From the response to real attacks, social systems reinforce the (equally legitimate) prevention from expected threats. As there is always another possible, not yet anticipated, looming threat, security becomes a neverending endeavor (Anderson, 2010), in an ideological shift that can easily create phenomena of collective anxiety and hysteria.
Belonging to a determined community/homeland, insiders are always assumed to be good unless the contrary is demonstrated. On the other side of the spectrum, those who stand outside are assumed to be bad and threatening, unless they keep demonstrating that they are good and friendly enough to deserve insider reappreciation. Less frequently the opposite happens: The insiders rarely threaten the outside; but when they do, it is for good reasons. When the inside is disturbed from within, "bad behaviors" are often associated with individual, othering reasons, such as mental illness. 24 Hysteria is the perfect sandstorm, a symbolical short-circuit that originates from a paranoid conception of what is deemed to stay out of the borders.
This prompts us to interpret the second mechanism in the definition of homeland/community as the pre-occupation of the "outside". From a psychosociological perspective, preoccupation means that the prediction of what is expected to happen occupies the mental state to a larger extent. Carli and Paniccia (2003) argue that preoccupation is a strong emotion that can organize social relationships, which can stem from feelings of powerlessness (or envy), occupied by suspicion, doubt and sorrow. Preoccupation is performed through an imperative to act, or prevent, in defense of the insiders.
Expanding homeland/community: The multi-scalar geographies of security This article took steps from the critical literature that emerged as a response to the consolidation of "homeland" as the central discursive engine of the US national security enterprise. We took issue with the interpretation, dominant in the literature, that the homeland signals a break from past understandings of the spatialization of US security.
We argued that the homeland builds on the same performative elements of a certain conception of "community", as it has become dominant in public safety policymaking on the local scale. In other words, recent transformations of US global intervention and imperialism find their roots in older transformations, typical of the neoliberal, post-New-Deal, shift of the US culture of public problems. We have interpreted these transformations through two mechanisms that shape the performance of homeland security and public safety: The inside is stripped of its conflictual dimensions, that is, depoliticized, thence shifting the focus toward the outside, which is then pre-occupied as a security/safety measure-an interpretation that resonates with, and gives a multi-scalar dimension to, Walters' twofold understanding of 'domopolitics' (2004) as the drive to secure the homeland and domesticate the forces that are said to threaten it.
This interpretation allows us, in conclusion, to reconsider the spatial dimensions of the processes of expansion of the US homeland security apparatus, and in particular their interpretation as the dilution of space and dissolution of inside/outside dichotomizations in national security (cf. Bialasiewicz et al., 2007).
The construction of "insides" at different scales (the local community, the national homeland) implies that every inside includes some components that are also outside on different scales-quite obviously, a (say) Black American soldier fighting in Afghanistan is performatively incorporated in the same socius that, once they have returned, excludes them as a Black youth in an "inner-city" neighborhood in (say) Memphis. At the same time, the expansion of the empire entails the progressive incorporation of former outsides-for instance, Barnett's "Pentagon's New Map" (see Figure 2) incorporates many countries once seen as enemies in the "functioning core". This double process of exclusion and incorporation resonates with Stuart Elden's interpretation of Foucault's understanding of the shift from "disciplines" to "security": While discipline operates through the enclosure and circumscription of space, security requires the opening up and release of spaces, to enable circulation and passage. [. . .] Discipline is centripetal, while security is centrifugal; discipline seeks to regulate everything while security seeks to regulate as little as possible, and, rather, to enable, as it is, indeed, laissez faire; discipline is isolating, working on measures of segmentation, while security seeks to incorporate, and to distribute more widely (2007: 46-47). Indeed, by looking at the performativity of security/safety from the local community to the planetary homeland, we could observe the coexistence of two spatial movements: a centrifugal one, the pushing of social problems "outside" any given "inside", from the neighborhood to the globe; and a centripetal, complementary one, the incorporation of any given "outside" within the "inside", from enemy states to problematic (racialized) communities. The "new" homeland, once observed through the lenses of the "old" community, entails, rather than a dilution of space, a continuous production of spaces, a never-ending, double movement of production of multi-scalar geographies of "good" and "evil". 6. Surprise probably due to the fact that, in general, European societies have political and cultural traditions that are more centered on the role of society (and hence on the social pact and welfare state), on the one hand, and individual positive rights, on the other, than on the value of the local community. See, for instance, the debate between T€ onnies (1887) and Weber (1968Weber ( [1958), the USA-Europe comparison by Norton and Bieri (2014), and Tulumello and Colombo's application (2018) of this reflection to the case of gated communities. 7. See previous footnote. 8. Incidentally, it is through this perspective that the modern US ideal of community has travelled back to the European context-for instance, in the political imagination of Italian "enlightened" capitalist Olivetti (see Cadeddu, 2012). 9. Community's etymology, from the Latin communitas, is composed of cum (with) and munus, a "gift" that needs be given, that which the individual owes to the socius-in opposition with immunity, immunitas, the property of being free from the communal duty. The meaning of community as that which is in common, according to Esposito, probably derives from a wrong etymological connection with the Greek munus, wall, according to which the community would signify a shared protection. 10. Ten interviews (in one case the interviewee provided written answers to a list of questions) and three work meetings with: academicians (University of Memphis, Criminology and Criminal Justice); civil servants from city departments (MPD, Parks and Neighborhoods, Housing and Community Development, and Planning and Development); a retired criminologist and former consultant of MPD; an activist from the Mid-South Peace and Justice Center; and a lawyer, chair of Memphis Crime Commission. 11. Participation in town-hall meetings and other public events on issues linked to public safety; and in an action-research partnership between the Department of City and Regional Planning of the University of Memphis and the Community Development Corporation of Klondike and Smokey City. 12. Critical criminology has long argued that crime data only depict a part of crime as a social phenomenon, and that police activity influences them heavily. 13. See https://memphiscrime.org/safe-community-plan/ (accessed 15 October 2020). 14. Granted, we are not at all suggesting that racial divides are directly and linearly correlated with spatial segregation. We do believe that the spatial organization of US cities plays a role in the specific US history of racial relations. Indeed, authors that study community-based, co-production initiatives share evidence on the risks associated with the construction of public values in societies that reflect this kind of spatial organization (Williams et al., 2016). 15. See www.washingtonpost.com/wp-srv/onpolitics/transcripts/sou012902.htm (accessed 15 October 2020). 16. Despite admitting in the same interviews that the largest share of violent crimes stems from "domestic violence" and "infra-communal" disputes. 17. A national initiative in which organized groups of citizens take responsibility for providing information to the police and communicating with neighbors about safety issues. 18. Two systems are active in Memphis, one managed by the police department and another by a nonprofit. See https://mdsas.memphispolice.org/cw/n/cyberwatch.php and http://memphis.crimestoppersweb.com/ (accessed 15 October 2020). 19. See the "gated communities" section of Layson Group -Keller Williams Realtywebsite, https:// www.memphisrealtysearch.com/gated-communities-memphis-tn/ (accessed 15 October 2020). 20. The rationale of predictive policing can be scrutinized from the critical criminological standpoint that crime stats (i.e., the crime known to the police and the judiciary) do not represent "crime" as a social fact, but are deeply influenced by several factors, including the likeliness that a crime is reported and police priorities, which are themselves historically skewed toward the policing of racialized minorities and the poor (Sutherland and Cressey, 1978;Reiner, 2016). The technological elaboration of crime may thus end up reproducing, at the same time as it gives a semblance of objectiveness, long-known problems linked with the idea that policing is, above all, an instrument for the reproduction of the liberal order (see, e.g. Neocleous, 2001). 21. One should remember that, unlike most European countries, criminal police is a local competence in the USA. 22. See, for instance, a 2014 episode of Last Week Tonight: https://www.youtube.com/watch? v¼KUdHIatS36A (accessed 15 October 2020). 23. According to Hardt and Negri (2001: 187), US imperialism can be understood as: inclusive because it overcomes individual differences, differences being considered unnecessary at a broader level; still so, differential, as once differences are allegedly neutralized, cultural differences (n.b. not socioeconomic or political differences) become a new appealing product for globalization; and managerial, as localisms are one of the ways through which the new global order manages and controls international relationships and trades, and creates new hierarchies between contexts and subjects. 24. An example of the practical consequences of this tension to divide the good inside from the bad outside is provided by recent immigration policies. Political discourses and practices of Western countries are increasingly grounded on the assumption that there is no chance for migrants (the outsiders) to have the same rights as the citizens (the insiders) unless they prove to deserve them through never-ending tests, which range from time-demanding bureaucratic procedures to exceptional demonstrations of attachment to the nation or community.